The Journal of Computing

Sciences in Colleges

Volume 26, Number 6 June 2011

 The Journal of Computing
Sciences in Colleges

Papers of the Sixteenth Annual CCSC

Northeastern Conference
April 15-16, 2011
Western New England College
Springfield, Massachusetts

John Meinke, Editor George Benjamin, Associate Editor

UMUC — Europe Muhlenberg College

Susan T. Dean, Associate Editor Michael Gousie, Contributing Editor

UMUC — Europe Wheaton College

Volume 26, Number 6 June 2011

The Journal of Computing Sciences in Colleges (ISSN 1937-4771 print, 1937-4763
digital) is published at least six times per year and constitutes the refereed papers of
regional conferences sponsored by the Consortium for Computing Sciences in Colleges.
Printed in the USA. POSTMASTER: Send address changes to Jim Aman, CCSC
Membership Chair, Saint Xavier University, Chicago, IL 60655.

Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission

to copy without fee all or part of this material is granted provided that the copies are not
made or distributed for direct commercial advantage, the CCSC copyright notice and the
title of the publication and its date appear, and notice is given that copying is by
permission of the Consortium for Computing Sciences in Colleges. To copy otherwise,
or to republish, requires a fee and/or specific permission.

ii
 TABLE OF CONTENTS

THE CONSORTIUM FOR COMPUTING SCIENCES IN COLLEGES

BOARD OF DIRECTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

CCSC NATIONAL PARTNERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x

FOREWORD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x
John Meinke, UMUC Europe

WELCOME TO CCSCNE 2011 FROM THE CHAIRS . . . . . . . . . . . . . . . . . . . . . . . 1

Stoney Jackson,Western New England College, Karl R. Wurst, Worcester State
University

BOARD MEMBERS — CCSC NORTHEASTERN REGION . . . . . . . . . . . . . . . . . 2

2011 CCSC NORTHEASTERN REGIONAL CONFERENCE COMMITTEE . . . . 2

REVIEWERS – 2011 CCSC NORTHEASTERN CONFERENCE . . . . . . . . . . . . . . 3

WHAT CAN WE DO? RAISING THE NUMBERS OF WOMEN IN COMPUTING

— FRIDAY PLENARY SESSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Jane Chu Prey, Microsoft Research

COMMUNITIES: OPEN SOURCE AND EDUCATION WORKING TOGETHER —

SATURDAY PLENARY SESSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Stormy Peters, Mozilla

GREENFOOT: INTRODUCING JAVA WITH GAMES AND SIMULATIONS —

PRE-CONFERENCE WORKSHOP / TUTORIAL PRESENTATION . . . . . . . 7
Adrienne Decker, University at Buffalo, SUNY, Frances P. Trees, Drew
University

AN OPEN SOURCE FIELD TRIP FOR FACULTY — PRE-CONFERENCE

WORKSHOP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Gregory W. Hislop, Drexel University, Clif Kussmaul, Muhlenberg College, Mel
Chua, Red Hat, Inc., Sebastian Dziallas, Olin College

iii
GETTING READY FOR A UNIT IN ALICE 2.2: WEEK 1 /
EVENT-PROCESSING IN ALICE 2.2 — PRE-CONFERENCE WORKSHOP /
TUTORIAL PRESENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Eileen M. Peluso, Lycoming College

INTRODUCING CASE STUDY BASED TEACHING IN COMPUTING

CURRICULA — PRE-CONFERENCE WORKSHOP . . . . . . . . . . . . . . . . . . 13
Massood Towhidnejad, Salamah Salamah, Thomas B. Hilburn, Embry-Riddle
Aeronautical University

IMPROVING THE STUDENT SUCCESS AND RETENTION OF UNDER

ACHIEVER STUDENTS IN INTRODUCTORY COMPUTER SCIENCE . 16
Carolee Stewart-Gardiner, Kean University

USING VISUAL LOGIC©: THREE DIFFERENT APPROACHES IN DIFFERENT

COURSES - GENERAL EDUCATION, CS0, and CS1 . . . . . . . . . . . . . . . . . 23
Dee Gudmundsen, The College of Saint Rose, Lisa Olivieri, Chestnut Hill
College, Namita Sarawagi, Rhode Island College

ALIGNING GENERATIONS TO IMPROVE RETENTION IN INTRODUCTORY

COMPUTING COURSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Christian Roberson, Plymouth State University

IMPROVING STUDENT PERFORMANCE WITH AN EXTRA-CURRICULAR

ROBOTICS CLUB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Charles N. Stevenson, Salem State University

INTRODUCTION OF VIRTUALIZATION IN THE TEACHING OF OPERATING

SYSTEMS FOR CS UNDERGRADUATE PROGRAM . . . . . . . . . . . . . . . . . 44
Kevin Grammer, Jack Stolerman, and Beifang Yi, Salem State University

ROBOTICS ACROSS THE CURRICULUM WITH dLife — DEMONSTRATION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Grant Braught, Dickinson College

JAVA VISUALIZATION OF STUDENT-IMPLEMENTED OS ALGORITHMS —

DEMONSTRATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Adam Fischbach, Widener University

iJava - AN ONLINE INTERACTIVE TEXTBOOK FOR ELEMENTARY JAVA

INSTRUCTION — DEMONSTRATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Robert Moll, University of Massachusetts

USING THE GOOGLE MAPS API WITH HIGHWAY MAPPING DATA AS A

PEDAGOGICAL TOOL — DEMONSTRATION . . . . . . . . . . . . . . . . . . . . . . 58
James D. Teresco, Siena College

iv
THAT'S NEAT -- HOW DO I DO IT? — DEMONSTRATION . . . . . . . . . . . . . . . 61
Alice E. Fischer, University of New Haven

EXPERIENCES WITH COMMUNITY-BASED PROJECTS FOR COMPUTING

MAJORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Jeffrey A. Stone, Elinor Madigan, Pennsylvania State University

CHANGING PERCEPTIONS OF COMPUTER SCIENCE AND

COMPUTATIONAL THINKING AMONG HIGH SCHOOL TEACHERS . . 71
Patricia Morreale, David Joiner, Kean University

AN INTERDISCIPLINARY APPROACH TO INJECTING COMPUTER SCIENCE

INTO THE K-12 CLASSROOM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
David Goldschmidt, Ian MacDonald, Judith O'Rourke, Brandon Milonovich,
The College of Saint Rose

METHODOLOGIES AND TOOLS FOR THE SOFTWARE QUALITY

ASSURANCE COURSE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Vladimir V. Riabov, Rivier College

A SOFTWARE ENGINEERING COURSE WITH A LARGE-SCALE PROJECT

AND DIVERSE ROLES FOR STUDENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Bonnie K. MacKellar, St. John's University

SOURCE CODE ANALYSIS AND ATTACK TECHNIQUES AS EDUCATIONAL

APPROACHES TO TEACH SECURE PROGRAMMING . . . . . . . . . . . . . . 101
W. Scott Harrison and Nadine Hanebutte, St John Fisher College

LEARN HOW TO USE EXECUTABLE FLOWCHARTS TO ENHANCE

LEARNING IN GENERAL EDUCATION, CS0, and CS1 COURSES —
TUTORIAL PRESENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Dee Gudmundsen, The College of Saint Rose, Lisa Olivieri, Chestnut Hill
College, Namita Sarawagi, Rhode Island

STUDENT INTEREST AND CHOICE IN PROGRAMMING ASSIGNMENTS

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
Lisa Torrey, St. Lawrence University

A CONTENT-BASED IMAGE RETRIEVAL PROGRAMMING ASSIGNMENTS

FOR INTRODUCTORY COMPUTER SCIENCE COURSES . . . . . . . . . . . 117
Michael Eckmann, Skidmore College

LANGUAGE CONSIDERATIONS IN THE FIRST YEAR CS CURRICULUM

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Jonathan D. Blake, Quinnipiac University

v
INTRODUCING BIOINFORMATICS INTO THE COMPUTER SCIENCE
CURRICULUM— PANEL DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Mark LeBlanc, Wheaton College, David Rilett, John Russo, Michael Werner,
Hongsheng Wu, Wentworth Institute of Technology

UNDERSTANDING NSF FUNDING OPPORTUNITIES — TUTORIAL

PRESENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Sue Fitzgerald, National Science Foundation

PROCESS-ORIENTED GUIDED INQUIRY LEARNING (POGIL) IN COMPUTER

SCIENCE — TUTORIAL PRESENTATION . . . . . . . . . . . . . . . . . . . . . . . . 135
Clif Kussmaul, Muhlenberg College

MOTIVATING STUDENTS TAKING CS1 BY USING IMAGE MANIPULATION

IN C AND C++ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Paul E. Dickson, Hampshire College

THE I-PHONE/I-PAD COURSE: A SMALL COLLEGE PERSPECTIVE . . . . . . 142

Lubomir Ivanov, Iona College

SOUNDLIB: A MUSIC LIBRARY FOR A NOVICE JAVA PROGRAMMER . . 149

Viera K. Proulx, Northeastern University

UNTANGLING THE MAZE OF PEDAGOGICAL STRATEGIES FOR

INTRODUCTORY PROGRAMMING — PANEL DISCUSSION . . . . . . . . 156
Frances Bailie, Iona College, Adel Abunawass, University of West Georgia,
Smiljana Petrovic, Iona College, Deborah Whitfield, Slippery Rock
University

SHARING YOUR INSTRUCTIONAL MATERIALS VIA ENSEMBLE —

TUTORIAL PRESENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Gregory W. Hislop, Drexel University, Lillian N. Cassel, Villanova University,
Lois M. L. Delcambre, Portland State University, Edward A. Fox, Virginia
Tech, Richard Furuta, Texas A&M University, Peter Brusilovsky,
University of Pittsburgh, Daniel D. Garcia, University of California,
Berkley

SHORT MOBILE GAME DEVELOPMENT PROJECTS FOR CS1/2 — TUTORIAL

PRESENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Stan Kurkovsky, Central Connecticut State University, Delvin Defoe, Rose-
Hulman Institute of Technology

MOTION CAPTURE IN A CS CURRICULUM — FACULTY POSTER . . . . . . 165

Bridget Baird and Ozgur Izmirli, Connecticut College

vi
COMPUTER SCIENCE EDUCATION IN SECONDARY SCHOOLS IN WESTERN
NEW YORK — FACULTY POSTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
Sarbani Banerjee, Buffalo State College

HOW COMPUTER SCIENCE DEVELOPS MATHEMATICAL SKILLS —

FACULTY POSTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
Reneta P. Barneva, State University of New York at Fredonia, Boris Brimkov,
SUNY University at Buffalo

ROBOTICS, AI AND VISION IN JAVA WITH THE DLIFE LIBRARY —

FACULTY POSTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
Grant Braught, Dickinson College

AN UPDATED TAXONOMY OF WRITING IN COMPUTER SCIENCE

EDUCATION — FACULTY POSTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Mark E Hoffman, Quinnipiac University

THE COMPUTER SCIENCE CLUB: BUILDING A STUDENT COMMUNITY

WITH PROJECTS AND ACTIVITIES THAT EXTEND BEYOND THE
CLASSROOM — FACULTY POSTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
Susan P. Imberman, College of Staten Island

CAPSTONE EXPERIENCE: ENGAGING STUDENTS IN SPEECH PROCESSING

TO EXCITE THEM ABOUT STEM — FACULTY POSTER . . . . . . . . . . . 180
Michael Jonas, University of New Hampshire at Manchester

A WEB SERVICE TO CONNECT DATA ANALYSIS AND VISUALIZATION —

FACULTY POSTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
Thomas Jyh-Cheng Liu, New Jersey City University

USING PENNY AUCTIONS TO INTRODUCE NON-CS MAJORS TO

MODELING AND SIMULATION — FACULTY POSTER . . . . . . . . . . . . 184
David Toth, Merrimack College

THE DESIGN AND COMPLEXITY ANALYSIS OF THE LIGHT-UP PUZZLE

PROGRAM — FACULTY POSTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Douglas Selent, Rivier College

DIGITAL HOME: A CASE STUDY APPROACH TO TEACHING IN

COMPUTING CURRICULA — FACULTY POSTER . . . . . . . . . . . . . . . . . 190
Massood Towhidnejad, Salamah Salamah, and Thomas Hilburn, Embry Riddle
Aeronautical University

MATHDOKU PROJECTS: MONTE-CARLO MEETS METHODOLOGY —

FACULTY POSTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
Linda Wilkens, Providence College

vii
MINING FORUMS FOR ASYNCHRONOUS LEARNING IN DISTANCE
EDUCATION — FACULTY POSTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
Songmei Yu, Sofya Poger, Alberto LaCava, Felician College

INDEX OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

viii
THE CONSORTIUM FOR COMPUTING SCIENCES IN COLLEGES
BOARD OF DIRECTORS

Following is a listing of the contact information for of Computer Science, Huntsville, Texas
the members of the Board of Directors and the 77341-2090, 936-294-1571, [email protected].
Officers of the Consortium for Computing Sciences Brent Wilson, Northwestern Representative (2012),
in Colleges (along with the year of expiration of their George Fox University, 414 N. Meridian St,
terms), as well as members serving the Board: Newberg, OR 97132, 503-554-2722 (O),
Bob Neufeld, President (2012), Professor Emeritus 503-554-3884 (fax), [email protected].
of Computer Science, McPherson College, P. O. Box Pat Ormond, Rocky Mountain Representative
421, North Newton, KS 67117, 316-283-1187 (H), (2013), Professor, Information Systems and
[email protected]. Technology, Utah Valley University, 800 West
Laura J. Baker, Vice-President (2012), Professor, University Parkway, Orem, UT 84058,
Computer Sciences Department, St. Edward's 801-863-8328 (O), 801-225-9454 (cell),
University, Box 910, 3001 S. Congress Ave, Austin, [email protected].
TX 78704, 512-448-8675, [email protected]. Lawrence D'Antonio, Northeastern Representative
Jim Aman, Membership Chair (2013), Assoc. (2013), Ramapo College of New Jersey, Computer
Professor, Computer Science, Saint Xavier Science Dept., Mahwah, NJ 07430, 201-684-7714,
University, Chicago, IL 60655, 773-298-3454 (O), [email protected].
630-728-2949 (cell), [email protected]. Linda Sherrell, MidSouth Representative (2013),
Bill Myers, Treasurer (2011), Dept. of Computer Associate Professor, Computer Science, University
Studies, Belmont Abbey College, 100 of Memphis, 209 Dunn Hall, Memphis, TN 38152,
Belmont-Mount Holly Road, Belmont, NC 901-678-5465 (O), [email protected].
28012-1802, 704-461-6823 (O), 704-461-5051, Serving the Board: The following CCSC members
(fax), [email protected]. are serving in positions as indicated that support the
John Meinke, Publications Chair, (2012), Collegiate Board:
Associate Professor, UMUC Europe, US Post: CMR Will Mitchell, Conference Coordinator, 1455 S
420, Box 3668, APO AE 09063; (H) Werderstr 8, Greenview Ct, Shelbyville, IN 46176-9248,
D-68723 Oftersheim, Germany, 011-49-6202-5 77 317-392-3038 (H), [email protected].
79 16 (H), [email protected]. George Benjamin, Associate Editor, Muhlenberg
Kim P. Kihlstrom, Southwestern Representative College, Mathematical Sciences Dept., Allentown,
(2011), Associate Professor of Computer Science, PA 18104, 484-664-3357 (O), 610-433-8899 (H),
Westmont College, 955 La Paz Road, Santa Barbara, [email protected].
CA 93108, 805-565-6864 (O), 805-570-6722 (cell), Susan Dean, Associate Editor, Collegiate Professor,
805-565-7036 (fax), [email protected]. UMUC Europe, US Post: CMR 420, Box 3669, APO
Elizabeth S. Adams, Eastern Representative (2011), AE 09063; Werderstr 8, D-68723 Oftersheim,
Associate Professor Emeritus, James Madison Germany. 011-49-6202-5 77 82 14, (H)
University - Mail Stop 4103, CISAT - Department of [email protected].
Computer Science, Harrisonburg, VA 22807, Robert Bryant, Comptroller, Professor &
[email protected]. Information Tech. Program Director, MSC 2615,
Deborah Hwang, Midwestern Representative Gonzaga University, Spokane, WA 99258,
(2011), Dept. of Electrical Engineering and 509-313-3906, [email protected].
Computer Science, University of Evansville, 1800 Paul D. Wiedemeier, National Partners Chair, The
Lincoln Avenue, Evansville, IN 47722, University of Louisiana at Monroe, Computer
812-488-2193 (O), 812-488-2780 (fax), Science and CIS Department, 700 University
[email protected]. Avenue, Administration Building, Room 2-37,
Scott Sigman, Central Plains Representative (2011), Monroe, LA 71209, 318-342-1856 (O),
Associate Professor of Computer Science, Drury 318-342-1101 (fax), [email protected].
University, Springfield, MO 65802, 417-873-6831, Brent Wilson, Database Administrator, George Fox
[email protected].. University, 414 N. Meridian St, Newberg, OR
Kevin Treu, Southeastern Representative (2012), 97132, 503-554-2722 (O), 503-554-3884 (fax),
Furman University, Dept of Computer Science, [email protected].
Greenville, SC 29613, 864-294-3220 (O), Myles McNally, Webmaster, Professor of Computer
[email protected]. Science, Alma College, 614 W. Superior St., Alma,
Timothy J. McGuire, South Central Representative MI 48801, 989-463-7163 (O), 989-463-7079 (fax),
(2012), Sam Houston State University, Department [email protected].

ix
 CCSC NATIONAL PARTNERS

The Consortium is very happy to have the following as National Partners. If you
have the opportunity please thank them for their support of computing in teaching
institutions. As National Partners they are invited to participate in our regional
conferences. Visit with their representatives there.
Microsoft Corporation
Turings Craft
National Science Foundation
Panasonic Solutions Company

FOREWORD
Once again we come to the final conference of the 2010-2011 academic year, and
once again Northeastern has outdone itself. When I look at the program I simply salivate,
wanting to be there for the presentations – but another item is networking with folks! I
see so many names of colleagues that I really would like to be there and chat with – it’s
interesting that as I look at the names of the presenters at the Consortium conferences I
see a number of names that I was in University with – and have not seen since – but they
remain professional colleagues, and enjoy communicating with them as we try to get the
conference proceedings in final form.
The Consortium is now over a quarter of a century old, and is still experiencing
growing pains! We are all volunteers, and that means that when one of our volunteers is
over-extended something has to slip. I apologize personally to all of you for those slips.
Our mailings in the last year have been behind schedule, and it’s due to a number of items
in the domino effect. We (the Board) are working diligently at trying to resolve such
problems. As a Board we try to deliver the best service that we can within constraints,
and growing pains adds an additional constraint.
As I look back at the last quarter century, thanks are due to many. First of all thanks
are due to the Consortium for their support over that period. Without their support there
would not be any conferences and there would not be a Journal. My sincere thanks to
the Consortium Board for all they have done to ensure that these regional conferences
continue and that we have a quality Journal that records the reviewed papers of the
presenters. Yes, all papers are peer reviewed using a double blind reviewing process to
insure that quality is maintained, and each issue contains a welcome statement from the
individual conference which indicates the acceptance rate.
Thanks are also due to the regional boards and conference committees for the hard
work that goes into preparing a regional conference. (I served as conference chair three
times, and can vouch for the amount of work that is involved.) All of these people are
essential in pulling off a successful conference. Needless to say my contact with the

x
regional conferences at this point is primarily with the contributing editors, and every one
of them (Mike Gousie for this conference) has been wonderful to work with.
Then, there’s the final production. Susan Dean and George Benjamin have been
wonderful to work with over the decades. Susan is very involved with final editing of
manuscript, which sometimes is a real challenge. Her contribution has become very
essential as we went from one conference per year a quarter century ago to now having
ten regional conferences. George handles the coordination with our printer, Montrose
Publishing, and that has been most helpful to me (as I’m located in Germany, a bit of a
geographical displacement.) George does the final checking of the blue line and
coordinates getting the final printing to the individual conferences. Without these two
people and their contributions it would be impossible to deliver the quality that is in the
Journal. Both George and Susan have been wonderful to work with for over the last
couple decades. They continue to be loyal troopers, supporting the Consortium – and I’m
most appreciative of their supprt
As I express thanks, I would be remiss in not thanking our National Partners listed
above. They help to supply the income that assists in keeping conference attendance
costs down, and the Consortium has always endeavored to make the conferences
accessible to those of us with minimal travel budgets. Please express your thanks to them
for allowing the continuation of the Journal and the conferences at minimal cost.
I must also express my thanks to Upsilon Pi Epsilon, the National Honor Society
for Computer Science. They have been contributing for many years allowing the
conferences to have activities for students. A primary example is the student
programming contest which is a part of almost every conference. We appreciate very
much their support of our students.
Of course, we cannot leave out ACM SIGCSE. They provide the “in cooperation”
support that is so very helpful. ACM also has been very supportive in posting the Journal
to the ACM Digital Library.
Thanks also are go to our printer, Montrose Publishing, and the staff there! As I
look at the printing this Spring, and this issue is the final issue, they have supported us
through multiple versions of the Journal – as the manuscript went out to the regions
minor problems were discovered, and each one resulted in minor changes in the final
manuscript, but the patience of the folks at Montrose Publishing must be recognized. My
sincere thanks to both Jerry Golis and Tammy Bonnice who have become real friends
over the many years that we’ve worked together, and they’ve provided support and
pataience to the Consortium.
Attend a regional conference (or two or three) as well as help execute the conference
of your choice. Check out the web site – ccsc.org – , the list of regions and the upcoming
conferences, as well as how to contact the region and indicate how you would like to
assist. The CCSC Board can also use your help. Feel free to contact any Board member.
We can always use the help.
John Meinke
UMUC Europe
CCSC Publications Chair

xi
xii
 WELCOME TO CCSCNE 2011 FROM THE CHAIRS

Welcome to the 16th Annual Conference of the Northeast region of the Consortium
for Computing Sciences in Colleges. The conference is held in cooperation with the ACM
SIGCSE and Upsilon Pi Epsilon Honor Society and is hosted by Western New England
College in Springfield, MA.
The program committee has worked hard to bring you a thought provoking
conference featuring a range of topics, presented in a variety of formats. During the
plenary sessions, Stormy Peters will present Communities: Open Source and Education
Working Together, and Dr. Jane Chu Prey will present What Can We Do? Raising the
Numbers of Women in Computing. We accepted 52% of the papers submitted. These will
be presented in six sessions. We have seven tutorial sessions, a demo session, and two
panel sessions.
In response to last year's record attendance, we have added an additional Friday
morning, pre-conference workshop, bringing the total to four workshops. We will have
a poster session for works in progress. And for our future educators, researchers, and
practitioners, we have a student poster contest and a student programming contest.
We thank the program committee for their dogged dedication, without whom none
of this would be possible. We thank the CCSCNE Board for their guidance and support.
We thank our talented reviewers, who provided prompt and thorough reviews of
submissions. We thank Western New England College for hosting the conference. Most
importantly, we thank those who participate and attend the conference, making CCSCNE
such a dynamic and provocative community.
We hope that you enjoy your visit to Springfield, MA and Western New England
College, and that you leave feeling inspired and fulfilled.

Stoney Jackson, Conference Co-Chair

Western New England College

Karl R. Wurst, Conference Co-Chair

Worcester State University

1
 BOARD MEMBERS — CCSC NORTHEASTERN REGION

Lawrence D'Antonio, Chair . . . . . . . . . . . . . . . . . Ramapo College of New Jersey, NJ

Richard Wyatt, Secretary, 2010-2013 . . . . . . . . . . . . . . West Chester University, PA
Adrian Ionescu, Treasurer 2008-2011 . . . . . . . . . . . . . . . . . . . . Wagner College, NY
Michael Gousie, Editor, 2010-2013 . . . . . . . . . . . . . . . . . . . . . Wheaton College, MA
Paul Tymann, Membership Chair, 2010-2012 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . Rochester Institute of Technology, Rochester, NY
Paul Tymann, Webmaster, 2008-2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . Rochester Institute of Technology, Rochester, NY
Frank Ford, Registrar, 2008-2011 . . . . . . . . . . . . . . . . . . . . . . Providence College, RI
Lonnie Fairchild . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SUNY Plattsburgh, NY
Timothy Fossum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SUNY Potsdam, NY
Scott McElfresh . . . . . . . . . . . . . . . . . . . . . . . . . . . Carnegie Mellon University, PA
Hemant Pendharkar . . . . . . . . . . . . . . . . . . . . . . . . . . . . Worcester State College, MA
Viera Proulx . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Northeastern University, MA
Ingrid Russell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . University of Hartford, CT
James Teresco . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mount Holyoke College, MA
Karl R. Wurst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Worcester State College, MA

2011 CCSC NORTHEASTERN REGIONAL CONFERENCE

COMMITTEE

Stoney Jackson, Conference Co-Chair . . . . . . . Western New England College, MA

Karl Wurst, Conference Co-Chair . . . . . . . . . . . . . . Worcester State University, MA
Mark Bailey, Papers Co-Chair . . . . . . . . . . . . . . . . . . . . . . . . . Hamilton College, NY
Aaron Cass, Papers Co-Chair . . . . . . . . . . . . . . . . . . . . . . . . . . . . Union College, NY
Darren Lim, Papers Co-Chair . . . . . . . . . . . . . . . . . . . . . . . . . . . . Siena College, NY
Lisa Hansen, Panels Co-Chair . . . . . . . . . . . . . . Western New England College, MA
Susan Imberman, Panels Co-Chair . . . . . . . . . . . . . . . . College of Staten Island, NY
Mark Hoffman, Demos Co-Chair . . . . . . . . . . . . . . . . . . . Quinnipiac University, CT

2
 CCSC: Northeastern Conference

Yana Kortsarts, Demos Co-Chair . . . . . . . . . . . . . . . . . . . . . Widener University, PA

Ingrid Russell, Speaker Co-Chair . . . . . . . . . . . . . . . . . . . University of Hartford, CT
Adrian Ionescu, Speaker Co-Chair . . . . . . . . . . . . . . . . . . . . . . Wagner College, NY
Sandeep Mitra, Posters Co-Chair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . The College at Brockport State University of New York, NY
Stanislav Kurkovsky, Posters Co-Chair . . . Central Connecticut State University, CT
Lonnie Fairchild, Posters Co-Chair . . . . . . . . . . . . . . . . . . . . SUNY Plattsburgh, NY
John Willemain, Tutorials and Workshops Co-Chair . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Western New England College, MA
Bonnie MacKellar, Tutorials and Workshops Co-Chair . . . St. John's University, NY
Jonathan Hill, Tutorials and Workshops Co-Chair . . . . . . University of Hartford, CT
Heidi Ellis, Registration Co-Chair . . . . . . . . . . . Western New England College, MA
Lori Postner, Registration Co-Chair . . . . . . . . . . . . Nassau Community College, NY
Darci Burdge, Registration Co-Chair . . . . . . . . . . . Nassau Community College, NY
Mihaela Sabin, Vendors Co-Chair . University of New Hampshire at Manchester, NH
Adam Fischbach, Vendors Co-Chair . . . . . . . . . . . . . . . . . . . Widener University, PA
Matthew Hertz, Publicity Co-Chair . . . . . . . . . . . . . . . . . . . . . . Canisius College, NY
Nadine Hanebutte, Publicity Co-Chair . . . . . . . . . . . . . . St. John Fisher College, NY
Frank Ford, Programming Contest Co-Chair . . . . . . . . . . . . . . Providence College, RI
Leh-Sheng Tang, Programming Contest Co-ChairWestern New England College, MA
Delbert Hart, Programming Contest Co-Chair . . . . . . . . . . . SUNY Plattsburgh, NY
Marc Corliss, Programming Contest Co-Chair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hobart and William Smith Colleges, NY

REVIEWERS – 2011 CCSC NORTHEASTERN CONFERENCE

Ziya Arnavut . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SUNY Fredonia
Barbara Bracken . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wilkes University
Warren Carithers . . . . . . . . . . . . . . . . . . . . . . . . . . Rochester Institute of Technology
Elizabeth Chang . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hood College
Yan Chen . . . . . . . . . . . . . . . . . . Borough of Manhattan Community College/CUNY
Mary Courtney . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pace University
Lawrence D'Antonio . . . . . . . . . . . . . . . . . . . . . . . . . Ramapo College of New Jersey
Bob Dugan . . . . . . . . . . . . . . . . . Department of Computer Science, Stonehill College
Michael Eckmann . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Skidmore College
Emanuel Emanouilidis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kean University
Lonnie Fairchild . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SUNY Plattsburgh
Chris Fernandes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Union College
Adam Fischbach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Widener University
Jill Gerhardt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stockton College
Michael Gousie . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wheaton College

3
JCSC 26, 6 (June 2011)

Delbert Hart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SUNY Plattsburgh

David Hemmendinger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Union College
Mark Hoffman . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Quinnipiac University
Donald Hsu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dominican College
Deborah Hughes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lyndon State College
Susan Imberman . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . College of Staten Island
Jennifer Kay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rowan University
Bo Kim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Daniel Webster College
Bradley Kjell . . . . . . . . . . . . . . . . . . . . . . . . . . . Central Connecticut State University
David Levine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . St. Bonaventure University
Joan Lucas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The College at Brockport
Bonnie MacKellar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . St John's University
Jeff Martens . . . . . . . . . . . . . . . . . . . . . . . . University of Maryland Baltimore County
Robert McCloskey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . University of Scranton
Scott McElfresh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wake Forest University
Barbara Nostrand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . de Moivre Institute
Eileen Peluso . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lycoming College
Sofya Poger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Felician College
Viera Proulx . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Northeastern University
Rathika Rajaravivarma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NYCCT
V Rajaravivarma . . . . . . . . . . . . . . . . . . . . . . . . . . SUNY Farmingdale State College
John Rieffel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Union College
Christian Roberson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Plymouth State University
Ian Sanders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . University of the Witwatersrand
Jeffrey Stone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Penn State University
James Teresco . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Siena College
Carol Traynor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Saint Anselm College
John Vaughn . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hobart and William Smith Colleges
David Voorhees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Le Moyne College
Xiaohong (Sophie) Wang . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Salisbury University
Beifang Yi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Salem State University
Robert Yoder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Siena College

4
WHAT CAN WE DO? RAISING THE NUMBERS OF WOMEN IN

COMPUTING*
FRIDAY PLENARY SESSION

Jane Chu Prey, PhD

Microsoft Research
co-sponsored by ACM/SIGCSE

Did you ever wonder why there are only 18% women studying computing in the
US? Where else are there women in the classroom and do they continue equal
representation in the workforce? From the work in the United States; the current studies
begun in Brazil; and brief overview of other countries, why are certain places successful
in certain levels but not at others? What are the lessons we can learn from our neighbors?
This interactive session will provide a brief overview of the gender diversity
situation including directions and potential best practices with ideas for strategies that can
be used by conference attendees.
It is hoped that the audience will provide information about their particular
communities and ask provocative questions to the audience at large.

SHORT BIOGRAPHY
Dr. Prey received her BS degree from the University of Illinois Urbana in biology
and her PhD from the University of Virginia in Instructional Technology. She was a
faculty member in the Computer Science Department at the University of Virginia for 11
years before joining Microsoft Research in January, 2004 where she led the Tablet
Technologies in Higher Education initiative. Jane is currently responsible for the
development and implementation of the Gender Diversity and Pipeline Strategy for
Microsoft Research. Dr. Prey also spent two years as the Computer Science Program
Manager at the National Science Foundation in the Division of Undergraduate Education.
She is a member of the IEEE CS Board of Governors, the ACM Education Board, and the
CRA Board as well as a number of university advisory boards.
Jane and her husband of 35+years have 3 children [all in computing careers]. She
loves to travel and read cookbooks and murder mysteries.

___________________________________________
*
Copyright is held by the author/owner.

5
COMMUNITIES: OPEN SOURCE AND EDUCATION WORKING

TOGETHER*
SATURDAY PLENARY SESSION

Stormy Peters
Head of Developer Engagement at Mozilla

Open source software is a relatively new development model that employs critical
computer science skills like working in virtual teams, code reading and communication
skills. Open source software projects get many of their new developers as students. While
programs like the Google Summer of Code internships help funnel students to open
source software projects, they also work with universities in a number of ways. Many
professors are teaching their students open source software skills by teaching them to
work on real projects. This requires collaboration between universities, professors and
open source software projects. Come learn and discuss how open source software
communities and educational communities are working together.

SHORT BIOGRAPHY
Stormy Peters is the Head of Developer Engagement at Mozilla where she is
passionate about open web technologies and the developers that use them. Previously she
worked as Executive Director of the GNOME Foundation. Stormy has experience with
open source communities and busienss. She set up OpenLogic Expert Community linking
the community to businesses that use their technology. She also worked at
Hewlett-Packard where she founded and managed the Open Source Program Office that
is responsible for HP's open source strategy, policy and business practices. She received
a Bachelors degree in Computer Science from Rice University. Stormy Peters is a
passionate and frequent keynote speaker on free software related topics.

___________________________________________
*
Copyright is held by the author/owner.

6
 GREENFOOT: INTRODUCING JAVA WITH GAMES AND

SIMULATIONS*

PRE-CONFERENCE WORKSHOP / TUTORIAL PRESENTATION

Adrienne Decker Frances P. Trees

Department of Computer Science & Mathematics and Computer Science
Engineering Department
University at Buffalo, SUNY Drew University
201 Bell Hall 36 Madison Avenue
Buffalo, NY 14260 Madison, NJ 07940
(716) 645-3184 (973)408-3552
[email protected] [email protected]

ABSTRACT
Greenfoot is an interactive Java development environment created by the
developers of BlueJ for use in introductory programming classrooms.
Greenfoot provides a graphical animated environment in which to develop
programs while students only need to use basic Java syntax to control the
actors in their scenarios. Neither students nor educators need to focus on the
graphical elements of the programs. They simply focus on the behaviors of the
actors and the system provides the graphics. This hands-on workshop is
intended to give educators (middle school, high school, and college) an
introduction to the Greenfoot environment, a demonstration of how it can be
used to introduce object-oriented programming to students, and a guided
approach to developing Greenfoot experiences that can be integrated into
existing curricula.The resources that are available to instructors using the
environment will also be discussed.

INTENDED AUDIENCE
This workshop is intended for those instructors (middle school, high school and
college) who have never worked with the Greenfoot environment before. The workshop
will introduce the environment and how it can be used in the teaching of introductory
programming. Knowledge of Java is preferred, but knowledge of another programming

___________________________________________

*
Copyright is held by the author/owner.

7
JCSC 26, 6 (June 2011)

language is sufficient. The workshop is not a workshop about Java, so Java-specific

syntax issues will not be explained, but Java is the language used in Greenfoot and in the
hands-on exercises introduced in this workshop.

PARTICIPANT REQUIREMENTS
If the workshop is not held in a computer laboratory, participants will need to have
a laptop in order to fully participate in the workshop. Prior to the workshop, it
participants should download and install the Java SDK and Greenfoot (both are free).
Instructions and information for all workshop participants are available at
http://www.greenfoot.org/hubs/newyork/events/20110415-workshop.html.

MATERIALS PROVIDED
Each participant will receive hard copies of
• the presentation slides

WORKSHOP AGENDA
[10 minutes] Introduction
[20 minutes] Greenfoot Demonstration
[60 minutes] First Scenario (Hands-on exercise)
• Movement
• Keyboard control
• Random behaviors
[60 minutes] Moving Beyond the first scenario: Advanced Features (Hands-on)
• Image control/advanced animations
• Sound
• Collision Detection
[15 minutes] Publishing Scenarios and Overview of Resources Available
[15 minutes] Questions, Discussion, and Wrap-Up

WORKSHOP LEADERS
Adrienne Decker is a teaching assistant professor in the department of Computer
Science & Engineering at the University at Buffalo, SUNY and the leader of the New
York Greenfoot Hub. She has been teaching introductory object-oriented programming
using Java for ten years to both majors and non-majors and has been teaching using
Greenfoot for two years. Adrienne is a past organizer of five Killer Examples Workshops
at OOPSLA 2005-2009. She currently serves as secretary for the Western New York
CSTA and has given several presentations to the local chapter about various computing
concepts. The University at Buffalo, under Adrienne's initiative was named one of the
first Greenfoot Hubs in the United States. Greenfoot Hubs are local representatives who
have been chosen to educate others about using Greenfoot in the classroom.

8
 CCSC: Northeastern Conference

Fran Trees teaches Mathematics and Computer Science at Drew University in

Madison, NJ. She primarily teaches introductory programming courses and has integrated
Greenfoot into these courses. Fran is actively involved with the AP CS program and has
conducted AP CS teacher workshops and institutes for many years. As CSTA Chapter
Liaison, Fran promotes CS education and professional development to various local CS
communities. Drew University, under Fran's initiative, has also been designated as a
Greenfoot Hub.

9
 AN OPEN SOURCE FIELD TRIP FOR FACULTY*

PRE-CONFERENCE WORKSHOP

Gregory W. Hislop Mel Chua

Drexel University Red Hat, Inc.
[email protected] [email protected]

Clif Kussmaul Sebastian Dziallas

Muhlenberg College Olin College
[email protected] [email protected]

ABSTRACT
Student participation in Free and Open Source Software (FOSS) projects and
communities offers excellent opportunities for learning and for developing
student interest in computing. This workshop will be a virtual field trip for
faculty wanting to know more about FOSS communities, practices, and tools
so they can take advantage of this opportunity. The workshop will provide an
overview of FOSS from a development perspective, hands-on activities to
introduce FOSS tools, and examples of student participation in FOSS projects.
Participants will also be introduced to a variety of resources that support
faculty and student participation in FOSS communities. The intended
audience is computing educators at the college or high school level.

AGENDA
This workshop builds on prior workshops related to student participation in FOSS.
These include an NSF-funded workshop on involving students in the development of
Humanitarian FOSS, and the Red Hat Professor's Open Source Summer Experience
(POSSE). To the extent possible given the venue, the workshop will involve hands on
work. The workshop will include discussion among participants. Topics will include:
1. The Landscape of Open Source Software
• The FOSS movement/concept, history, principles and economics of FOSS
• Educational potential of student participation in FOSS communities
2. Virtual Field Trip: FOSS forges and project sites
• Common features
• Evaluating a FOSS project for possible student participation

___________________________________________

*
Copyright is held by the author/owner.

10
 CCSC: Northeastern Conference

3. Virtual Field Trip: FOSS tools and techniques

• Collaboration and knowledge sharing tools
• Project control tools
4. Examples of FOSS in Education
• Approaches to incorporating FOSS participation
• Examples for students with strong computing background
• Examples for non-majors and students with less computing background
• FOSS for social good.
5. Roadblocks and Resources
• Potential roadblocks and resources related to student FOSS participation
• Getting started
6. Wrap-up
• Discussion about potential for FOSS in computing education.
• Discussion about collaborating on projects.

PRESENTER BIOGRAPHIES:
Mel Chua is part of the Red Hat Community Architecture team. She has worked on a
variety of open source projects including Fedora, Sugar Labs, and One Laptop Per Child.
She is currently organizing the Red Hat Professor's Open Source Summer Experience
(POSSE) to help faculty gain FOSS skills and knowledge.
Sebastian Dziallas is the engineering manager for Sugar on a Stick (SoaS), a
Fedora-based Linux distribution for the Sugar Learning Platform, originally deployed by
the One Laptop Per Child (OLPC) project. Sebastian is also a Fedora packager and the
founder of Fedora's Education SIG. In his free time, he is a student at Olin College of
Engineering in Needham, MA.
Gregory Hislop holds faculty appointments in Information Science and Technology and
Computer Science at Drexel University. He PI on the HumIT project and co-PI on
SoftHum project (see xcitegroup.org). Dr. Hislop has 15 years experience leading
development of curricula in SE, IS, and IT, and spent almost 20 years as an IT
professional.
Clif Kussmaul is Associate Professor of Computer Science at Muhlenberg College and
formerly Chief Technology Officer, Elegance Technologies, Inc. and has 15 years of
experience in software development, FOSS, and computing education He has contributed
to a variety of FOSS projects including Drupal, Moodle, Twiki and Trac. He has also
guided student participation in FOSS.

11
 GETTING READY FOR A UNIT IN ALICE 2.2: WEEK 1 /

EVENT-PROCESSING IN ALICE 2.2*

PRE-CONFERENCE WORKSHOP / TUTORIAL PRESENTATION

Eileen M. Peluso, PhD

Mathematical Sciences Department
Lycoming College
700 College Place
Williamsport, PA 17701
Telephone: (570) 321-4135
Fax: (570) 321-4371
Email: [email protected]

ABSTRACT: Carnegie Mellon's Alice is a freely downloadable teaching tool that uses
3D graphics to create a fun and engaging first programming experience. Incorporating
a unit using Alice in an applications, graphical art, or web design course will introduce
students to fundamental programming concepts and hopefully interest a diverse group of
students in taking additional coursework in computing. Participants in this hands-on
workshop/tutorial sequence who have no Alice experience will learn the Alice 2.2
environment, and all participants will leave with materials for a 2-week instructional unit
in Alice 2.2. As participants work through the unit materials, pedagogical considerations
will be presented and discussed.
The primary audience for the workshop and tutorial sequence is middle school and
high school technology teachers and college professors interested in using Alice in a
course for non-majors. With that said, anyone simply wanting to learn Alice will enjoy
the workshop and tutorial. No previous experience with Alice is needed; however,
participants with knowledge of Alice will find value in the unit materials and other
pedagogical aspects of the workshop. The workshop stands on its own; tutorial
participants should either attend the workshop or have some familiarity with Alice.
Each participant will receive a hard copy and the link to a soft copy of lesson plans,
handouts, and rubrics for four structured activities and a lesson plan and rubric for a fifth,
open-ended activity, all using Alice. Additionally, one general information handout and
two tips-and-tricks handouts will be included. The soft copy will also include completed
activities and additional sample Alice worlds/programs.

___________________________________________

*
Copyright is held by the author/owner.

12
 INTRODUCING CASE STUDY BASED TEACHING IN

COMPUTING CURRICULA*

PRE-CONFERENCE WORKSHOP

Massood Towhidnejad, Salamah Salamah, Thomas B. Hilburn

Embry-Riddle Aeronautical University
600 S. Clyde Morris Blvd.
Daytona Beach, Fl, 32114
[email protected], [email protected]. [email protected]

ABSTRACT
The use of case studies is an effective method for introducing real-world
professional practices into the classroom. Case studies have become a proven
and pervasive method of teaching about professional practice in such fields as
business, law, and medicine. The term “case study” is used in a variety of
ways. In its most naive form, it simply refers to a realistic example used to
illustrate a concept or technique. Although the use of case studies in education
has shown success in the above mentioned disciplines, it is yet to be adopted
in any significant way in the computer science education. One of the reasons
for the lack of use of the case-study approach is the lack of sufficient material
for this purpose. The main aim of the workshop is fourfold. First, the
participants will be introduced to a comprehensive set of case study (the
DigitalHome case study) that can be used throughout a computing curriculum,
emphasizing full software development life cycle. Second, the workshop will
include an interactive discussion of how to use the case study artifacts and the
associated teaching support materials to complement their teaching artifacts
throughout their computing curriculum. Third, the attendees will be asked to
participate in the assessment of case study material and make
recommendations for improving both the material and the delivery. Fourth, we
intend to follow up with the attendees who are interested to participate in our
project, by incorporating some of the case study material as part of their course
work during the fall 2011 semester.

___________________________________________

*
Copyright is held by the author/owner.

13
JCSC 26, 6 (June 2011)

INTRODUCTION
Many computing programs have a software engineering course that involves a
software development project in which students are grouped into teams to work on a
semester or year-long software development project. Unfortunately, this is too often
isolated from the rest of the curriculum and does not form a real-world basis for the entire
curriculum. As a result, these programs produce graduates who are familiar with the basic
theoretical concepts in software development, but lack the skills to apply these concepts
in real-world development environments. Therefore, it is imperative that computing
curricula introduce professional and real-world education into the academic programs.
The use of case studies is one widely-used method for introducing real-world professional
practices into the classroom.
Although many computing text books include the use of case studies to explain
ideas, these cases seem to serve a specific purpose (e.g., discussing programming
constructs, software planning or requirements analysis, design, risk analysis, construction
or testing issues by using simple examples that are quasi-realistic), they often lack the
following:
• Realistic artifacts (often space does not allow providing a complete requirements or
design document)
• Completeness (covers only a portion of the life ]cycle, and not an end ]to ]end),
with a focus on design and implementation
• Ability to decouple from the text and apply in ways not intended by the author
• Techniques for integration into course activities or into the curriculum as a whole
• A scenario format that would motivate students to get engaged in problem
identification and solution.
• Guidance to the instructor on how to use the case study to teach a course topic or
concept
The DigitalHome case study [1, 2] is intended to address these shortcomings by
providing a complete set of artifacts associated with software development as well as
providing case modules that can be used by faculty in teaching different subjects in a
computer science curriculum. Each case module represents a mini case study and is
associated with a specific teaching subject and learning objectives (e.g. requirements
analysis, object oriented design, testing, team building...). Case modules also include an
exercise booklet and a set of guidelines to assist the instructor in teaching the session
.
WORKSHOP GOALS
The aim of this workshop is to advance the understanding of case study based
teaching and to introduce the participants to a comprehensive case study in software
development. The goals of the session are:
1. Describe the background and effectiveness of case study techniques in professional
education.
2. Introduce the DigitalHome software development case study material.
3. Initiate the use of case study material (artifacts and case modules) by faculty in a
variety of computing programs and courses.

14
 CCSC: Northeastern Conference

4. Assess the quality and utility of the case study material by the participating faculty.
5. Assist interested participants in developing of plans to incorporate case study
material as part of their course offering during fall 2011 and beyond.

WORKSHOP DETAILED AGENDA

The workshop’s detailed agenda is as follows:
1. (20 minutes): Facilitators and participants introductions and backgrounds
2. (15 minutes): Introduction and background to case study teaching, previous uses,
and advantages.
3. (30 minutes): Description of the DigitalHome case study, its origins, artifacts, and
case modules.
4. (50 minutes): Group exercise on the use of the DigitalHome case modules.
Participants will be broken into groups of 4-5, based on their major area of teaching
(i.e., programming, data structure algorithm, software engineering, etc). Each group
will be provided an appropriate case module to discuss and evaluate. The
participants will be asked to discuss the utility of the case module as well as suggest
necessary changes to improve both the particular case module as well as the
DigitalHome case study as a whole.
5. (40 minutes): Group discussion; groups report back on their activity in # 4 above.
6. (25 minutes): Workshop summary, final thoughts and discussions.
7. (60 minutes, after the completion of workshop): Development of initial plan for
fall 2011 for interested parties.

ACKNOWLEDGMENT
Initial work on the DigitalHome case study was funded as part of the NSF project:
“The Network Community for Software Engineering Education” (SWENET) (NSF
0080502). In addition, the current work on the case study is funded through NSF’s (DUE-
0941768) “Curriculum-wide Software Development Case Study”.

REFERENCES
[1] Thomas B. Hilburn, Massood Towhidnejad, “A Case for Software Engineering”.
In the proceeding of the 20th Conference on Software Engineering Education and
Training (CSEET), Dublin, Ireland, July, 2007, 107-114.
[2] Thomas B. Hilburn, Massood Towhidnejad, Salamah Salamah, “The
DigitalHome Case Study Material h In the proceeding of the 21st Conference on
Software Engineering Education and Training (CSEET), Charleston, SC, April,
2008, 279-280.

15
 IMPROVING THE STUDENT SUCCESS AND RETENTION OF

UNDER ACHIEVER STUDENTS IN INTRODUCTORY

COMPUTER SCIENCE *

Carolee Stewart-Gardiner
Kean University
1000 Morris Ave., Union, NJ 07083
(908) 737-3795
[email protected]

ABSTRACT
In many universities, student success and retention in computer science are an
ongoing concern to faculty. Finding best practices to improve the success and
retention rate of less than the best students, has been a particular challenge.
This study, conducted at a public teaching university, has focused on
enhancements to introductory computer science courses to aid in the retention
of students in STEM (Science, Technology, Engineering, and Mathematics)
majors. The majority of these students had no prior programming or computer
science experience. All computer science majors and some other STEM majors
are required to take the first course in programming (CS0). The PLTL (Peer
Led Team Learning) methodology, successful with Chemistry majors
elsewhere, was used to improve student success in CS via collaborative
workshops. Quantitative grade data and attitudinal surveys are combined to
show the success of this study on the student success rates, which contributed
to the retention of these students over the past 3 years. Over 80% of the
students who attended PLTL workshops, said that future students should attend
PLTL workshops in this and follow on CS courses. The progress of the
students' success, and the enthusiasm of the students in introductory CS classes
shows the contribution PLTL has made in the retention of STEM students.

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

16
 CCSC: Northeastern Conference

INTRODUCTION: Prior Work in Peer Led Team Learning

The development of Peer Led Team Learning at CCNY was for improvement of
student learning and reduction of attrition in Chemistry. [2] CCNY is a public teaching
university with comparable student profile and demographics to Kean University. PLTL
at CCNY involved three overarching principles:
P Peer leaders were facilitators of student learning, not instructors, or answer givers,
to reduce dependency on instructors, and increase self and group scientific
discovery.
P Workshops were to consist of well formed problem solving exercises, highly
integrated to the course material, for small group collaborative problem solving.
P Trained by faculty, Peer Leaders and faculty developed the PLTL exercises to aid
in peer leader growth, and assurance that exercises were tuned to the course
material.
There is significant literature that demonstrates the benefits of adding PLTL
workshops to a scientific course like Chemistry. [3] CCNY contributed materials to help
us learn the PLTL workshop concepts. Computer Science is relatively new to using this
PLTL model. Two successful studies have used PLTL in Computer Science, and are
documented. One was a consortium of competitive colleges and universities, which used
the Emerging Scholars model and recruited strong minority students.[4] Rutgers
University was a member of that group. We started with a set of PLTL workshops from
Rutgers, for CS1 classes, and we used a few at the beginning of our process. The other
study at Georgia Tech [1] was with students whose mean SAT score was 1340 to 1356
in their 3 year study. Kean is a public teaching university. We wanted to see if PLTL
could help ordinary students (with Math SAT of 500-599), and under achiever students
(at risk of earning C or D, but had potential to earn B or C). We wanted to improve their
understanding of algorithms, and problem solving, and their success in introductory
Computer Science. We also wanted to see if PLTL could improve student interest, so they
would want to stay in their STEM majors. It is one thing to assist scholar Computer
Science majors who are generally doing ok in the course to do better, but it is another
level of challenge to involve the under achiever students at risk of earning a "C" or "D",
to help them learn enough to want to stay in their STEM major.

PLTL AT KEAN UNIVERSITY

According to Kean University's Mission statement: "Kean offers a wide range of
demanding programs that provide the high quality of instruction and academic support
services necessary to assure its diverse students the means to reach their full potential,
including students from academically disadvantaged backgrounds, students with special
needs, and adults returning or entering higher education." So the challenge for us in the
sciences is how do we create a program of high quality with students, many of whom are
from academically disadvantaged backgrounds? We provide various means of academic
support. In the past this has been primarily tutoring. But in the sciences, we tend to find
that while students do get help to solve the particular homework they are trying to
accomplish, tutoring does not generally produce self reliance. And the tutored students
often withdraw, or get "D"s. Also many students who are getting by with a "C" do not go

17
JCSC 26, 6 (June 2011)

to tutoring. So in the Fall 2006, we started the PLTL program at Kean, funded by the
Epsilon Corps grant from NSF(DUE #431637). It was to provide an alternative to
tutoring for math and computer science introductory courses. It would be for enrichment
for all students, as a means to encourage students to continue as STEM majors. The math
PLTL workshops did not continue, but the computer science PLTL workshops did.
In 2006, and 2007 overall DFW rates (% of students with D, F, or W grades) in
introductory CS classes were higher than we were comfortable with. PLTL might be a
way to aid in improving student outcomes, lowering DFW rates, while raising the success
rates of the entire class. So those were our goals, to lower DFW rates, to improve the
grades of all the class in the process, and to improve retention in the STEM majors.
Because PLTL focuses on all of the students, not just the weakest or the strongest
students, we felt it was a good candidate tool.
Kean is mainly a commuter university where about 2/3 of the students work for pay.
About 1/3 of our CS0 students work 15 to 25 hours a week and 1/3 over 25 hours a week.
Convincing them to participate in enrichment is difficult. We used surveys to determine
what they thought of PLTL. The main reason why they did not attend was they had
conflicting times for work or classes, and they had many other commitments. [7]

Participation Was Key

It took 6 semesters of trial and error at Kean until we found a way to improve the
voluntary participation by the students. In Fall 2007, the workshops were offered in CS0
(Computing Fundamentals - with Java programming) classes, but almost no one came.
The students thought these workshops were tutoring for D and F students, so the students
could pass exams, or turn in solutions to the assignments. In January 2008, the PLTL
leaders were given 20 minutes in each section, to lead a demonstration workshop problem
with the class, but almost no one came to the regular PLTL workshops. The faculty and
PLTL leaders spent the early semesters creating and improving workshop problems. In
Fall 2008 the department agreed that all faculty of CS0 would give a small amount of
extra credit to students who participated in PLTL workshops. Even with extra credit, very
few students came. But just before the midterm, 4 students, concerned that they might not
do well on the midterm, asked if they could have PLTL sessions on Saturday. We found
a Saturday PLTL leader. This was the first time a group of 6 students (4 regulars) began
to act like a PLTL group was supposed to. They came on a regular basis, and although
4 of the 6 were in jeopardy of a "D", they got a "C", and the 2 better students got a "B".
But those 6 were not common. In Spring 2009 only 2 students out of 80 came to PLTL
workshops regularly. So in Fall 2009, in 1 section, the first half hour of the PLTL
workshop was scheduled inside the regular class time. That worked! One section had
everyone exposed to a PLTL workshop in class time, and the students could elect to stay
or go at the end of class time. The other 2 PLTL workshop problems were immediately
following the class, for those who chose to stay. Each student in that section got a chance
to experience the value, the fun, and the collaborative learning of the PLTL workshop
problems, in a peer led environment. About half of that section stayed. And of course at
exam weeks, the workshop swelled. So students who have to work 20 to 30 hours a week,
took an hour out of their week to stay for PLTL workshop, because they experienced the
value in it. The Spring 2010 semester we did it for 2 sections. It was the first time we had

18
 CCSC: Northeastern Conference

50% participation of 2 sections. Word got around, and about 6 students from another
professors' class came regularly. PLTL was finally on the way to real participation.
Peer leaders took attendance at the workshops, and were charged with keeping
everyone participating. So we had all the data of who went to workshops, and we had
their grades. This enabled us to do some comparative grades between the half of the class
who chose to stay, and the half that did not. Of course, the half who did not stay for PLTL
did get 1 short workshop problem. And the students who did stay for PLTL were self
selected, and could be called the more motivated half of the class. Because most of the
students from the other professors' classes were so few, the grades data here for Fall 2009
and Spring 2010 is for my 2 sections of 21 each, totaling 42 students a semester.

PLTL Workshops Improved Student Ranking

One challenge of the study was to prove that our students could improve during the
semester with PLTL workshops. Since the first half of the course is easier, students who
only earn a "C" or a "D" on the midterm, are at risk of failing the final exam. So we
looked at how the PLTL workshops in the second half of the course could assist all
students, and especially the at risk students, to do their best in the more difficult half of
the course. Figure 1 shows that 60% of PLTL students clearly improved their rank from
midterm to final, whereas 50% the non-PLTL half declined in their class ranking by the
time of the final exam.
The students often come from a background where teachers gave predigested, top
down concepts with pre-solved problems, and students were expected to be able to follow
the solution and mimic the solution on a similar problem in the lab. Students often said
that the word problems on the midterm were too difficult to solve, even if they knew the
basic skills of programming. Getting these students to think for themselves is a real
challenge. The PLTL leaders and the faculty create leading questions for the workshop
exercises that enable the students to discover the solutions. For example "What do you
know about the problem? " or "What are the main steps this solution needs to do?"

PLTL Improved the Success For All Students in Intro Computer Science (Fig 2)
Algorithmic thinking is a very important part of this course, because CS0 is required
for many STEM majors in their curriculum, as well as Computer Science majors.
Therefore our emphasis is on algorithmic thinking in PLTL enrichment and problem
solving skills for all students. Because of the collaborative methods used in PLTL

19
JCSC 26, 6 (June 2011)

workshops, and the fact that there is only one PLTL leader for 6 to 20 students, the better
students were encouraged to assist their peers also, in the problem solving efforts.
Because PLTL workshop problems are not graded, these efforts were strictly focused on
enrichment for all participants. Some students thought it was just "more work" but that
is why the PLTL leaders, with the professor, worked every week to try to make the
problems both relevant to the material of the week of the course, and also some degree
of fun.[6] We used pairs with think/pair/share, or groups of 3 or 4 to work on an
algorithm of various games, and frequently had two small groups share their solutions
with the others on the board. Then students critique what was different about the two
algorithms.
In Fall 2009 the best students were not staying for the PLTL sessions. But by the
final exam, the PLTL students caught up to the non-PLTL students and had the same
average on a difficult final exam. One way to deal with students who do not do well on
exams is to make the final easier, but we were determined to improve the student
understanding instead. So in Spring 2010 midterm and final exam grades, the two halves
of the class were almost the same at the midterm, but by the time of the final exam, the
PLTL attendees had 10 more points average on the final exam.

PLTL Workshops Reduced the DFW Rates

The best illustration of the progress of our PLTL program through the semesters is
the line graph above (figure 3). The lines show that through the semesters, the progress
of students without PLTL was non-existent, but the progress of the students with the
PLTL was significant. Those who rarely or never attended PLTL had aggregate DFW
rates of over 30% each semester, but the F and W rates went to 0% for those attending
PLTL, and to 5% to 12% for the D grades. And the C grades went from around 30% each
semester without PLTL, to about 45% to 58% for the frequently attending PLTL students.
The A's and B's improved for the PLTL students, but were not as significant.

Student Assessment of PLTL Workshops

The anonymous student attitude surveys clearly showed a positive assessment about
PLTL workshops. Improving the interest of under achievers has a clear link to improved
success. In figure 4 over 80% of the PLTL attending students felt that future students in

20
 CCSC: Northeastern Conference

the same introductory computer science course should attend PLTL, as well as follow on
computer science courses should provide PLTL workshops. In our commuter college,
55% felt that PLTL workshops improved a sense of community. And in figure 5, over
70% of the PLTL attending students said PLTL helped them do better on exams, and over
65 % said it helped them in the course overall. The long term attitude was that about 65%
of the students attending PLTL workshops felt it helped them remain a STEM major,
which is one of the original goals of the PLTL introduction. (The only students who
strongly disagreed, went to PLTL but were not STEM majors. ) Each year that the
students attended in stronger numbers, the surveys also improved. The students clearly
indicated that the conflict of student commitments of work and other classes, does remain
a difficult problem to solve.

PLTL Improved Leadership Of the Leaders

With weekly leader workshops, the PLTL leaders learned how to ask questions,
instead of answering them, and become small group facilitators, rather than to give
solutions to students. When the students had difficulty creating algorithms, the PLTL
leaders learned to guide the students by having them re-read the problem, and discuss the
start of a solution among the small group. It is imperative that PLTL leaders help the
professor to create/improve the workshop problems. As PLTL leaders work each week
to improve the exercises with leading questions, they help direct students to discover the
solutions to problems in the process.

The leaders learn to see the problems as a discovery tool, and the leader role to
enable the discovery by the students. One of the leaders said in the leader survey that "I
have begun to learn what leadership and facilitator really mean. I used to think it meant
showing the student the solution."

CONCLUSIONS
Our experience with PLTL in early Computer Science classes was very encouraging.
When PLTL is used with typical students, who have many commitments, and less than

21
JCSC 26, 6 (June 2011)

excellent backgrounds, the success rates of all students can improve, with lower DFW
rates, and improved retention attitudes of STEM majors. There are some requirements.
P PLTL workshops need to be clearly integrated into the course.
P Computer Science focuses on algorithmic thinking & collaborative problem solving.
P PLTL requires the participation of the students! The workshops must be of clear
benefit to the students.
P PLTL leaders need to be trained as facilitators, not tutors, and PLTL can grow new
leaders from among the ranks of the PLTL students.
P PLTL workshops can be grown, and improved, and shared, so that everyone does
not have to invent them from scratch.
P PLTL needs support, because the student leaders need to be paid, and at least one
faculty member needs to be a coordinator of the students and workshops.
Creating the workshops is a major task. And tailoring the workshops to encourage
algorithmic thinking has been a serious effort of our leader workshops. So, we began with
some workshops from Rutgers, and others are available from University Wisconsin,
Milwaukee and Indiana University.[5] Our PLTL workshops are available for sharing
with other universities who wish to use them. (www.kean.edu/~cstewart/cps1231.html)

REFERENCES
[1] Biggers, M., Yilmaz, T., Sweat, M., Using collaborative, modified Peer Led
Team Learning to improve student success and retention in intro CS ,
Proceedings of 40th SIGCSE ACM Symposium on Computer Science Education,
Vol 41, (1) ,2009.
[2] Gosser, D., Cracolice, M., Kampmeier, J., Roth, V., Strozak, V., Peer Led Team
Learning: A Guidebook. Upper Saddle River, NJ, Prentice Hall, 2001.
[3] Hockings, S.C., DeAngelis, K.J., and Frey, R.F. Peer-Led Team Learning in
general chemistry, Journal of Chemical Education, vol 85, (7), 2008.
[4] Horwitz, S., Rodger, S.H., Using Peer-Led Team Learning to increase
participation and success of under-represented groups in introductory computer
science , Proceedings of 40th SIGCSE ACM Symposium on Computer Science
Education, Vol 41, (1), 2009.
[5] Horwitz, S., Rodger, S. H., Huss-Lederman, S., Peer-Led Team Learning in
computer science, www.pltlcs.org, 2007.
[6] Huss-Lederman, S., Chinn, D., Skrentny, J., Serious fun: Peer-Led Team
Learning in computer science, Proceedings of 39th SIGCSE ACM Symposium on
Computer Science Education, Vol 40, (1), 2008
[7] Stewart-Gardiner, C., Using Peer Led Team Learning to assist in retention in
computer science classes, The Journal of Computing Sciences in Colleges 25th
CCSC Eastern Conference, Vol 25, (3), 2009.

22
 USING VISUAL LOGIC©: THREE DIFFERENT APPROACHES

IN DIFFERENT COURSES - GENERAL EDUCATION, CS0, and

CS1 *

Dee Gudmundsen Lisa Olivieri Namita Sarawagi

The College of Saint Chestnut Hill College Rhode Island College
Rose Philadelphia, PA Providence, RI 02908
Albany, NY 12203 19118 401-456-9865
518-454-2105 215-248-7092 [email protected]
[email protected] [email protected]

ABSTRACT
One of the main challenges students in introductory programming courses face
is learning the syntax of a particular programming language, such as C++ or
Java, while they simultaneously try to understand the fundamental logic of
programming constructs. Visual Logic© (www.visuallogic.org) is an
interactive, graphical, flowchart tool that can be used to address this challenge.
In this paper, we present three different courses, developed independently at
three different colleges, using Visual Logic with completely different
approaches. The first course is a General Education CS course introducing
algorithmic thinking. The second is a CS0 course which uses Visual Logic for
six weeks and then transitions to Python. The third course is a CS1 course
which uses Visual Logic as a tool to teach programming concepts and to
brainstorm solutions. Visual Logic is not taught separately, but integrated with
the various Java control structures. Once students gain an understanding of
basic programming logic concepts, the transition to a programming language
such as Python or Java, in the same course or the next course, is often much
easier for them. Though these are different approaches in three different
courses, using Visual Logic has shown an increase in the interest level of the
students and enhanced their learning.

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

23
JCSC 26, 6 (June 2011)

FEATURES AND ADVANTAGES OF VISUAL LOGIC

Visual Logic is an interactive, graphical
software tool that enables a novice programmer
to develop interactive, executable flowcharts
[9]. Visual Logic is extremely user-friendly,
requires minimal syntax, and is very simple for
a student to learn. Students are introduced to the
logical structures found in most programming
languages: input and output (console, dialog
box and file), arithmetic and logical operators,
assignment statements, if/else statements, loops,
arrays, functions and user-defined procedures.
Students select a statement from a drop-down
menu and enter only constants (numeric and
string), variable names, arithmetic expressions
or Boolean expressions. Visual Logic includes a graphics package containing LOGO style
commands which can be used to create colorful graphics programs.
Although other good programs, such as Alice and Scratch have been developed for
introducing programming, Visual Logic includes some unique features. The program
introduces students to the fundamentals of flowcharting, a traditional method of
graphically illustrating a program. The "visualness" of the flowchart symbols and
connecting arrows clearly illustrate the logic flow. Students can easily correct the logic
of a program, rerun the flowchart, and determine if it produces the desired output.

RATIONALE FOR USING VISUAL LOGIC

The purposeful, graphical flowchart symbols provide a visual perspective to a
conceptually abstract discipline. Since most students are visual learners [2], attempting
to teach abstract concepts using a traditional programming language, which is very
textual, adds to the complexity of the learning experience of many students. The
simplicity of the Visual Logic program and its straightforward correlation with the
essential structures of most programming languages provides a good visual introduction
to programming concepts [3]. The analysis, algorithm development, and documentation
skills acquired through flowcharting can be applied to other languages in future courses.

IN A GENERAL EDUCATION COURSE

"Introduction to Algorithmic Thinking" is a new, General Education, Computer
Science course at Rhode Island College. It provides an alternative to the traditional
Introduction to Computers computer literacy course, for students of all majors.
Computational Thinking is emerging as a basic skill that should be developed in K-12 and
General Education [6], and this course in Algorithmic Thinking addresses this need at the
introductory Computer Science level.

24
 CCSC: Northeastern Conference

The course introduces algorithmic thinking, the study of step-by-step procedures for
solving problems, with simple applications from various disciplines. The major topics
covered are: algorithms, flowcharts, top-down design, selection, repetition, and
modularization. Students interested in programming may continue on to take the CS1
course which introduces Object Oriented Programming in Java.
Students taking this General Education course come from varied backgrounds and
lack prior programming experience. This gives rise to two major challenges: (i) finding
simple, real-world, interesting problems for in-class examples and student assignments
and (ii) finding the appropriate programming language/tool for students to develop
programs and test their logic. Using Visual Logic solves (ii) very effectively due to the
advantages and interesting features described above.
Some examples of interesting problems from various disciplines (e.g. arts, sports,
math, science) used in the course, and the underlying algorithmic concepts in their
solutions, were on display in a CCSCNE 2010 poster [7]. It also included an example to
teach procedures in a simple program using Visual Logic's LOGO capability.

The first programming assignment is

a Mad Libs program, shown below, which
introduces the concepts of input, output
and variables in a fun and familiar game.
Visual Logic makes it easy to come up
with quick interactive solutions, thus
allowing students to try out many more
problems than they would have using a
traditional programming language.

Students from various majors as well as undecided students have the opportunity to
successfully learn programming fundamentals in a (almost) syntax-free environment, thus
focusing on the real content of the course, "algorithmic thinking" applied to problems
from various disciplines. The same reason makes this one of the most enjoyable and
satisfying courses to teach.

25
JCSC 26, 6 (June 2011)

IN A CS0 COURSE WITH PYTHON

"Introduction to Programming Logic", the CS0 course at Chestnut Hill College is
a prerequisite for Computer & Information Science/Technology majors and other
interested students who have had no previous programming experience.
The goal of most introductory programming courses is to develop students' problem
solving skills and to express algorithm solutions using a computer programming language
[5]. As the complexity of the programming languages has increased, students' struggles
with the syntax of the languages have made them lose sight of the main purpose of the
course - solving problems through algorithm development. This problem is not unique
to students at CHC[8].
To introduce prospective computer science and computer technology majors and
minors to the essence of computer programming without overwhelming them with tedious
syntax, the department revised their CS0 course so that students could focus on algorithm
development using Visual Logic in the first part of the semester and then implement these
algorithms in Python during the second half of the semester. The rationale for introducing
a programming language as a complement to Visual Logic and for choosing Python in
particular reflects an effort to meet the needs of the student population, adhere to
curriculum requirements, and prepare students for the workplace. Students are often
limited in the number of major courses that can be required of them. In order to provide
students with exposure to a variety of programming languages, the department decided
to introduce students to a "real" programming language, believing that, as Agarwal et al
[1] state "the use of a higher-level programming language gives the students a firmer
foundation".
Much consideration was given to choosing the language that would complement
Visual Logic. Python was chosen for its simple syntax and its growing use in industry.
Many of Python's features, such as not requiring variable declarations, coincide with the
way that Visual Logic operates. Students are introduced to Python through a simple
manual developed by the instructor that emphasizes the relationship between the concepts
learned in Visual Logic and reintroduced in Python. As students learn Python, the basic
programming structures introduced in Visual Logic are reinforced as they examine them
in a different context. Students revisit programming problems they completed in Visual
Logic and recode them in Python, helping them see the connection between flowcharting
and programming in Python.
In general, students welcome the ease of Python as they begin learning the
language-- though some miss the graphical interface of Visual Logic. The following
simple Visual Logic
flowchart program
and its Python
program equivalent
demonstrate the
coding of a FOR loop
using a negative step,
random number
generation, and screen
output.

26
 CCSC: Northeastern Conference

The implementation of a few programming structures in Visual Logic differs from

their implementation in Python. For example, although Visual Logic provides a structure
to represent arrays, Python does not support arrays, but instead supports a more powerful
list structure. The list structure cannot be applied in Visual Logic, but the exposure to
arrays in Visual Logic helps students understand the differences between arrays and lists
and gives them an appreciation for the power of lists.
By the end of the course students are generally comfortable with the basic structures
of programming and are ready for CS1, which focuses on an early introduction to objects
in Java. Although the use of Visual Logic in the first CS course has not solved all the
problems of preparing students for Java, many students find the transition less daunting.

IN CS1 (with JAVA)

"Introduction to Programming", currently taught in Java, is the first required course
in the Computer Information Systems and Computer Science majors at The College of
Saint Rose. This objects-late course is also the required general education course for
mathematics majors, all science majors, math adolescence education majors, elementary
education math concentrators, and criminal justice forensics majors.
Students enrolled in "Introduction to Programming" have had very little preparation
for learning a syntax-driven programming language and often have difficulty progressing
from a problem statement to a sequence of logical steps. Visual Logic can be used to
demonstrate the how a control structure works before teaching the related Java syntax.
Visual Logic does an excellent job of illustrating difficult concepts such as when to use
a decision versus a loop and the transfer of execution that happens when you call one
method from another. The student sees separate flowcharts and the interaction between
the flowcharts. This visual tool makes the concept come to life!
Visual Logic can also used as a tool when brainstorming a solution to a problem as
a class and also when students do independent homework. Instead of proceeding directly
to writing the program in Java, students develop a Visual Logic flowchart. They are able
to incrementally test this solution by "running" the flowchart. Once students have arrived
at the correct sequence of steps to produce the desired output, then they can proceed to
writing the program, using appropriate Java structures/techniques for each step. The very
last step is dealing with syntax errors!
To illustrate, one problem that was solved as a class: "Write a program that
calculates the amount a person would earn over a period of time if his or her salary is one
penny the first day, two pennies the second day, and continues to double each day. The
program should display a table showing the salary for each day, and then show the total
pay at the end of the period. The output should be displayed as a dollar and cents amount,
not the number of pennies. Input validation: Do not accept a number less than 1 for the
number of days worked."[3]
This Visual Logic flowchart was developed incrementally. The final flowchart
shows a do-while input validation loop and a for loop to generate tabular output.
Students in this course struggle a little with Java syntax as all students do, but it is
at a much later point in the process, only after they have developed the correct logic.

27
JCSC 26, 6 (June 2011)

After using Visual Logic, they

have confidence in the steps and
sequence of their solution. Using
Visual Logic encourages them to
categorize Java control structures
and techniques (Scanner input,
JOptionPane output, etc) as
building blocks that can be used to
create the overall solution of any
problem.

CONCLUSION
This paper describes three
different CS courses using Visual
Logic with different approaches.
In all courses, novice
programmers were able to learn
basic programming concepts in a
visual, (almost) syntax free
manner, thus focusing on the logic
of the problem's solution. This
minimized student frustration and
enhanced learning. In student course evaluations, most of the comments were extremely
positive. A typical example: "Visual Logic was really helpful when it came to learning
the material in class and working on it by myself" speaks of the ease and success of this
versatile tool. From the instructor's perspective, using this visual tool generates far more
student enthusiasm and understanding than the traditional approach of explaining syntax
and fixing errors.

REFERENCES
[1] Agarwal, K., Agarwal, A., Celebi, M., Python puts a squeeze on Java for CS0
and beyond, J. Comput. Small Coll., (23:6) 49-57, 2008.
[2] Cardellini, L., An Interview with Richard M. Felder, Journal of Science
Education 3(2), 62-65, 2002.
[3] Crews, T., & Butterfield, J., Using technology to bring abstract concepts into
focus: A programming case study. Journal of Computing in Higher Education,
13(2), 25-50, 2002.
[4] Gaddis, T., Starting Out with Java: From Control Structures Through Objects,
4e, Boston, MA: Addison Wesley, 2010.
[5] Gomes, A., Mendes, A. An environment to improve programming education. In
Proceedings of the 2007 International Conference on Computer Systems and

28
 CCSC: Northeastern Conference

Technologies, (CompSysTech '07), ACM, New York, NY, USA, Article 88 , 6

pages
[6] Qualls, J., Sherrell, L., Why computational thinking should be integrated into the
curriculum, Journal of Computing Sciences in Colleges, 25(5), 66-71, 2010.
[7] Sarawagi, N., A general education course - "Introduction to Algorithmic
Thinking" - using Visual Logic, Journal of Computing in Sciences in Colleges,
25(6), 250-252, 2010.
[8] Shackelford, R., LeBlanc, R., Introducing Computer Science Fundamentals
Before Programming, Proceedings of FIE '97, 285-289, 1997.
[9] Visual Logic, © PGS Systems, LLC, www.visuallogic.org/, retrieved Nov. 11,
2010.

29
 ALIGNING GENERATIONS TO IMPROVE RETENTION IN

INTRODUCTORY COMPUTING COURSES*

Christian Roberson
Computer Science and Technology Department
Plymouth State University
Plymouth, NH 03264
[email protected]

ABSTRACT
Incoming students at higher education institutions exhibit different core traits
than those of students from past generations. The misalignment that can occur
between the traditional course approach and the expectations of students is one
of the contributing factors to low student retention, especially in Computer
Science and Information Technology programs. The author's school observed
a clear misalignment of this nature in a first semester introduction to
computing course. After researching the core traits of Generation Y and
Generation Z, the department underwent a process of redesigning their course
to better meet the needs of today's students. This paper provides an overview
of the core traits of the current generation of incoming higher education
students, describes in detail the major changes involved in the restructuring of
the introduction to computing course, and examines the positive results of their
implementation during the first offering of the revised course.

INTRODUCTION
Retention of first-year students has been a problem for many higher-education
institutions, especially for computer science and information technology programs. At
Plymouth State University we have observed a similar trend in the Computer Science and
Technology Department. Many students come to the university and sign up to be a
Computer Science major or an Information Technology major, but end up changing
majors after only one or two semesters. Students tend to refer to our freshmen courses

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

30
 CCSC: Northeastern Conference

as "weed-out" courses, implying they are more of a barrier to entry into the program than
courses designed to stimulate interest in the field.
One issue that we have observed at Plymouth State is that there is a clear
misalignment between the expectations of faculty teaching the introductory courses and
the students taking the courses. This issue is compounded by the generation gap between
these two groups. Most faculty members are either from the Baby Boomer generation or
Generation X while most new students entering college are at the trailing edge of
Generation Y or the leading edge of Generation Z [7]. Because the core traits of these
two groups are fundamentally different, it becomes difficult for faculty to effectively
work with students because of differences in learning styles.
At Plymouth State we have recently taken a long and detailed look at our first
semester course for both Computer Science and Information Technology majors:
Computing Fundamentals. This paper details the changes made during the course
revision in an attempt to address the challenge of teaching computing to this new
generation of students and discusses the results observed from the initial offering of the
newly revamped Fundamentals course.

STUDENT GENERATIONS
Most students in college today are from Generation Y, but we are starting to see
some students from Generation Z. Generation Y (Gen-Y , Generation NeXt, or the
Millennial Generation) consists of people born between 1981 and 1994, and Generation
Z (Gen-Z or the Neo-Millennial Generation) consists of people born between 1994 and
now [7]. These students, sometimes referred to as digital natives, have been exposed to
technology their entire lives. They have been inundated with choices and flexibility since
childhood and in many cases have grown up with a very different support system than the
one most faculty were raised with.
These students exhibit many traits that are positive including: confidence,
team-oriented thinking, and a strong sense of achievement [5]. They can be extremely
confident in their actions, and are motivated by future success. They naturally are
inclined toward team-based activities, and favor in many cases being able to interact and
work with others. They also are motivated by achievement, and enjoy a sense of
accomplishment on completion of tasks. These are all traits valued by professionals in
the computing industry, especially given the increase in the need for employees that have
these types of "soft" skills [3], and should be encouraged throughout their course of study
in higher education.
There are also several perceived weaknesses among Gen-Y and Gen-Z students [5].
They tend to be very pragmatic, and will adapt to find a solution that works rather than
seek out the "best" solution. They can be very skeptical by nature and may question
information they receive from a faculty member. There is an expectation of excellence
for students and they have been brought up to believe that this success may come with
little effort on their part. Students may not exhibit the politeness and deference expected
to the faculty. They may express interest only in achieving the minimum expectations
associated with a particular goal, and need to understand the importance of a task before

31
JCSC 26, 6 (June 2011)

being convinced to undertake it. These students are used to and expect frequent and
immediate feedback on their activities.
A misalignment is likely to occur when a faculty member teaching a course,
especially at the introductory level, fails to take into consideration the motivations and
weaknesses of their target audience (Gen-Y and Gen-Z students) [4]. Modern students
are accustomed to gathering information rapidly from various multimedia-driven sources,
where faculty tend to deliver information in a more slow and controlled fashion. Based
on the modern, hyperlinked Internet and social network model of sites like Facebook and
Twitter, students are trained to expect information in a random-access model and receive
updates automatically from various sources. Faculty members tend to plan and organize
lectures and courses in a very linear and sequential way. Many Gen-Y and Gen-Z
students have a strongly visual or auditory learning style, but find most textbooks and
lecture material to be text-based. Students are motivated and interested in practical, fun,
and relevant activities as opposed to the more standards-based approach of most courses.
Students seek instant feedback and gratification for their efforts but most course
evaluation tools provide delayed feedback. When a course is misaligned students tend
to become uninterested and fail to engage while the faculty member can become quickly
frustrated and disenchanted with the students in the course.

A TRADITIONAL INTRODUCTORY COURSE

At Plymouth State we have developed an introductory-level course that is taken by
both Computer Science and Information Technology majors during their first semester
in the program. Computing Fundamentals (CS 2010) is a three credit course which
includes a weekly two-hour lab session. This course assumes no prior knowledge of
computing or programming, and is designed to generate interest in the fields of computer
science and information technology. The primary goals of the course are to provide
students with exposure to different sub-areas of computing, to help students become
comfortable with software and technology used in the department, to facilitate
critical-thinking skills, and to prepare students for their first programming course. Upon
completion of the Computing Fundamentals course students take Programming in Java
(CS 2370), which follows the traditional introduction to programming course model.
The major topics for the traditional Computing Fundamentals course were HTML
and JavaScript, computer number systems, machine structures and assembly language,
diagramming program flow, basic algorithm structure, variables and functions, logic
diagrams, recursion, basic object-oriented programming concepts, search and sort
algorithms, and basic artificial intelligence algorithms. Each of these topics was offered
primarily at the introduction-level and provided a breadth-first model for exposing
students to a variety of common computing concepts and issues.
The topics in the course were delivered as a combination of lecture and lab
activities. Students would have a weekly lab assignment as well as homework problems
to help them master the material. During the lab sessions students had the opportunity
to experiment with several different technologies. They developed simple web pages
using HTML, wrote simple assembly programs and simulated their execution, played
with a simple robot simulator, built and programmed Lego MINDSTORM robots, and

32
 CCSC: Northeastern Conference

animated 3D worlds with Alice. The intent was for these hands-on experiences to help
students grasp the fundamental concepts of computing and algorithm design so when they
moved into the Java programming course they would be adequately prepared.
Several issues with students were observed over the course of several offerings of
Computing Fundamentals. In many cases students would become overwhelmed trying
to learn and master each language needed to complete the labs in a timely fashion.
Because each module in the course was very short in duration (only two or three weeks
long), students would become frustrated that as they would start to become comfortable
with one particular language the course would move on to the next module and they
would have to essentially start all over again with a new technology or language. This
caused some students to become very discouraged and disengaged during class activities.
Another issue that appeared was the lack of interest in some topics because of a perceived
disconnect between activities and any real-world application. Students would complain
that the lab assignments were pointless and as a result would put in little effort to
complete them. Finally many of the students who did continue on in the program seemed
unprepared for the concepts covered in the Java programming course, despite having seen
them before in the Alice programming language.

Year CS 2010 CS 2010 CS 2370 Retention

Enrolled Completed Enrolled Percent
2005-2006 38 27 9 33%
2006-2007 35 35 14 40%
2007-2008 27 26 15 58%
2008-2009 34 34 19 56%
2009-2010 32 20 15 75%
Table 1. Plymouth State University First Semester Retention Data.

Currently Computing Fundamentals is offered every fall semester and the follow-up
Programming in Java course is offered every spring semester. Table 1 summarizes
enrollment data from the previous five offerings of these two courses. On average just
over 52% of the Computer Science and Information Technology majors who successfully
complete the Fundamentals course continue on to the Java course. With the additional
attrition seen from students continuing on to their sophomore year, the low enrollments
seen at the upper-division level were of great concern. To attempt to address these
retention issues, the Computing Fundamentals has been completely re-designed for the
Fall 2010 offering of the course.

AN UPDATED INTRODUCTORY COURSE

Based on the observed problem with the misalignment of the course approach used
in Computing Fundamentals with the motivations and interests of Gen-Y and Gen-Z
students, major changes were made to the course goals and topics. Carter [1] observed

33
JCSC 26, 6 (June 2011)

that many students in introductory computer science courses have little or no idea what
computer science actually is. In addition we noticed that students were not very
committed to producing quality work when they failed to see the reason for learning the
material. To try and provide more context for the topics covered in the course, more
focus was placed on providing discussion and examples of how the technologies covered
were used in industry.
While some of the topics were retained from the previous version of the course,
many were dropped and replaced with different topics. Currently the list of topics
includes an introduction to computer science and information technology, careers in
computing, computer number systems, basic logic and circuits, simple programming
concepts, mobile computing, computer security, robotics, artificial intelligence, and
simple analysis and design. To help maintain a connection to real-world computing each
unit includes discussion of careers and opportunities for the area of discussion. Because
Gen-Y and Gen-Z students are motivated by seeing the practical application of their work
[6], several of the lab assignments were modified to provide examples of real-world
activities. Students develop a simple mobile device application using Google's App
Inventor, conduct simple experiments using several different artificial intelligence
algorithms, and play with simple security tools to better understand how computer
security is implemented in practice.
To address issues with students being unprepared for the programming class and
being overwhelmed by learning many different programming languages, the set of
languages used in the lab assignments has been streamlined. While some research
indicates that taking an objects-early approach can help students better understand
programming concepts using a visual environment with drag-and-drop commands like
Alice [2], it has been our experience that students are then poorly prepared to switch from
such an environment to a text-based programming environment. To provide the students
more experience with this model of programming, we chose to use Python for most of the
programming components of the course. This allowed for more consistency in learning
the syntax of a single language and could more easily be translated to similar constructs
in the Java course.
Finally the overall structure and tone of the course lectures was modified to better
align with the learning style of Gen-Y and Gen-Z students. Several lectures were
overhauled from a one-way lecture to more of a free-flowing discussion model. Students
were encouraged to contribute to the conversation and were posed questions frequently
as opposed to being allowed to sit passively during the entire lecture period. Additional
active learning activities were built in to engage students and more in-class group work
was added to drive discussion and help keep students interested.

RESULTS
Plymouth State's newly revised Computing Fundamentals course was offered for the
first time during the Fall 2010 semester. There were 34 students enrolled in the course,
with about a 50/50 mix between Computer Science and Information Technology majors.
Roughly 80% of the students in the course are first-year students, with the remaining 20%

34
 CCSC: Northeastern Conference

made up of sophomores and a few juniors, most who were switching into the department
or were transfer students.
The reaction from students to the new course has been overwhelmingly positive.
Compared to previous semesters, student attendance has increased significantly. The
students have been more involved and engaged in classroom discussions and several have
expressed interest in learning beyond the material covered in the course. Overall grades
on both labs and homework assignments have improved on average compared to previous
semesters. On multiple occasions students have commented on how much they have
enjoyed the course, which was almost never heard of with the previous version of the
course.
We have observed a boost in retention based on students continuing into the
Programming in Java course for the Spring 2011 semester. Of the 34 students enrolled
in Computing Fundamentals, 25 enrolled in the Java course for a retention rate of 74%.
We hope to continue to see this higher level of retention for students for future offerings
of the Computing Fundamentals course.

CONCLUSIONS AND FUTURE WORK

The Computer Science and Technology Department at Plymouth State University
was facing similar retention issues and a lack of interest in introductory course material
seen in many institutions. Much of the evidence indicated that part of the problem is
coming from a misalignment of the course model with the expectations of the newest
generation of students. By restructuring our Computing Fundamentals course to better
align with Gen-Y and Gen-Z students, we have noticed an increase in satisfaction,
involvement, and retention in our first year courses. With only one offering's worth of
data at this point it is too early to draw any conclusions about the impact on overall
retention in our degree programs, but the results observed are encouraging.
We hope to continue to track students as they progress through their programs to
gather additional data on retention rates over the longer term. In addition we are planning
to develop additional metrics for analyzing the effectiveness and impact of these changes
in the Computing Fundamentals course. Finally, we have seen similar concerns in
Programming in Java, and would like to undergo a similar process to correct any
misalignments in that course as well.

REFERENCES
[1] Carter, L., Why Students with an Apparent Aptitude for Computer Science don't
Choose to Major in Computer Science, Proceedings of the 37th SIGCSE
Technical Symposium on Computer Science Education, 27-31, 2006.
[2] Cooper, S., Dann, W., Pausch, R., Teaching Objects-First in Introductory
Computer Science, Proceedings of the 34th SIGCSE Technical Symposium on
Computer Science Education, 191-195, 2003.

35
JCSC 26, 6 (June 2011)

[3] Drexel, P., Roberson, C., Using MBTI in a Project-Based Systems Analysis and
Design Course, Proceedings of the International Conference on Engineering and
Meta-Engineering, 92-96, 2010.
[4] Guzdial, M., and Soloway, E. Teaching the Nintendo Generation to Program.
Communications of the ACM, 45(4), 17-21, 2002.
[5] Howe,N., Strauss, W., Millennials Go to College, Great Falls, VA: Lifecourse
Associates, 2007.
[6] Layman, L., Williams, L., Slaten, K., Note to Self: Make Assignments
Meaningful, Proceedings of the 38th SIGCSE Technical Symposium on Computer
Science Education, 459-463, 2007.
[7] Taylor,M.L., Generation NeXt comes to college, 2006,
http://www.taylorprograms.org/images/Gen_NeXt_article_HLC_06.pdf,
retrieved November 11, 2010.

36
 IMPROVING STUDENT PERFORMANCE WITH AN

EXTRA-CURRICULAR ROBOTICS CLUB*

Charles N. Stevenson
Computer Science Department
Salem State University
Salem, MA 01970
978-542-6616
[email protected]

ABSTRACT
We developed an extra-curricular robotics club in order to strengthen the
academic performance of our computer science students and to increase their
understanding of different career paths that are open to them. Despite the small
size or our program, around 100 students, the club has the determined
participation of six to eight students, including women and minority students.
The club's activities increased the student's practical knowledge of embedded
systems by developing low-cost line-following robots using inexpensive RC
cars and 8-bit microcontrollers. Some of the technical content created by the
club's activities will now be used in the first offering of a robotics course.

BACKGROUND
Trudy Howles describes the educational and cultural background of many of today's
computer science majors and the resulting academic problems with which the faculty
must contend [1]. These problems include an overexposure to media, poorly understood
career paths, and the distraction of part-time jobs. As a result, students are often unable
to persist in their studies to graduation. Our computer science majors certainly suffer from
these problems, and they also frequently lack an appreciation for the general usefulness
of strong analytical skills, or even the value of being a strong programmer.
Many of our majors regard a degree in computer science as a stepping stone to
employment as a network administrator rather than as a professional programmer or a

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

37
JCSC 26, 6 (June 2011)

developer of new technology. While these students express an interest in learning to use
existing applications, they are relatively uninterested in developing new content,
applications, or ideas. Many of these students have never experienced the personal
satisfaction that comes from creating new or unique ideas or applications.
Part-time, or even full-time, employment reduces both the time available and their
desire to apply themselves to course work. Many students seem to suffer from a poor
engagement with their course work because they fail to see its relevance to their later
professional careers. They often express uncertainty that the languages that they are
learning, such as Java, are actually used for software development.
Their many distractions exacerbate the academic problems that today's students face,
which undermines their chances for success. As indicated by Howles, many students
doubt the value of formal learning and the abstract nature of computer science further
discourages them from continuing with their education[1]. For these students, it is a long
way from their programming classes to developing the skills necessary to create useful
software applications, and even farther to seeing themselves as contributing to such
efforts.

EXTRA-CURRICULAR APPROACH
Our approach has been to improve our student's chances of academic and
professional success by mitigating their background problems with extra-curricular
activities designed to supplement and reinforce classroom learning. We were successful
in increasing student interest in embedded systems and robotics with our student-run
robotics club.
The students selected a project for themselves, with faculty guidance, and then
developed their own goals. We decided to build low-cost, autonomous robots that were
capable of line following for their first project. This project gave the students experience
with C programming, microcontrollers, sensor interfaces, motor drivers, and high-level
path planning, all topics found in embedded systems courses or other computer science
areas.
Our computer science program had slightly more than one-hundred majors last year,
and we had the regular participation of about six to eight students, who were typically
either junior or seniors. Some of the most dedicated members are woman or minority
students, indicating that one possible way for increasing the diversity of our enrollment
is to supplement classroom activities with extra-curricular activities that combine social
and technical activities. This result may also indicate that such students may be more
interested in computer science as a major if they learn that professionals in the field
usually work in teams that involve a good deal of social interaction. These students saw
a student club that was focused on technical work as a natural supplement to their
classroom work, a way to enhance both their credentials and their professional skills.

TECHNICAL APPROACH
Given the student's limited financial resources, it was necessary for the autonomous
robots to be constructed at minimal cost. Furthermore, while such a project emphasizes

38
 CCSC: Northeastern Conference

the interdisciplinary nature of computer science, it was essential to recognize that the
students are not mechanical or electrical designers, and lacked the necessary skills to
design a suitable platform. Consequently, inexpensive radio-controlled cars, available at
major retailers for under six dollars, were selected for the platform.
A similar approach was taken by Brauni [2], who added a custom controller to
model cars to construct autonomous vehicles for playing "robot soccer." This custom
controller is based on a 32-bit controller from Motorola. In contrast, our students decided
to develop low-cost controllers for their cars using 8-bit microcontrollers from the
Microchip Corporation, combined with a readily available quad package of half
H-bridges for controlling the drive and steering wheels. The microcontrollers, H-bridges,
and a voltage regulator were combined on an inexpensive card for under two dollars.
The use of a targeted controller, rather than a commercially developed general
controller, kept costs minimal and offered educational benefits. The students were able
to develop the controller at an architectural level, rather than simply utilizing a
pre-packaged solution with libraries that managed all the low-level tasks.
We selected an 8-bit controller from Microchip for with on-board pulse-width
modulation (PWM) module and analog to digital converters. Microchip's line of
controllers can be programmed in the C language using a cross-compiler offered by
Hi-Tech, and a reduced version of this compiler is available free to students. Microchip
offers extensive, free technical training for educators, a resource that was essential for
developing our autonomous cars because they taught us how to use the cross-compiler,
the PWM module, and the A/D modules.
In particular, we selected the PIC 16F690 from Microchip for our controller, which
is a 20 pin, flash-based 8-bit CMOS microcontroller with an internal clock, 12 10-bit A/D
channels, and a PWM module with up to four output channels. A closely related product,
the PIC 16F687, is described in detail, along with accompanying C source code, by
Bates[3].
For the line-following sensor, we used an integrated infrared emitter and detector,
the QRE1113 manufactured by Fairchild Semiconductor, which is available commercially
in sample quantities for under two dollars. The combined cost for the two sensors
required to detect the left and right edges was about twice the cost of the microcontroller
and PWM package combined. The total cost of the robot cars was around twelve dollars.
The students adapted existing, open-source C code for use in their robot cars.
Several web sites offer sample code for 8-bit PIC microcontrollers [4][5]. To utilize the
PIC microcontrollers requires setting various register flags, which beginning students
often find confusing. By studying the open-source code, students were able to see
examples which they could then adapt to their own use. From their experiences in the
club, these students also came to understand that existing solutions often cannot be simply
copied, but rather must be understood and adapted for the solution of new problems.
None of the existing open-source code could be directly used; rather it had to be modified
for our purposes. Additional C code had to be developed to interface the infrared line
detectors to the microcontroller.

39
JCSC 26, 6 (June 2011)

A portion of the C code that uses the PWM module to control two DC motors is
given in Table 1. The first task that this code accomplishes is to set the registers and flags
that are provided by the proprietary architecture. All of these registers and flags are
referenced by specific names that are peculiar to this particular product line. Without the
sample code provided by Microchip and various open-source examples, students would
have experienced great difficulty in identifying these settings.
int main(void)
{
unsigned char input;
ANSEL = 0b00000001;
ANSELH = 0b00000000;

INTCON=0b01000000; // purpose of disabling the interrupts.

PIE1 =0b00100000;
init_comms(); // set up the USART - settings defined in
usart.h
init_adc();
Stop();

/* Set up output ports */

TRISC = 0; /* all bits output */
PORTC =0;
TRISA = 0x03;

/* Set up PWM frequency and timer */

PR2 = 0b10000001;
T2CON = 0b00000111;
CCP1CON = 0b00001100;

Table 1. Code Fragment for Setting Microcontroller Registers

In order to move the cars, it is necessary for students to complete the development
of higher-level path following software. The club currently has integrated the infrared
sensors with the microcontrollers, which are able to execute higher-level instructions to
move left or right, or so forth. This software remains uncompleted and not fully tested.

IMPACT OF CLUB
Our robotics club exposed students to new career possibilities in embedded systems
and created a groundswell of interest in robotics. The club had an active membership of
six to eight students during the course of the year, out of around 100 computer science
majors.
Two of the club members are now working professionally as embedded systems
programmers or about to accept offers of such employment. In both cases, potential
employers were interested in their accomplishments as members of the club as well as

40
 CCSC: Northeastern Conference

their accomplishments as student leaders. With six to eight graduates per year, helping
to place two students in embedded systems companies is a significant accomplishment
for the club.
Our curriculum requires students to complete a two-semester senior project. For
several years, the department faculty were concerned because almost all of the senior
projects were in web design, where we were hoping that students would consider a broad
array of topics for their projects. During the last year, we have been able to get students
to work on projects from a more diverse set of topics, in part because of their experiences
in the robotics club. Three students are currently completing robotics projects, and three
others had either completed or are working on an embedded systems project using the
Microchip line of embedded controllers that they used in the robotics club. Judged solely
on its ability to expose students to different career options, the club has been an
outstanding success.

IMPACT ON ROBOTICS COURSE

We will be launching our robotics course next fall, and some material developed by
the robotics club will be incorporated into the curriculum. One of the early experiments
in the course will to be to complete the work begun by their predecessors in the club by
finishing the programming necessary for line-following cars to actually follow lines. Such
work will give students experience in sensor interfacing and writing embedded software
for closed-loop control.

CONTRIBUTIONS TO THE CURRICULUM

The knowledge gained from the robotics club led one student to develop a small
circular autonomous robot, similar in mechanical design to existing products, but based
on the Microchip line of microcontrollers. There are a number of existing inexpensive
robots available for purchase, but their use would require introducing a new product line
into our curriculum. With two out of six faculty members working in embedded systems,
there simply isn't the time necessary to develop and administer laboratory projects using
microcontrollers from several different firms. We have chosen instead to seek the
efficiencies that come from specializing in the products from one vendor.
Our own design of an autonomous vehicle uses both 8-bit and 16-bit
microcontrollers from Microchip Co. The 8-bit controllers drive the twin motors using
the C code previously developed by the robotics club for their autonomous cars. The
16-bit microcontroller interfaces to ultrasonic range sensors, communicates with a remote
controller using wi-fi, and sends commands to the 8-bit controller. The higher-level 16-bit
controller sends motion control commands to the lower-level 8-bit controller, such as to
steer left or right.
This new autonomous robot design, a direct result of the robotics club, meets the
needs of our curriculum better than existing designs. It utilizes a controller technology
that we are familiar with, and its 16-bit processor enables sophisticated laboratory
projects. The downside of this design is that students must assemble it from commercially
available parts, rather than simply purchasing an integrated unit.

41
JCSC 26, 6 (June 2011)

PERILS AND PITFALLS

In an undergraduate setting, every student club is going to require a continual influx
of new members as the student body naturally turns over due to graduation. However, our
club is limited in its ability to attract new members because it is so strongly focused on
technical activities that require students to be fairly far along in their educational
program. Students who have not completed our C programming course find it difficult
to participate in the development of autonomous robots based on microcontrollers that
are programmed in C. This is a direct consequence of the club's formation by a group of
students who wanted to improve their knowledge of C and embedded systems.
The largest problem is that the club has a much lower priority in the student's lives
than their coursework or part-time jobs. The undesirable consequence is that it has been
extremely difficult to bring even the simple project of the line-following robot to
completion. The primary incentives to completing club activities are personal satisfaction
and the hope of placing such accomplishments on a resume.

FUTURE WORK
Our new robotics course will be offered for the first time in the fall of 2011, and our
upcoming challenge will be to integrate the classroom and extra-curricular activities of
our students. While both course and club will be used to stimulate interest in the other
activity, essential development of the autonomous vehicle laboratory portion of the course
was done through the club and related activities.
The most important and difficult challenge will be to draw into the club students
who are still early in their academic careers. It was noted earlier that some of the most
determined members are woman and minority students; perhaps other students in the
program would benefit from the social and technical interactions in the club as well.
Students may be more interested in pursuing computer science as a major if they realize
that professionals in the field spent much of their time working in teams with a good deal
of social interaction.
Our extra-curricular technical club can assist students in persisting to graduation if
we keep in mind Howles results from her surveying at the Rochester Institute of
Technology. She notes that most of their students select computer science for their
college major because of a genuine interest in computing, a result that would seem to
apply to our majors as well[1]. She also cites a psychology paper that affirms the
well-known result that intrinsically motivated individuals are more likely to persist to
accomplishing their goals despite adversity than individuals whose primary motivation
is external factors. The robotics club will continue to contribute to our curriculum only
as long as it taps into and reinforces our student's fascination with computing which is
their intrinsic motivation for enrolling in the program.

REFERENCES
[1] Howles, T., Preliminary Results of a Longitudinal Study of Computer Science
Student Trends, Behaviors and Preferences. Journal of Computing Sciences in
Colleges, 22(6), 18-27, 2007.

42
 CCSC: Northeastern Conference

[2] Brauni, T., Embedded Robotics: Mobile Robot Design and Application with
Embedded Systems. Berlin, Springer-Verlag, 2008.
[3] Bates, M., Programming 8-bit PIC Microcontrollers in C with Interactive
Hardware Simulation. London, Newnes Press, 2008.
[4] http://www.micro-examples.com/public/microex-navig/doc/097-PWM-calculator
.html, retrieved November 15, 2010.
[5] http://www.best-microcontroller-projects.com/PWM-pic.html, retrieved
November 15, 2010.

43
INTRODUCTION OF VIRTUALIZATION IN THE TEACHING OF

OPERATING SYSTEMS FOR CS UNDERGRADUATE

PROGRAM*

Kevin Grammer, Jack Stolerman, and Beifang Yi

Computer Science Department
Salem State University
352 Lafayette Street, Salem, MA 01970
(978) 542-7426
kevingrammer@gmail, viper11@gmail, and [email protected]

ABSTRACT
For many colleges, virtualization is a low-cost solution for providing hands-on
lab activities for computer science courses. This paper describes the design and
implementation of a series of projects for an undergraduate operating systems
course. By utilizing Linux virtual machines on students’ personal computers,
these projects teach reinforce students’ understanding of operating system
concepts as well as teach students the basics of virtualization. After completing
the projects, students were surveyed about their experience and a summary of
their responses is presented here.

INTRODUCTION
The study of operating system (OS) principles is an integral part of the computer
science undergraduate curriculum. Beyond lectures and reading textbooks, hands-on
activities add significantly to a student’s understanding of computer science topics. The
CS2008 Review Taskforce emphasizes that the study of operating systems should include
a “laboratory component to enable students to experiment with operating systems” [1].
While some major universities are able to provide dedicated computer labs for this
purpose, smaller colleges must use what they have at hand and often cannot provide labs
for every course [4, 6]. The challenge for these colleges is to find other ways to provide
students with high-quality, affordable hands-on experience [2, 3].

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

44
 CCSC: Northeastern Conference

Virtualization offers a lightweight, low-cost, practical solution to this problem. The

student installs a virtual machine on his or her personal computer and it runs on top of the
computer’s host operating system [12]. Separation between the host OS and the virtual
machine allow the students to manipulate these virtual machines without the risk of
damaging the host OS [4, 8].
In this paper we will discuss applying virtualization in teaching an undergraduate
OS course with hands-on projects. Projects were designed to introduce students to the
Linux operating system. The main objectives of the projects were to have students update
the Linux kernel, add a system call, and work on process-and thread-related projects in
the Linux environment. We have also included the comments and feedback we received
from the students in this paper.

BACKGROUND
As personal computer speeds and storage space increase rapidly and OS
virtualization becomes more easily accessible and commonly used professionally,
colleges are more seriously considering adopting virtualization in the classroom. Some
schools have already developed virtualization techniques to bring students more
accessibility to lab-based activities for both university students and distance learning
students [7]. Not only does it cut lab costs for colleges that use virtualization as a teaching
tool, but it gives students necessary exposure to the latest real-world technologies. Most
students’ personal computers meet the minimum requirements for running at least one
virtual machine, making it a practical alternative to using university computer labs. Many
universities have already implemented virtualization to give their students hands-on
experience in system-oriented computer science courses such as operating systems,
system administration, and networks courses [4, 5, 7].
OS virtualization allows a CS professor to give his or her students exposure to
different operating systems that they may not have prior experience with. Rather than
simply learning the software of a single OS, using a variety of operating systems
emphasizes the fundamental and lasting concepts of operating systems by revealing the
commonalities in function between them [2, 10]. In addition, virtualization gives the
professor flexibility as to which OS to use to teach each concept. For example, a
professor might use Linux to help students understand the distinction between user and
super-user security levels [7].
At Salem State University, Operating System Principles is a 3-credit hour
undergraduate course. While many CS courses here have separate lecture and lab time,
this course has no accompanying in-class lab hours. From our past experience in teaching
OS, students often complained of the theoretical descriptions of OS topics and algorithms
without lab practice to help solidify an understanding of the internal structure of the OS.
Due to the limited resources in our department (students do not have administration rights
on the machine) and course setting (no in-class lab time), we must find non-traditional
solutions to overcome these challenges— to design OS projects that can be implemented
on students’ personal computers.

45
JCSC 26, 6 (June 2011)

PROJECT GOALS AND PREPARATIONS

We set the goals of designing the OS projects as: (1) introduction to virtualization
— the students will not only learn a new technology but also complete the OS projects
without affecting their personal computer environments; (2) one installation is a Linux
operating system — the students will get experience with different operating systems than
the norm (Windows/Mac) and work with free, open source software (most CS students
here have a very limited knowledge of Linux); (3) patching the kernel with a newer
version — the students will acquire direct knowledge of compilation, configuration, and
construction of Linux; and (4) adding a system call to the new kernel and practicing with
the new kernel through a simple C program — the students will have a chance to look
inside the OS to obtain a basic understanding of internal structure of one operating system
and to reinforce OS principles.
To achieve these goals, we polled the students in an initial survey. Questions from
the survey included inquiries about the students’ experience with computers, their
computer usage (such as which operating systems and programming languages they had
used) and configurations of their personal computers (such as RAM/hard-drive capacity).
From the feedback (of 25 replies at the beginning of 2010 Fall), we noticed the following:
• All the students own one or more computers: all students except one have laptops;
about 40% have both desktop and laptop.
• Only two students (8% of total) have a computer with 1 GB RAM; 20% with 2GB
RAM; 60% with 4GB RAM; 12% with 6 or 8GB RAM.
• 40% of students did not have any knowledge about Linux; 40% have very limited
experience with Linux.
• 20% of students did not use any command line environment; 40% had limited
experience through the using of Windows’ DOS.
• More than 80% of students used only Windows operating systems.
Previously, we took a similar survey in the Fall of 2009 for the OS class and
received very similar results indicating a little lower percentage in the student population
that had computers with 2GB or more of RAM.
These statistics provided us with guidelines to design OS projects. First, we decided
to use VMware Workstation as the virtualization software package for virtual machines.
Our considerations were (1) compared with other virtualization solutions, VMware is
more reliable, more widely adopted in industry, easier to set up, and provides more
management tools [9] and (2) each student’s personal computer meets the hardware
requirements needed to install VMware. We also encouraged students who had
experience with virtual machines to use alternative virtualization packages such as
VirtualBox.
Next, we made use of the concepts and topics from the course textbook [11] for the
implementation of the projects. Thus, students should be familiar with these topics from
reading the textbook. We focused on the following points in designing the projects: (1)
any student with no background in Linux should still be able to complete them, (2)
students should be given a chance to “see” the inside of Linux and to make some
modifications to it, and (3) some projects which covered the most important concepts of
the OS (such as processes and threading) should be simple enough to be implemented in

46
 CCSC: Northeastern Conference

both C and Java programming languages (Java is the first and required programming
language taught at Salem State University. Most students had little to no prior experience
coding in C when taking the OS course).

PROJECT IMPLEMENTATION AND RESULTS

Project 0: Open Source OS and Virtualization
This project begins with a guide to downloading VMWare (or another virtual
machine software) and installing the Ubuntu operating system. The Ubuntu file, although
readily available for free download on the Internet, is also provided on the university
network and distributed on CDs and thumb drives to students in class. Ubuntu is booted
within the virtualization software and from here on, students work within the Ubuntu OS
rather than the host OS.
After a basic introduction to the Ubuntu GUI and some of its preinstalled
applications (Firefox, Open Office, Gimp, etc.), the student is given instructions to use
the command line interface. The student tries some commands, such as cp, mv, and
uname to get some information about the OS. Next the student makes a directory, changes
directories, and lists parent files. Once the student is familiar with some basic commands,
he or she is asked to create a hello.cpp file and enter some lines of code (provided in the
project description). In order to compile and run the code as a C++ program, the student
must first download and install the g++ software package from the command line. As the
student completes each step of Project 0, he or she takes screenshots for submission.
The goal of Project 0 is to give the students some familiarity with the virtual
machine, Linux, and the CLI. The project focuses on how to do a variety of basic
operations with the Linux virtual machine, such as how to install a virtual machine
running Linux, how to navigate Ubuntu’s GUI, and how to get basic functionality from
the CLI. With the completion of Project 0, the student not only has a bit of experience
with Linux and the command line, but also has a “how to” reference for the next project.

Project 1: Updating the Linux Kernel and Adding a System Call

This project is a step-by-step guide to updating the Linux Kernel with a later version
and adding a system call to the kernel. Much of what the student is asked to do in Project
1 is complex and most likely beyond his or her basic understanding of operating systems
at the time when the project is assigned. Project 0 has given them some understanding of
the Linux environment and command line to better follow along.
Initially, the student opens up the CLI and logs in as the root user. From here on, the
student can write to the system¡|s directories and must be careful not to accidentally
change file or directory names. Because the student is doing this on a virtual machine, if
they do make an error, they only run the risk of corrupting the virtual machine, and
cannot directly affect their host operating system. Project 1 then takes the student through
updating the Linux package and downloading the new kernel.
In order to add a system call to the kernel, the student must act as the root user and
edit files within the kernel. Project 1 guides the student through each step of the process,

47
JCSC 26, 6 (June 2011)

editing the appropriate system files and adding a simple system call function to the
kernel. In the case of Salem State University, most OS students have very little
experience coding in C and C++. Project 0 gave students a chance to write and compile
a C/C++ program. Project 1 requires the students to manipulate C files. The student
should gain enough knowledge from Project 0 to successfully manipulate these files and
at the same time become more familiar with the structure of C files.
After adding the system call, the student must configure the kernel using the kernel
configuration menu. The student loads the new configuration and then builds the new
kernel, appending his or her first initial and last name to the kernel version string. The
compiling process takes anywhere from two to five hours. By witnessing the compiling,
students said they gained an appreciation for how massive a task compilation really is.
The student then installs the kernel and tests the newly added system call with a C test
program. Although it is only a simple C program, it re-enforces the basics of C and its
uses in the Linux kernel. Writing a working C program that makes a student-created
system call can be a very powerful tool for convincing Java-based students that at least
a basic understanding of the C programming language is important.
At the conclusion of the project, the student submits screenshots of the process to
the professor. The greatest satisfaction the student gets is seeing his or her name
appended to the kernel version name while getting a good look at the composition of the
kernel of how it functions. The student also gains valuable experience utilizing the CLI
and gains an appreciation for the root user power of the command line.

Project 2 and Project 3: Processes and Multithreading in C/Java

Project 2 builds on the basic C concepts learned from the previous two projects. In
this project, students must create multiple processes that communicate with each other
using pipes. A message is sent from a parent process to a child process. The message is
then modified and sent back to the parent process. Although a rather simple program, it
demonstrates synchronization and the implementation of pipes.
Project 3 is a more in depth look at Java threads. Besides giving students experience
using threads, this project directly contrasts Project 2's use of processes. Threads share
data in a very different way than processes. Project 3 gives students a basic understanding
of threads by designing a Java program using threads that adds and multiplies all the
numbers 1 through N. In addition to this simple threading program, the student is asked
to then write a more complex program multiplying two matrices. A separate thread is
used to multiply each element of the solution matrix. The students are then challenged to
rewrite this program in C by using PThread. A more challenging multithreading project
— modeling of one of the OS classic examples, Producer-Consumer problem — was
given as a bonus project.

CONCLUSIONS AND FUTURE WORK

Students showed great interest in these projects, particularly those involving practice
with virtual machines and Linux -- updating the Linux kernel and the addition and testing
of new system calls. In addition to everyone completing the first two projects, many

48
 CCSC: Northeastern Conference

students provided positive feedback. In order to better design for future projects, we gave
another survey to the students after they completed the above projects. We invited
students to critique their learning experiences and we asked for suggestions to help with
future modifications of the projects. The following gives a brief summary of the survey.
• Virtualization and virtual machines: more than 80% of students replied that the
projects greatly helped them understand the related subjects and use the latest
technology. About 85% of students used VMware for the projects, while others used
different virtualization software packages with success.
• Linux and open source software: more than 80% of students considered that the
projects provided a good opportunity for them to learn Linux and to make use of
open source software. More than half of them said that without the projects they
would not have even considered using Linux.
• C programming language: half of the students acknowledged that the projects
increased their knowledge of the C programming language.
• OS kernel and system call: more than 85% of students responded that the projects
helped them not only understand the OS internal structures “in a very elementary
way” but also learn the basics of system programming. Many of them recognized
that without the projects they could not fully understand the kernel functions of the
operating system.
• Additional comments: we also asked students to provide feedback on other aspects
of the projects. Many of them recognized the “complexity” of the operating system
(it takes about “3-4 hours to compile” — quoted from a survey) and the superior
functionality of the Linux command-line to the GUI. They particularly mentioned
the thread/process-related projects, which greatly reinforced their understanding of
the concepts of process and threading.
Students not only favorably responded to the survey questions but also provided
many suggestions that we will benefit from in future versions of these projects. We will
continue to use the projects but will spend a bit more time teaching the basics of Linux
before assigning the Linux related projects. We will first ask students to do some
investigation on the topic of virtualization and adopt different virtualization solutions. We
will design more projects in other areas of OS such as multithreading program but
continue to have each project build off the knowledge gained from previous projects.

REFERENCES
[1] ACM and IEEE Computer Society, Computer Science Curriculum 2008: An
Interim Revision of CS 2001,
http://www.acm.org//education/curricula/ComputerScience2008.pdf, 2008.
[2] Bergman, M., Low-cost compute clusters in virtualized lab environments,
Journal Computing Sciences in Colleges, 25 (1), 159-166, 2009.
[3] Bower, T., Experiences with virtualization technology in education, Journal
Computing Sciences in Colleges, 25 (5), 311-318, 2010.

49
JCSC 26, 6 (June 2011)

[4 Brylow, D., An experimental laboratory environment for teaching embedded

operating system, Proceedings of the 39th SIGCSE technical symposium on
Computer Science education, 192-196, 2008.
[5] Eastman, E. G. Exploring Linux as an operating system in the CS curriculum,
Journal Computing Sciences in Colleges, 21 (4), 83-89, 2006.
[6] Gaspar, A.,Langevin, S., Armintage, W. D., Rideout, M., Enabling new
pedagogies in operating systems and networking courses with state of the art
Open Source kernel and virtualization technologies, Journal Computing Sciences
in Colleges, 23 (5), 189-198, 2008.
[7] Hickman, G. D., An overview of virtual machine (VM) technology and its
implementation in I.T. student labs at Utah Valley State College, Journal
Computing Sciences in Colleges, 23 (6), 203-212, 2008.
[8] Laadan, O., Nieh, J., Operating system virtualization: practice and experience,,
Proceedings of the 3rd Annual Haifa Experimental Systems Conference (SYSTOR,
10), 2010.
[9] Li, P., Selecting and using virtualization solutions--our experiences with
VMWare and VirtualBox, Journal Computing Sciences in Colleges, 25 (3),
11-17, 2010.
[10] Shade, E., Operating Systems on a Stick, Journal Computing Sciences in
Colleges, 24 (5), 151-158, 2009.
[11] Silberschatz, A., Galvin P. B., Gagne G., Operating System Concepts, 8th ed.
John Wiley & Sons. Inc, 2009.
[12] Stackpole, B., The evolution of a virtualized baboratory environment,
Proceedings of the 2008 ACM Conference on Information Technology Education
(SIGITE¡|08), 243-247, 2008.

ACKNOWLEDGEMENT
The authors of this paper would like to thank Professor Joseph Kasprzyk,
chairperson of the Computer Science Department of Salem State University, for his
support and guidance in the writing of the paper. The corresponding author of the paper
is Beifang Yi ([email protected]).

50
 ROBOTICS ACROSS THE CURRICULUM WITH dLife*

DEMONSTRATION

Grant Braught
Department of Mathematics and Computer Science
Dickinson College, Carlisle, PA 17013
(717) 245-1401 [email protected]

dLife is a Java library designed to support teaching and research in the fields of
Artificial Intelligence, Artificial Life, Robotics and Computer Vision. A central design
principle for the creation of dLife was that it be appropriate for beginning students while
also being powerful enough for use in upper level courses and research projects. This
demonstration will illustrate this flexibility by highlighting uses of dLife from across the
curriculum.
The demo will begin with an application to be used in a First-Year seminar titled
"Can a Machine have a Mind?" that allows non-programmers to teach a robot to navigate
a maze. The structure of a simple dLife robot controller will be shown to illustrate how
dLife might be used in a CS1 course when introducing conditional statements. This part
of the demonstration will also highlight dLife's interactive features including the live
display of robot sensor and effector information. A brief discussion of how dLife can be
used when introducing other CS1 content such as variables, arrays, iteration and objects
will also be included at this point. Several examples of assignments using dLife in an
upper level artificial intelligence course will be described, and time permitting
demonstrated live or via recorded video. These assignments may include path planning,
vision based object recognition and tracking and adversarial search. The demonstration
will conclude with the execution of some research grade code that uses a genetic
algorithm to evolve a neural network controller for collision free robot navigation. A list
of other research projects in which dLife has been used will be included to further
highlight its flexibility.

ADDITIONAL INFORMATION ABOUT dLife:

dLife includes packages for neural networks, genetic algorithms, computer vision
and robot control. The neural network package includes feedforward, backpropagation,
Elman and CMAC neural networks. The genetic algorithms package provides a
generational genetic algorithm with a variety of selection mechanisms and reproduction
operators. The robotics packages support the following robots: MobileRobots Pioneer

___________________________________________

*
Copyright is held by the author/owner.

51
JCSC 26, 6 (June 2011)

3DX; K-Team Hemisson, Khepera 3 and Khepera 2; and Sony Aibo. Robot simulations
are supported through the Player/Stage system. The computer vision package provides
a collection of basic image filters: color matching; blobification; blurring; edge detection;
etc. Images for processing can come from a variety of sources: image file (jpg, gif, etc.);
sequence of image files (e.g. img1.jpg, img2.jpg, …); streaming video from the Aibo; and
streaming video from any V4L device (Linux box required). All of dLife's packages have
been designed with extensibility in mind facilitating the creation of new types of neural
networks, genetic algorithms, robots, robot devices, image filters and image sources.
The dLife ControlCenter is a GUI application that provides a convenient means of
experimenting with the robotics and computer vision features of dLife. Using the
ControlCenter, users can see the values of robot sensors, manipulate robot effectors and
observe and modify the effects of image filters without writing any code. The
ControlCenter also allows users to load a user defined Controller object (a Java class) that
can interact with robots and computer vision features programmatically. Typically the
ControlCenter is run on a workstation and commands are sent wirelessly to the robot.
dLife has been developed and fully tested on Mac OS X and Linux.

Biographical Information: B.S., Dickinson College, 1990. M.S., Pennsylvania State

University, 1995; Ph.D., Nova Southeastern University, 2005. Prof. Braught's current
research interests are in the area of artificial life and robotics. His specific interests
include interactions between learning and evolution; co-evolutionary systems;
evolutionary and developmental robotics; and biologically inspired artificial intelligence.
He is also active in computer science education. His recent projects in this area include:
the development of the dLife library to support teaching and research in the fields of
Artificial Intelligence, Artificial Life, Robotics and Computer Vision; and an evaluation
of the effects of pair programming on student programming ability in the first computer
science course.

Supporting Materials: The dLife documentation, examples, binary files and source files
will be made available via http://users.dickinson.edu/~braught/dlife. In addition a single
page flyer highlighting dLife's major features will be available for distribution.

52
 JAVA VISUALIZATION OF STUDENT-IMPLEMENTED OS

ALGORITHMS*

DEMONSTRATION

Adam Fischbach
Department of Computer Science
Widener University
Chester, PA 19013
(610) 499-4035
[email protected]

In an operating systems course, students have the opportunity to learn a variety of

different algorithms designed to solve problems such as process scheduling and page
replacement. Students can learn these algorithms on a conceptual level, but they gain a
stronger understanding of the algorithms by implementing them and seeing them work.
For students who lack a strong background in C, studying and modifying the code of an
actual operating system such as Linux may not be feasible in an undergraduate operating
systems course. Furthermore, with the variety of topics than can be covered in such a
course, it may not be desirable to assign major month-long programming projects. As an
alternative, a simple simulation has been developed for the undergraduate operating
systems course at Widener University that allows students to implement and test their
algorithms in Java, the language that they are most familiar with.
The students are given the source code of the simulation, which implements a simple
algorithm. They are then asked to modify the code to implement a more complex
algorithm. For example, students are given an implementation of the First-Come,
First-Served scheduling algorithm and asked to implement a multi-level feedback queue.
Students may be given an implementation of the First-In, First-Out page replacement
algorithm and asked to implement a 2-bit clock algorithm or Least Recently Used. By
running the simulation, they are able to see the results in a simple GUI window and verify
that their implementation is correct. Students are given one to two weeks to complete the
assignment, depending on the difficulty of the algorithms they are asked to implement.
The simulation is not complex and only simulates the algorithm implemented by the
student. These assignments have been used effectively for the past three years. Students
have successfully completed the assignments in the given time frame and have
demonstrated a greater understanding of the algorithms.

___________________________________________

*
Copyright is held by the author/owner.

53
JCSC 26, 6 (June 2011)

During the demo, recent assignments given to undergraduate operating systems

students will be presented and sample implementations of scheduling and
page-replacement algorithms will be demonstrated.

BIOGRAPHY
Adam Fischbach is an Assistant Professor of Computer Science at Widener
University. He has a Ph.D. in Computer Science and Engineering from Penn State
University and has taught an undergraduate operating systems course each fall for the
past five years.

54
 iJava - AN ONLINE INTERACTIVE TEXTBOOK FOR

ELEMENTARY JAVA INSTRUCTION*

DEMONSTRATION

Robert Moll
Department of Computer Science
University of Massachusetts
Amherst MA 01003
[email protected]

iJava is an online interactive textbook for elementary Java instruction. Its principal
feature is an embedded evaluator, which means that throughout the text students are asked
to solve simple coding problems that reflect textbook material. The text is integrated with
a remote-server-based automated homework system and learning management system -
the OWL system.
The iJava text is online, and is organized around a set of ~ 175 embedded questions.
These are mostly very simple coding problems. They are tied to the surrounding text. The
work cycle goes this way: a student reads a few paragraphs, and then works one or
several exercises that are tied to the just-discussed material. Since problems are easy,
answers come quickly, students are patted on the back, and - most importantly - material
leading up to the embedded problems has been read. The embedded problems have a
due-date, and to ensure that students keep up, students get triple extra-credit for timely
completion. Kids in computer classes are maniacal about "points", and by and large they
can't resist a triple credit offer.
iJava runs from a remote server (right now at UMass) and on that server records are
kept of every action a student takes, for example how many times a particular problem
has been attempted, when attempts were made, and what those attempts are.
The embedded question component is the crucial and unique feature of iJava - no
other book, to my knowledge, works in this way. The four figures below illustrate how
these problems work. However the book also relies on a number of other interactive
features that make the book more lively - short movies, mouse-over interactions, and so
forth.

___________________________________________

*
Copyright is held by the author/owner.

55
JCSC 26, 6 (June 2011)

In addition the book is supported by a corpus of 200+ automatically graded

homework problems. These regular (and generally more difficult) problems are divided
into weekly problem sets that are tied to chapters in the book. Results here too are
recorded on the remote server.
While the class is demanding, it has greatly improved retention rates for our
introductory Java class. Only 8-20% of students do not pass my final exam.
Finally, the class has done a good job of: 1) making student more effective readers
of technical materials (most students do indeed read the book); 2) making students
independent learners - less than half the students ever come to lecture, since the book is
sufficient for most students to learn Java on their own, and on a final survey, more than
55% of students said the course did indeed make them better independent learners; and

56
 CCSC: Northeastern Conference

3) arousing and keeping students interested in programming - though computer science

majors make up just 20% of my beginning class this term, roughly 50% of the class has
signed up for our next CS class, Data Structures.
Robert Moll is Associate Professor and Associate Chair for Academic Programs in
the UMass Computer Science Department. He is the author of five CS texts and two
children's books, He holds a PhD in Mathematics from MIT (1973) and has had a
long-standing interest in CS teaching and instruction.
The first three chapters of the textbook can be viewed at
http://ijava.cs.umass.edu/

57
 USING THE GOOGLE MAPS API WITH HIGHWAY MAPPING

DATA AS A PEDAGOGICAL TOOL*

DEMONSTRATION

James D. Teresco
Visiting Assistant Professor of Computer Science
Siena College
Loudonville, NY 12211
[email protected]

Nearly any topic in a data structures or algorithms course will be more interesting
to undergraduate students if they can apply what they are learning using real-world data
and visualize results in a meaningful way. This demo will show how to use highway data
obtained from the Clinched Highway Mapping (CHM) Project [3] to explore graph
structures and algorithms and to display results using the Google Maps API [1].
The CHM Project has gathered information about the routes of highways in North
America and Europe. The project allows travelers to track the segments of roads they
have traveled (or, in the project's terminology, "clinched"), to see representations of their
travels in map form, and to compare the extent of their travels with those of other project
users. The data that describes the highways needs to be at a fine enough granularity to
allow users to specify their travels with a reasonable level of accuracy, but coarse enough
to be able to present (in both list and map form) the endpoints of road segments available
to be marked as clinched. Each route is added to the project ("plotted") by listing the
latitude/longitude coordinates of the route's endpoints and at major intersections along
the route (the "waypoints" of the route). Project volunteers plot the highways at an
appropriate level of granularity using public domain data sources such as route logs from
state departments of transportation, map data from the OpenStreetMap project [2], and
government satellite images.
The CHM data for a highway system is used to generate a graph that represents that
highway. A preprocessing program developed by the presenter reads the set of waypoints
for each route and matches up shared intersections based on coordinates. The waypoints
(of all routes in a system) are the graph vertices and the road segments connecting
adjacent waypoints are the graph edges. Each vertex is labeled with a string based on its
route name(s) and waypoint name and stores its coordinates; edges are labeled with the

___________________________________________

*
Copyright is held by the author/owner.

58
 CCSC: Northeastern Conference

name of the route that connects its endpoints. Distances (edge lengths or weights) are
computed easily from the coordinates
The highway systems available in the CHM project vary in size, meaning graph data
can be generated at a variety of scales. For example, the Interstate highways in Hawaii
generate a graph with only 47 vertices and 48 edges, while the New York Interstate, U.S.,
and state highway systems combine to form a graph with 7265 vertices and 8416 edges.
These graph data files can be used as input for class examples or assignments in a
data structures or algorithms course. This data was used in a laboratory assignment for
a data structures course at Mount Holyoke College where students were required to build
a graph structure representing a highway system then perform Dijkstra's algorithm to
compute the shortest route between two specified waypoints. Student programs generated
output files listing the road segments along the shortest path. These were then uploaded
to a course web server, where an instructor-provided program used the Google Maps API
to visualize their results. Students were able to develop and debug their programs using
small data sets like the Hawaii Interstates, then use those programs to compute much
more complex routes using the larger data sets.
The software has been updated and extended for use during the Spring 2011
semester at Siena College in an algorithms course. Many more highway systems have
been added to the CHM project and those will be made available to students.

DEMONSTRATION PLAN
The "demo" presentation will begin with a brief overview of the CHM project and
of a selection of graph algorithms that are appropriate for use with the CHM data. It will
then describe the CHM data formats, the desired graph data format, and the process used
to convert CHM data into these graphs. Next, it will focus on the Javascript program that
plots the graph data either in its original form or as output of graph algorithms such as
Dijkstra's algorithm. The presentation will then demonstrate examples and assignments
that use the system, from both the instructor and student perspectives.

PRESENTER BIO
James D. Teresco earned his Ph.D. in Computer Science from Rensselaer
Polytechnic Institute in 2000. He has held faculty positions at Williams College and
Mount Holyoke College and is currently a member of the faculty in Computer Science
at Siena College.

SUPPORTING MATERIALS
At the time of this writing, available web resources include the lab assignment page
(http://courses.teresco.org/cs211_f09/labs/maps/) from the first assignment to use this
project, and the mapping site used by the students to visualize their results from that
assignment (http://courses.teresco.org/cs211_maps/). A new site is under development
that will 1) more thoroughly describe the project from the instructor's point of view and
2) provide additional tools to be used by students in the Spring 2011 algorithms class to

59
JCSC 26, 6 (June 2011)

visualize their own results in the Google Maps API. The new site will be available before
the demo presentation and its URL will be given at the demo. It will include all the
assignments and examples that make use of this project. The software that converts CHM
data into graph format and the Javascript programs that plot the data in the Google Maps
API will be made available on the instructor portion of this site. They are currently
available from the presenter upon request.

REFERENCES
[1] Google, Inc., Google Maps API Reference, 2010,
http://code.google.com/apis/maps/, retrieved December 10, 2010.
[2] OpenStreetMap: the free wiki world map, 2010, http://www.openstreetmap.org/,
retrieved December 10, 2010.
[3] Reichard, T., The Clinched Highway Mapping Project, 2010,
http://cmap.m-plex.com/, retrieved December 10, 2010.

60
 THAT'S NEAT -- HOW DO I DO IT?*

DEMONSTRATION

Alice E. Fischer
University of New Haven
[email protected]
80 Killdeer Road, Hamden CT 06517

This presentation is about teaching techniques for a course that introduces GUI
interfaces and event-driven programming. The material was developed for a
sophomore-level Java course but could be adapted to other languages or other levels.
As student programmers advance, the programs they study and those they must write
become more complex. At this level, it becomes both impossible and undesirable to
develop an entire program in class. Impossible because programs span two or three files
and become too long for one class period. Undesirable because all event-driven programs
have large portions that are just like all the others, and therefore not appropriate to cover
repeatedly in lecture. The instructor, then, needs to show how each new concept or
technique fits into a larger, relatively routine, context.
Even simple programs in the areas of GUI programming, event-driven
programming, and multimedia require use of two or more Java classes. Further, elements
relating to one new concept can be distributed in different parts of those classes.
Presenting the required elements could be both boring and confusing. I use a demo-first
method to unify the material while, simultaneously, providing a way for students to
engage with the new topics. This approach is especially helpful when used to make
abstractions visible.

PRESENTATION METHOD:
• Start each class with a colorful demo that illustrates the day's topic. Then ask, "Do
you know how to do this?", "What's new here?". Relate what they see to what you
want to teach.
• Briefly introduce a new feature illustrated by the demo and show the portion of the
demo code that implements the new feature. Stress usage and semantics, not just
syntax.
• Elicit questions and suggestions for changes. Compile and run the results.
• Summarize the new material, adding anything the discussion has missed.

___________________________________________

*
Copyright is held by the author/owner.

61
JCSC 26, 6 (June 2011)

When a demo introduces more than one distinct concept (e.g. layouts and button
clicks), they should be presented in sequence.
Two examples (boldface) from the following list will be briefly presented. All have
been used in my class and are available on my website at
http://eliza.newhaven.edu/ccsc11/
1. Moving: concurrency, event-driven, interactive programming, fonts and screen
coordinates. (Visualize screen coordinates and events)
2. Caterpillar: colors, timers, multiple windows, awt drawing primitives.
3. Balloon: colors, shapes, scale, buttons
4. Lantern: radio buttons, Model/View/Controller concepts.
5. Tree: layouts, check-boxes. Responding to checkbox events.
6. Egg: mouse, window, and keyboard events. (Visualize events)
7. Pumpkin: class derivation and polymorphism. (Visualize derivation)
8. Tic-Tac-Toe: multimedia
9. Cluck vs. Hiss: threads, an interactive game

ASSESSMENT OF THIS TEACHING METHOD:

The simple but fun examples help minimize lecture time and maximize learning by
questioning. The approach has been well received by students and leads to minimal or
better mastery for almost all of them. This year, the final exam included a practical
portion (2 hours) during which students were asked to create a GUI with controls + an
animated color graphic scene. All students succeeded in doing either the color graphic
portion or the GUI portion. Half of them completed both parts.
Students like this mode of presentation. They are enthusiastic about the material
being presented and make positive comments on course-end surveys. A few examples:
• I loved learning Java.
• The most enjoyable thing is writing programs you can see working.
• The computer code examples are helpful.
• I got a good understanding of object-oriented design.

SUMMARY
We are using a demo-first method, with color-graphic demos:
• To motivate students to engage with the material.
• To help then visualize concepts.
• To make the class time fun (at least a little bit).
• To focus attention on one or two sets of concepts per class period, even when those
concepts are embedded in a complex program.
• To provide a complete program that can be modified in class, to answer student
questions and experiment with variations.
• To develop each student's appreciation for what they can do with Java.

62
 CCSC: Northeastern Conference

HANDOUTS FOR THE TALK:

• Summary of the goals and the method.
• For the demos presented in the talk, a page, each, listing the teaching objectives and
showing two or three views during execution.
• A URL and index to the website supporting the talk, listing all the demos and their
purposes.

BIOGRAPHY:
Alice E. Fischer
Professor of Computer Science, University of New Haven, West Haven, CT
Areas of interest: programming languages, software design, systems programming.
BA in Mathematics, 1964, University of Michigan,
MS in Applied Mathematics, Harvard University, 1967
Ph.D. in Computer Science, Harvard University, 1985

63
 EXPERIENCES WITH COMMUNITY-BASED PROJECTS FOR

COMPUTING MAJORS*

Jeffrey A. Stone Elinor Madigan

Pennsylvania State University Pennsylvania State University
200 University Drive 200 University Drive
Schuylkill Haven, PA 17972 Schuylkill Haven, PA 17972
570-385-6267 570-385-6267
[email protected] [email protected]

ABSTRACT
Projects which compel students to work in team situations with community
partners are one method for building so-called "soft skills" in computing
majors. The use of team-oriented, community-based service learning projects
can provide meaningful and productive experiences for students but must be
selected and managed carefully by faculty. This paper discusses a sample of
recent community-based projects at one campus of a large, multi-campus
research university. Experiences with community-based projects involving
Web design, systems design and implementation, and digital storytelling are
discussed, as well as the benefits and challenges encountered in the projects.

INTRODUCTION
There has been a growing recognition in recent years that so-called "soft skills" -
teamwork, oral and written communication, and social skills - are becoming increasingly
important for graduates in the Computing Sciences. Projects which compel students to
work in team environments with community-based partners are one method for building
these skills. The use of team-oriented, community-based projects throughout a four-year
degree program may also be a powerful tool for retaining majors as well as providing
students with a competitive advantage for future job searches. These service learning
projects can provide meaningful and productive experiences for students but must be
selected and managed carefully by faculty. This paper will discuss a sample of recent

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

64
 CCSC: Northeastern Conference

community-based projects at one campus of a large, multi-campus research university.

The benefits and challenges encountered in these projects will also be discussed.

BACKGROUND
Community-based projects are undertaken for many reasons. For example, these
types of projects foster positive relationships between the university and the surrounding
community, provide a tangible means for students to relate their studies to the "real
world", and allow faculty members to integrate students into community-based public
scholarship projects [2]. "Service Learning" has become an institutionalized part of
university life at some colleges and universities, offering opportunities for students to
serve local communities while providing meaningful learning experiences [1]. Computer
Science Education has recognized the value that community-based projects/service
learning can offer to computing majors ([3], [6], [7]). Prior research into service learning
and community-based projects for Computer Science students has focused on experiences
with "traditional" computing projects such as systems analysis, design and development
[6], database development [5], and software engineering [3]. However, Sanderson [4] has
argued that further work is needed to offer guidance on how to effectively integrate
service learning into the CS curriculum.
This paper will describe several community-based projects undertaken by computing
majors at a single campus of Penn State University. These projects were undertaken in
a rural area of Pennsylvania where the predominant local employers involve educational
services, health care, and social assistance. The students were enrolled in Penn State's
Information Sciences and Technology (IST) program, which combines elements of
Computer Science and MIS. The IST program is active learning-oriented and stresses
teamwork and oral and written communication skills in addition to technology skills; as
a result, projects involving community-based partners are common across the curriculum.
This paper describes a sampling of these "real world" projects, including a discussion of
the benefits and challenges of these projects.

COMMUNITY-BASED PROJECTS WITH DIGITAL STORYTELLING

One course in the IST program where community-based projects have been
integrated is entitled "Information and Organizations" (IST 301). The purpose of this
course is to introduce IST students to the more basic elements of business and
organizational behavior through the "lens" of information technology use and integration.
This junior-level course is unlike any major-specific course IST students will have
encountered. Prior to this course students will have completed more traditional
technology-oriented courses in C++ and Java programming, databases, networks, calculus
and discrete mathematics. Students typically take this course in the same semester as a
complementary course in Project Management. These courses are meant to prepare
students for future coursework in systems analysis, design, and integration.
Beginning in the fall of 2009, IST faculty at the Penn State Schuylkill campus
partnered with a local community revitalization organization (Schuylkill VISION) to
build digital storytelling projects into the IST 301 course. The goals of these projects

65
JCSC 26, 6 (June 2011)

were to provide our students with a "non-traditional" set of technology projects and to
provide a means by which students could collaborate with community-based partners and
exercise their creativity. These projects were also tied to a campus teaching theme for
2009-2011, "Diversity within Community", which strived to showcase the diversity
inherent within the campus community as well as the surrounding communities. The
following sections describe two main types of projects which resulted from this endeavor.
One method of integrating community-based digital storytelling into the IST 301
course focused on event-driven stories. At the request of Schuylkill VISION, IST 301
students were assigned a series of local community events to film for fall 2009. Each
team of three students constructed a 3-5 minute digital story for each of four individual
community events. Students attended and filmed the events in addition to interviewing
important stakeholders. These events included such things as town craft festivals,
Halloween parades, and charity fundraising events. The events were chosen for their
diversity, both in terms of the host community and the event's content. The result was a
series of short, Web-based videos which could be used by Schuylkill VISION to highlight
the diversity of the local community.

Digital Stories as Oral Histories

During the fall 2010 semester another group of five IST 301 students embarked on
a larger digital storytelling project than had thus far been attempted. The project, done in
partnership with two local non-profits, involved constructing a digital history of the local
shoe manufacturing industry. In the 20th Century, Schuylkill County, Pennsylvania had
one of the largest shoe manufacturing industries in the United States. The industry has
since declined, but one of the former shoe manufacturing facilities was recently turned
into a local Arts Center. The Arts Center management asked our students to build a set
of short digital stories around an oral history of Schuylkill County's shoe manufacturing
industry. The purpose of these videos was to educate the public about this rich heritage
through archival materials and first-person oral histories of former shoe industry workers.
The shoe manufacturing industry project provided students with an opportunity to
learn about many different academic and practical research methods, including historical
research. Since the industry has long since declined, much of the historical information
necessary to tell these stories could not be found via a simple Google Search. Visits were
made to local historical agencies, campus archives were searched, and digital library
databases were scoured for information. In addition, students were given the opportunity
to interview former workers from the local industry as well as individuals whose families
had a long history of involvement with the industry. These interviews - conducted both
individually and in a focus group - formed the basis for a series of short stories on the
Schuylkill County shoe industry. One video involved a summary of the industry, while
others were shorter, more focused stories using interviews with former shoe workers.

Challenges and Benefits

The challenges involved with these digital storytelling projects were significant. The
primary challenge was training and equipment. The students were by no means
professional videographers. The class involved a non-trivial amount of training in the

66
 CCSC: Northeastern Conference

technical use of the video equipment (digital video cameras, microphones, iMovie) as
well as the more stylistic aspects of video production (proper lighting, ambient sound
management, shot framing). Thankfully, a local professional videographer provided both
tutorial materials and in-class advice for the students. The limited amount of equipment
available to the students was also a challenge: only a few cameras and microphones were
available to students through the campus library on a first-come, first-serve basis.
The experience was a change of pace for students. Unlike the traditional projects
they experience in other courses (e.g. software development, systems analysis) these
digital storytelling projects provided students with a window into their community, its
people, and its events. Teamwork, communication, and client interaction were critical
project elements. These projects also allowed the more creative-oriented students to flex
that creativity using multimedia and "new" technology - e.g. iMacs, iMovie, and other
multimedia software. As word spread about the Shoe project in the wider community,
other parties (former shoe industry workers or their relatives) contacted the instructor
offering their services as sources. Consequently, the Shoe project is ongoing.
More than anything however, these digital storytelling projects provided a vehicle
for students to see how multimedia technology can be used to impact their community.
The digital story videos involving event synopses allowed various non-profit and
community groups to spread information about their activities to a wider audience than
otherwise may have been available. The digital history of the Shoe Industry provided a
means by which the local area can learn about and share a rich manufacturing history.
Students were thus able to realize the ways in which computing and multimedia
technology can be used to educate the community and achieve a wider social impact.

COMMUNITY-BASED PROJECTS WITH WEB AND SYSTEMS DESIGN

Community based projects help to connect student's academic work to the
community needs. The IST program at the Penn State Schuylkill campus partners with
a variety of local small businesses and non-profit organizations as well as campus units.
One of the more common requests from these organizations is to provide some level of
Web design, integration, and e-commerce services. For these projects, one key challenge
is the balance between what the community partner needs and the skill set of the students.
In our Web Design course, the majority of the students have no background in Web-based
languages. Most students have never created a Web page nor do they have any design
skills. Our community partners need to have patience and must be willing to work with
the students for both sides to have a positive experience.
Through careful management of the student-client relationship by faculty - as well
as positive results from the student teams - community-based Web design projects have
been a successful tool for engaging and retaining students. For example, one team of
students designed a small site for a local farmer ("The Veggie Stand"). The team not only
did the basic site design, but they also researched and recommended a Web hosting
service. Once the site was complete the team uploaded it and showed the farmer how to
maintain it. The site is still active and the farmer is still pleased with the efforts of these
students. Small efforts such as these have enabled larger projects to be undertaken, most
often with successful results for the client and positive learning outcomes for students.

67
JCSC 26, 6 (June 2011)

Emergency Management Planning System

Students in the IST program have a senior-level, writing-intensive capstone in
systems design and integration. This capstone course requires students to design and
implement a usable IT system for external clients. One of the most successful of these
projects involved a partnership with the Schuylkill County Emergency Management
Agency (SCEMA). The project goal was to produce an online Emergency Operation Plan
database and report generator. These plans are mandated by the Federal Government and
each Pennsylvania municipality must create and maintain this document. Each
municipality in Schuylkill County was expected to have access to the new system. The
system was to be used to allow municipalities to enter all required data for shelters, dams,
bridges, essential personnel and critical infrastructure. The system would then take that
data and dynamically create the state- and federal-required document.
The SCEMA project began with a two-student team but, due to unforeseen
circumstances with one of the students, ended with one student doing the bulk of the
project with faculty support. The resultant system was completed on time and worked as
originally intended. SCEMA then asked the student and the instructor to demonstrate it
at the annual Pennsylvania Emergency Management Agency meeting that fall, where the
system was well received. Unfortunately, it became obvious that some technical problems
with the system and errors in the initial requirements from the client would necessitate
some further project work. Despite this disappointment, this continued work will be used
as an additional course project in the near future.

Small Projects Can Bloom Into Larger Student Opportunities

Not only can community partners provide classroom projects for faculty, they often
provide internships for students who work on these projects. For example, a local
cemetery was looking for a Web site and requested some help from the IST students. As
faculty talked with the cemetery about their requirements, it became clear that what the
cemetery envisioned was larger than a simple class project. As a result, the project was
transformed into an internship experience for one student. The student assigned to the
project quickly became attached to his work and has continued working with the client
well beyond the time of his internship. The student has created a number of information
products for the client's Web site - video content, a slideshow, downloadable documents,
and an online donation form. The student will soon have spent two academic years
working on this project, with one of those years as a volunteer.
Another successful project that started out small was a partnership with the local
county Cooperative Extension Office. The Horticulture Educator asked the IST program
to help put a map of local farms online. The Web Design students assigned to this project
decided to create an "experience" for Web site visitors. Students researched different
methods of providing views of local farms and their produce. The students ultimately
decided upon a Google Maps solution. The students' Web site contains a map with
pushpins for the different local area farms. When a pushpin is selected by the user a
popup provides more information about the farm and a link to the farm Web site. Should
the farm not have a Web site, a simple page containing database-driven information about
the farm appears. The students also created a user manual (using Google Docs) as well

68
 CCSC: Northeastern Conference

as administrative Web pages which allow the Horticulture Educator to maintain the site.
This successful project has also lead to a related project about the trees on campus.

COMMON CHALLENGES IN COMMUNITY-BASED PROJECTS

There are several challenges involved with utilizing community-based projects for
computing majors. Finding projects that both fulfill an external organization's needs as
well as providing a manageable and productive learning opportunity for students can be
challenging. Undergraduate students are not yet professionals, and depending on their
academic progress may have limited skills for a given project. A certain amount of
student training may be necessary to prepare students for "real-world" expectations,
standards, and client interaction behaviors. Faculty members must therefore be careful
to manage the expectations of community-based partners, choose projects that align with
course goals and student skill levels, and provide preparatory training for students.
Faculty should decide on the acceptance criteria for proposed projects in advance.
These criteria will help to "weed out" unsatisfactory proposals from local organizations.
These criteria may include pedagogical goals as well as the nature and past history of the
client. Exceptions can and should be made; for example, we will take on a lesser project
for a non-profit that is considered to be making a difference in the local community. The
ability of faculty to get the word out to the local community is also critical. One method
is for faculty to actively participate on the boards of community organizations. This
visibility heightens the profile of the program and can lead to quality projects. Faculty
should also ensure that their academic administrators and public relations staff are aware
of their community-based activities.
Faculty should be aware of the significant time commitment involved in managing
these projects. Communication with student teams is critical: status reports and managed
deliverables are two methods by which faculty can keep communication channels open.
Perhaps the biggest communication challenge involves community partners. Local
organizations do not always realize how time consuming it can be to work with students,
many of whom may be learning the technology needed for the project. Sometimes
community partners fail to be communicative with student teams - e.g. they will not
answer emails, return phone calls, or do not provide the necessary information/materials
to complete the project. Faculty must sometimes intervene to get the project to move
forward and should have a backup plan if the client abandons their responsibility. Of
course, there will also be student teams who have grandiose ideas or simply do not follow
through, and faculty must be prepared to deal with those situations as well.

CONCLUSION
Community-based projects partnerships can be a win-win for computing programs,
students, and community partners. Students are able to use classroom skills while helping
their community, computing faculty and programs can establish important community
relationships, and community partners gain the benefit of technology services at little or
no cost. However, the experiences described in this paper demonstrate that careful faculty
management of these projects is required.

69
JCSC 26, 6 (June 2011)

REFERENCES
[1] Maddux,H., Bradley, B., Fuller, D., Darnell, C. & Wright, B. Active Learning,
Action Research: A Case Study in Community Engagement, Service-Learning,
and Technology Integration. Journal of Higher Education Outreach and
Engagement, 11, (3), 65-79, 2006.
[2] O'Meara, K. Motivation for Faculty Community Engagement: Learning from
Exemplars. Journal of Higher Education Outreach and Engagement, 12, (1),
7-29, 2008.
[3] Olsen, A. A service learning project for a software engineering course. J.
Comput. Small Coll. 24, (2), 130-136, 2008.
[4] Sanderson, P. Where's (the) computer science in service-learning?. J. Comput.
Small Coll. 19, (1), 83-89, 2003.
[5] Scorce, R. Perspectives concerning the utilization of service learning projects for
a computer science course. J. Comput. Small Coll. 25, (3), 75-81, 2010.
[6] Tan, J. & Phillips, J. Incorporating service learning into computer science
courses. J. Comput. Small Coll. 20, (4), 57-62, 2005.
[7] Traynor, C. & McKenna, M. Service learning models connecting computer
science to the community. SIGCSE Bull. 35, (4), 43-46, 2003.

70
 CHANGING PERCEPTIONS OF COMPUTER SCIENCE AND

COMPUTATIONAL THINKING AMONG HIGH SCHOOL

TEACHERS*

Patricia Morreale David Joiner

Department of Computer Science NJ Center for Science, Technology,
Kean University and Mathematics
Union, NJ 07083 Kean University
[email protected] Union, NJ 07083
[email protected]

ABSTRACT
Interest in specific college majors begins in high school. The influence of high
school math and science teachers is important, both for high school preparation
of future computer science undergraduates, and for the understanding and
interest in prospective majors high school teachers can encourage in their
students. With the simultaneous decline in high school computer science
curriculum offerings and undergraduate majors in computer science, an effort
has been made to update the perception of computer science and computational
thinking among high school teachers by hosting local non-residential summer
workshops for high school teachers and their students.
Pre- and post- workshop surveys show that the summer workshops have
changed the perceptions high school teachers had regarding careers in
computer science, resulting in a 50% increase in the likelihood that a high
school teacher would recommend computer science, computational science,
or information technology to their students as a career. By changing teacher
perceptions of computer science and computational thinking, high school
teachers are more likely to include computer science as one of the college
majors suggested to their students.

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

71
JCSC 26, 6 (June 2011)

INTRODUCTION
Exposure to computer science and computational thinking is frequently absent from
the high school experience in the United States, due other curriculum expectations,
budget constraints, or teacher experience. One theory is that today's classroom teachers
are not aware of the materials available to them for use with students and the professional
opportunities available to their students upon graduation. While using computers may
be part of the learning experience for students, problem solving and an understanding of
what computer scientists, computational scientists, and information technologists do after
college is not universally held among the teaching population in high schools [1, 2].
The importance of computer science to the national curriculum has been addressed
by some states with teacher certification programs [3]. The larger topic of K-12 teacher
certification in computer science is not considered here. Rather, an experience report on
university faculty interaction with high school teachers, focused on changing teacher
perceptions of university study in computing and computer science, is provided. By
holding summer workshops for high school teachers and their students, a public
comprehensive university has updated and enhanced ideas current high school faculty
have regarding applications of computers in the sciences and the utility of computational
thinking in high school. Pre- and post-workshop surveys of teachers and students have
shown that math and science teachers are more likely to recommend computer science,
computational science, and information technology to their students as a career choice
after having attended a summer workshop.

BACKGROUND
The decline in computer science (CS) majors in the United States is known [4, 5].
An additional challenge these days is the removal of computer science and information
technology courses from high school curriculum offerings in states where a CS/IT course
is not required for high school graduation. Often, CS and IT high school offerings have
been provided as mathematics or business technology electives, and with the nationwide
decline in federal and state funding for local public school systems, CS and IT electives
have been reduced or, in many cases, completely removed from high school curriculum
offerings, further limiting opportunities for high school students to be exposed to
computer science and computational thinking prior to college. Over the years, a number
of university outreach efforts have been used to attract students to the major, with little
success. University faculty are challenged when it comes to keeping high school student
awareness of computer science high.
An alternative paradigm was considered: what if the emphasis of university and
college faculty moved from engaging high school students to engaging their teachers?
The importance of high school teachers in recruiting undergraduate computer science
students has been recognized by others, and well-regarded multi-day regional residential
workshops for teachers only, such as those held by the CS4HS group over the past five
years [6], demonstrate the utility of teacher workshops in enhancing teacher effectiveness
in high school computer science education. The week-long non-residential workshop
presented here is more inclusive, as math, science, and computer science teachers from
high schools are welcome, not just computer science teachers. Additionally, the

72
 CCSC: Northeastern Conference

workshop outlined here includes teachers who would otherwise not been able to travel
away from home for a week of professional development to attend, but are able to attend
a regional event which they can commute to daily. A unique feature of the summer
workshop presented here is that teachers are encouraged to bring up to four students from
their high school. This gives teachers attending the workshop an opportunity to use
materials provided with their students, resulting in immediate feedback on the utility of
the materials offered and student experiences specific to the teacher's actual environment.
As a result of this thinking, a week-long university-based summer program for high
school math and science teachers and their students was developed. Teachers were
offered continuing education units (CEUs) if they attended and a modest stipend for their
summer participation.

RECRUITMENT AND ACTIVITIES

The goal of the summer workshop was to address the educational ramp from high
school to university - the ramp which is the approach for students considering computer
science or any discipline involving computational thinking. In light of NSF's "10,000
school project" [7, 8] which seeks to develop a new pre-AP secondary school curriculum
in fundamental concepts of computing and computational thinking, with the goal of
having this new curriculum taught in 10,000 schools, a professional development pipeline
to current math, science, and computer science teachers is vital.

Recruiting Summer Workshop Participants

With the professional development model in mind, summer workshop notices were
circulated to potential workshop participants, with a note that teachers could bring teams
of up to 4 students with them for the week-long workshop. The workshop notice was
directed at high school math, science, and computer science teachers and the workshop
participants had primary responsibilities teaching mathematics, physics, chemistry or
biology. Secondary school budget cuts have removed many computer science teachers
from high school programs in the region.
In addition to the notices being sent to students, a mailing list of SAT students in the
region was also used, with the request to the students that they consider forming a team
of four students, and bring a teacher to the summer workshop. The student appeal was
very effective.

Summer Workshop Activities

The summer workshop was scheduled in the weeks immediately following the
conclusion of the academic school year. The structure of the summer workshop included
a group event in the morning, such as a demonstration or illustration of a scientific
concept, followed by a lunch break, with individual or small team activities in the
afternoon, reinforcing concepts presented in the morning.

73
JCSC 26, 6 (June 2011)

Table 1. Summer Workshop Activities

Software Tool Activity
VenSim Easy Java Simulations
Predator-Prey
Easy JavaSim Falling Rock
Bouncing Ball
Earth-Sun Model
N-Body Model
Harmonic Oscillation
Supercomputer Environment Running N-Body Simulations
POV-Ray Creating Animations
Stereo Project - 3D
The primary material for the summer workshop was modeled after materials
developed and successfully used at Shodor [9]. In addition, a tour of the campus
supercomputer, and individual projects are included. Table 1 lists the Summer Workshop
activities.
The activities were selected to illustrate different scientific and computational
concepts, taking into account that teachers and students are from a range of science and
mathematics backgrounds.

SUMMER WORKSHOP SURVEY RESULTS

The summer workshop includes science and mathematics teachers, with their teams
of up to four students per teacher. This provides the teachers with not only an
opportunity to acquire new ideas, but also an opportunity to use the new ideas
immediately, with students whose learning skills and styles are familiar to the teachers.
By having new curriculum materials and activities, as well as an opportunity to use the
new materials with students outside their formal classroom, it is hoped that the teachers
will be receptive to incorporating the ideas into their actual classroom materials during
the next academic year.
The student teams were made up of students from the teacher's most advanced class
or a group of students who expect to continue on in the next year with a specific science
teacher. Often, teachers reported that they had been recruited by their students for the
event, as the student teams formed first, usually from the appeal to the SAT e-mail list,
and then looked for a teacher to attend. Surveys were administered to both students and
teachers before the summer workshop and at the conclusion of the workshop in 2010.
The results are summarized here.

Student Survey Results

Due to the non-residential nature of the summer workshop, all students and teachers
attending were from the surrounding urban and suburban environment. The students
were 40% male and 60% female, and were evenly split between those that had just

74
 CCSC: Northeastern Conference

completed their sophomore year and those that had completed their junior year. The
student participants were all from public high schools and did not have any prior
programming language knowledge.
The student post-survey results ranked the VenSim Easy Java Simulations and
Predator-Prey highest, closely followed by the EasyJavaSim materials (Bouncing Ball,
Harmonic Oscillation, etc.). Students were surprised by the wide variety of programs
which were available. A table of selected comments is provided in Table 2.
Table 2. Student Post-Workshop Survey Comments
Q: How has the Summer Workshop changed your perceptions of computer science
and computational thinking as a learning tool?
A: I had no idea how complex computer science really was until I participated in this
workshop.
A: It made me realize it (computer science) can be fun and somewhat easy.
A: Computational science is vastly applicable.
A: It definitely raised the importance of computer science and computational thinking
for me.

Q: How has the Summer Workshop experience changed your perceptions of student
team learning in the sciences as a learning tool?
A: It is effective in small groups.
A: It made me realize the power of student team learning.
A: Student team learning is a great way to solve problems.
A: Teamwork seems very beneficial to such type of work.
Overall, all students felt they had learned more about computing and that
computational thinking is important. The importance of computational thinking to math
and science in high school was also strongly supported. The Summer Workshop was
highly recommended to others. Individually, the post-workshop student survey showed
more enthusiasm for computer science, computational science and information
technology as a career, as well as interest in applying to the host university for college
than had been demonstrated by the students in the pre-survey administered before the
workshop.

Teacher Survey Results

The teachers attending did not teach computer science courses or languages. Half
the teachers had been invited to attend the workshop by students - the student teams
assembled, based on the appeal to the SAT email list, and then the students looked for a
teacher to accompany them. This was quite interesting, as it showed strong
self-motivation on the part of the students, despite the student interest in mathematics and
scientific areas other than computer science. Teachers enjoyed the modules presented as
the students had. Table 3 provides representative examples of post-workshop teacher
comments.

75
JCSC 26, 6 (June 2011)

Table 3. Teacher Post-Workshop Survey Comments

Q: How has the Summer Workshop changed your perceptions of computer science
/computational thinking as a learning tool?
A: I can do it!
A: Should become more a part of our existing high school curriculum.
A: Various computational techniques can be used to solve complex mathematical
relations.
A: It has enhanced my ability to use different visuals to explain certain concepts to
my class.

Q: How has the Summer Workshop experience changed your perceptions of student
team learning in the sciences as a learning tool?
A: Reinforce my positive perceptions.
A: The techniques learned can be used in almost any science field.
A: That team learning may be the better way to learn and present certain concepts.
A: The students are amazing and can jump right in.
The most important part of the summer workshop was the change in perception of
computer science as a major and the host university as a choice for their students. After
the workshop, there was a 50% increase in the number of teachers who would recommend
CS, computational science, or IT to their students as a career and recommend that their
students consider attending college at the host university.

CONCLUSION
The results from the one-week summer workshop are very encouraging. Working
with local high school math and science teachers to increase their awareness and
understanding of computer science and computational science is an effective way of
reaching the educators teaching the next generation of computer scientists. Furthermore,
by keeping these educational professionals apprised of how computer science and
computational thinking infuses scientific and mathematic curriculums, future students
will be better prepared for computer science, computational science, and information
technology study at the university level. In a comparison between one-day in-service
professional workshops for teachers-only offered earlier in the year [10], the longer
(one-week) summer workshop offering changed the teacher perceptions to a greater
extent. This was probably due in part to the greater period of time for the summer
workshop, as well as providing teachers with the opportunity to work with their students
on the concepts and curriculum materials presented. However, the absence of computer
science teachers in the summer workshop may also be a factor in the greater change in
perception by the summer workshop teacher attendees. Both the one-day and one-week
workshops were effective in changing teacher perceptions.

76
 CCSC: Northeastern Conference

SUMMARY
Overall, the local, non-residential teacher-student summer workshops have been
very successful, with participants suggesting ideas for future summer workshops and
being impressed with the wide range of scientific fields which would lend themselves to
the computational techniques presented. A follow-up survey of both teachers and
students is planned, to determine the lasting impact on both teachers and students as a
result of their attendance at this summer workshop.
In addition to the significant increase in positive perceptions of computer science,
computational thinking and information technology expressed by teachers attending the
summer workshop, the student-led recruiting for the workshop, resulting from the use of
a SAT email list, was one of the most interesting outcomes. Like many other innovations
in computer science and information technology, a "bottom-up" approach to changing
teacher perceptions may be demonstrated to be highly effective in developing the next
generation of computer scientists.

REFERENCES
[1] Margolis, J., Estrella, R., Goode, J., Holme, J., and Nao, K. Stuck in the Shallow
End: Education, Race, and Computing. The MIT Press, Cambridge, MA, 2008.
[2] Cooper, S., Pérez, L. and Rainey, D. "K-12 Computational Learning",
Communications of the ACM, Volume 53, Number 11, November 2010, p. 29-29.
[3] Wilson, C., Sudol, L.A., Stephenson, C. and Stehlik, M., Running on Empty: The
Failure to Teach K-12 Computer Science in the Digital Age, ACM and CSTA,
2010. www.acm.org/runningonempty/
[4] Vesgo, J. "Enrollments and Degree Production in US CS Departments Drop
Further in 2006-2007", Computing Research News, Volume 20, Number 2,
March 2008, p.4.
[5] Lazowska, E. "Computing Research and Human Resources: The Current
Situation", CRA Computing Leadership Seminar, February 2005,
[6] www.cs.cmu.edu/cs4hs/ Explorations in Computer Science for High School
Teachers (2006-2010).
[7] Cuny, J. "Finding 10,000 Teachers", CSTA Voice, Volume 5, Issue 6, January
2010, p.1.
www.csta.acm.org/Communications/sub/CSTAVoice_Files/csta_voice_01_2010.
pdf
[8] www.computingportal.org/cs10k
[9] www.shodor.org and www.shodor.org/activities/
[10] Morreale, P., Joiner, D. and Chang, G., "Connecting Undergraduate Programs to
High School Students: Teacher Workshops on Computational Thinking and
Computer Science", Journal of Computing Sciences in Colleges, Volume 25,
Number 6, June 2010, pp. 191-197.

77
 AN INTERDISCIPLINARY APPROACH TO INJECTING

COMPUTER SCIENCE INTO THE K-12 CLASSROOM*

David Goldschmidt, Ian MacDonald, Judith O'Rourke

Department of Computer Science
The College of Saint Rose, 432 Western Avenue, Albany, New York 12203
{goldschd, macdonai, orourkej}@strose.edu

Brandon Milonovich
Department of Mathematics, Department of Teacher Education
The College of Saint Rose, 432 Western Avenue, Albany, New York 12203
[email protected]

ABSTRACT
As society's use of technology expands, students are using computers and
technology at increasingly younger ages; however, enrollment and graduation
rates in college-level computer science programs are decreasing [2,8]. As a
result, the need for professionals in the field of software development is
rapidly outpacing the supply of graduates [9]. According to The National
Association of Colleges and Employers, computer science ranks as one of the
top five highest-paying career paths and is projected to further expand within
the next decade [6]. To bridge the gap between poor enrollment and
technology-related job force needs, computer science topics must be
introduced at both the elementary and secondary education levels. In general,
students are exiting the K-12 program with a lack of skills or even exposure
to computer science and other STEM (Science, Technology, Engineering, and
Mathematics) fields [3,4,5]. Introducing creative and fun interdisciplinary
study into K-12 curricula holds promise for bringing more young scholars to
the field of computer science. In this paper, we explore potential computer
science topics and how they can effectively be integrated into K-12 curricula
in an interdisciplinary fashion to support STEM initiatives, including efforts
to also increase enrollment numbers for women and underrepresented
minorities [3,4,5].

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

78
 CCSC: Northeastern Conference

INTRODUCTION
Enrollment in degree-bearing computer science programs is failing to meet the
demand for computing professionals in the American workforce [9]. Overall, enrollment
and graduation rates at the college level in computer science are decreasing [2]. High
schools are reducing or eliminating computer science programs, as evidenced by the
College Board's eradication of the Advanced Placement (AP) Computer Science AB
exam this year [2,8]. Based on reports from the College Board and the Association of
Computing Machinery (ACM), as noted in [8], the number of students taking the AP
Computer Science exam has remained essentially constant since 1998. Other reports
show a rapidly declining trend in the number of secondary schools offering AP courses
or program in computer science (e.g. [4,5]).
In contrast to these disturbingly flat or declining trends in participation, AP exams
in Calculus, Biology, Statistics, Chemistry, Physics, and Environmental Science have all
shown steady increases since 1998 [8]. Regarding why computer science has seen such
stagnant numbers, NSF program officer Janice E. Cuny observes in [8] that "high-quality
computer-science instruction is all too rare in public schools."
The above trends are surprising, as computer science ranks third amongst the
highest-paying career paths in the United States and open positions are projected to
expand within the next decade [6]. A serious gap exists between education at the K-12
level and the available "pool" of computing professionals, which has led to a variety of
recent STEM (Science, Technology, Engineering, and Mathematics) initiatives, including
efforts to increase interest in computer science in women and underrepresented minorities
[3,4,5].
We believe that these trends are due to how much society takes the use of
technology for granted. Because our younger generations more readily and more
naturally use technology, there is a feeling that minimal computer and technology
instruction is required at the K-12 level. This is counterproductive, as evidenced by the
aforementioned lack of growth in students taking the AP Computer Science exam and the
decline in college-level computer science graduates. As further anecdotal evidence, we
often hear of veteran teachers and administrators at the K-12 level lacking the knowledge
or expertise in computer science to understand its importance in today's technological
world.
We propose that increased exposure to computer science topics via interdisciplinary
study at the K-12 level will encourage more young scholars to enter the field of computer
science. Though establishing computer science as a field of study at the K-12 level may
be a far-reaching goal, an interdisciplinary approach will hopefully enlarge the "pipeline"
of incoming college students in computer science and other STEM-related fields. This
paper presents a number of interesting and instructional topics for use in K-12 curricula
in support of STEM initiatives. While these topics may not be new, what is new here is
how they can be integrated in an interdisciplinary fashion, in some cases without students
knowing that they are about to learn a core topic of computer science.

79
JCSC 26, 6 (June 2011)

Interdisciplinary Computer Science in Physical Education: the "Binary Tango"

Computer science concepts can be integrated into physical education by combining
the binary number system with a physical activity synchronized by rhythm. The goal is

Figure 1: A snapshot of the "Binary Tango" exercise video showing binary

number 00101101
to illustrate counting using binary representation as it relates to standard 4/4 rhythms.
Student participants wear custom-made binary T-shirts with a large "0" printed on the
front and a large "1" printed on the back.
Capitalizing on the fact that 4 is a power of 2, we are able to relate each single
quarter note beat to the least significant bit (20), every other beat to the 21 bit, each
measure to the 22 bit, every other measure to the 23 bit, and so on. Participants stand
side-by-side and are assigned a bit position. Upon hearing their rhythm cue, students turn
180 degrees, flipping their bit from 0 to 1, or vice versa.
This process is repeated until we have reached the maximum binary value for all bit
positions (i.e. all 1s). And while this can correspond to music, it certainly does not have
to. A snapshot of this fun exercise performed without the musical connection described
above is captured in Figure 1 by eight of our undergraduates (with corresponding video
available online at http://www.youtube.com/CSatSAINTROSE).
Interdisciplinary Computer Science in Art using Alice
The New York State Education Department (NYSED) Standards for the Arts
provide school districts the basic structure for the development of local curricula that link
instruction and assessment to the content standards [7]. In particular, the New York State
Board of Regents outlines four learning standards. The first standard is "Creating,
Performing and Participating in the Arts." For the visual arts, part of what constitutes
meeting the standards for the elementary level is to experiment and create works of art
in a variety of mediums, including computer graphics. At the higher learning levels, it
is reflecting on the effectiveness of selected media to convey the artist's intended
meaning.
The second standard is "Knowing and Using Arts Materials and Resources," in
which it is suggested that students can use a drawing or paint program on the computer
to create graphic images and use the computer as a design tool to communicate visual
ideas. The third and fourth standards are "Responding to and Analyzing Works of Art"
and "Understanding the Cultural Dimensions and Contributions of the Arts."

80
 CCSC: Northeastern Conference

We propose that by using Alice [1], an introductory programming environment that

allows students to create three-dimensional visual stories, teachers can teach fundamental
computer science concepts while teaching art and also meet the required content
standards. Developed at Carnegie Mellon, Alice is a free software program in which
students can create visual stories or interactive games by dragging and dropping objects
into a scene, then directing them and their behavior. Students can select from hundreds
of graphical "objects" that can be rotated or moved in any direction.
Once the scene is developed, students create actions that the objects take; to do so,
students drag and drop instructions that correspond to programming statements in such
high-level object-oriented programming languages as Java, C++, and C#. Students using
Alice are able to visually express their stories and ideas, view the work of others, and
assess the effectiveness of computer graphic art form. Applicable at just about any grade
level, Alice provides a gentle introduction to fundamental concepts of computer science.
Further, teachers need little instruction or computer science expertise to use Alice; in
other words, learning to use Alice is straightforward and intuitive.

Interdisciplinary Computer Science in Music

In a typical undergraduate data communications course, students learn a variety of
methods for digitally encoding analog data. One of the most common examples of analog
data, and one of profound interest to students, is music. In the past decade, CDs have
been replaced by iPods and other portable MP3 players, yet many students are
disappointed and puzzled by the fact that CDs often sound better than their digital MP3
counterparts. Seldom do students learn that the answer lies in the encoding techniques.
For linear digital encoding, the digitization process approximates the height of the
wave at regular sample intervals, as shown in the progression from Figure 2(a) to Figure
2(b). Fortunately, digital encoding is a straightforward concept, accessible at many grade
levels. A simple explanation of the relationship between sounds and analog waveforms
(e.g. frequency components tied to pitch, amplitudes tied to volume) could be
comfortably introduced. An instructor can illustrate the difference between a
course-grained (i.e. low-bit) sample size and a fine-grained (i.e. high-bit) sample size.
Likewise, an instructor can demonstrate the effect of low versus high sample rates (i.e.
samples taken per second).
In both scenarios, of course, an instructor can play the resulting encoded tracks,
allowing students to hear the differences and relate them to the visual depictions shown
in Figures 2(a) and 2(b). A variety of fun exercises can be used to further illustrate these
concepts. At lower grades, a connect-the-dots game could be applied; at higher grades,
students could determine, by ear, where the optimal point is between compression and
sound quality.

81
JCSC 26, 6 (June 2011)

(a) (b)
Figure 2: A graphical illustration of quantization error in digital encoding: (a)
the original analog waveform with digital samples taken at regular intervals; and
(b) the reconstructed analog waveform taken from the digital samples, clearly
illustrating distortion

Interdisciplinary Computer Science in Mathematics using Scratch

Applying the use of the computer to mathematics instruction is not new. We add
to the evidence a case study at the Arbor Hill Elementary School within the Albany City
School District in Albany, New York. In 2010, over the course of three lessons in two
third-grade classes, Scratch [10] was integrated into the classroom, especially for
mathematics instruction, by author Brandon Milonovich, a secondary education major
with a focus on mathematics and computer science at The College of Saint Rose.
According to the teachers involved, there is observable evidence that several students
who were usually disengaged in their lessons found Scratch to be exciting and
interesting-and not just on first use-thus providing a means to motivating these students
to perform better in the classroom.
Through the program, students were exposed to 45 minutes of instruction in Scratch
centered on integrating mathematics into sequential programming. The first lesson
introduced students to the Scratch interface, while also allowing students to manipulate
movement of a Sprite within an x-y coordinate plane. This not only engaged students in
higher-order thinking skills, but also strengthened connections between what they have
seen in their mathematics instruction and the computer games many are used to playing
daily (or perhaps see older siblings playing).
The second week further increased the skill levels of students by introducing them
to manipulations of effects on the Sprites. This brought a higher-level of abstract
thinking to the number line concepts introduced with the coordinate plane in the first
lesson. During the third week, students started to develop the beginnings of their first
game, a word-counting game that provided students the means to develop sequential
programming that interacts with a user and counts words. Figure 3 shows an example
screen shot of the "programs" they developed.

82
 CCSC: Northeastern Conference

In future lessons on Scratch, students will complete the game and other challenges,
such as determining combinations of words to sum to various totals, thereby increasing
their understanding of number sense. Students will also develop the project further by
incorporating decision-making if-statements to verify answers given by players of the
game and create interactive learning games that will continue to develop their
understanding of mathematics and computer programming.

Figure 3: A screen shot of computer programming using Scratch, as

developed by third-graders

Based on the promise and success of Scratch, we now offer a 1-credit course in
Scratch for our elementary education majors at The College of Saint Rose, thus preparing
them to offer the same benefits, as evidenced above, to their future students at all grade
levels in mathematics, as well as other subjects.

Interdisciplinary Computer Science in Social Studies

From the ancient Mayan base-20 number system to the cryptic German Enigma code
during World War II, to the rapid expansion of the Internet and its impact on modern
cultures, computer science concepts have often played a pivotal role in history. Consider
the current online "social network" revolution that has taken hold of us all.
Unfortunately, many of these topics, often outside of an instructor's comfort zone, are
glossed over in social studies, history, and civics curricula.
Covering the details of how the Enigma code worked may be beyond a reasonable
scope; however, a light introduction to encryption may be accomplished though an
explanation of the classic Caesar cipher, which closely resembles today's cryptogram
puzzles often found in newspapers. Students may find it both educational and

83
JCSC 26, 6 (June 2011)

entertaining to explore the puzzles on their own, practicing a technique that reinforces
fundamental computer science material, logical thinking, and problem solving.

Interdisciplinary Computer Science in English and Foreign Languages

Students of English or foreign languages often learn specific rules of the language
through the application of a grammar. The grammar specifies how sentences are
constructed in the language, which directly parallels grammars for specifying computer
programming languages. While spoken languages tend to have numerous exceptions and
auxiliary rules, grammars used in computer science typically do not.
As students of a spoken language learn their grammar rules, an interesting topic to
incorporate here is how grammars for programming languages also produce valid
programming instructions or validate such instructions. Further, grammars for
programming languages typically do not require expertise in any spoken language.
Computer programmers may not speak the same spoken languages, but they are able to
communicate via the same programming language and successfully build a large software
system.

CONCLUSIONS
As noted above and in [8], there is a profound lack of educational resources
deployed to support computer science within the K-12 curricula. Unfortunately, such an
approach has led to a sharp decrease in enrollment and graduation rates of college-level
Computer Science programs. In support of STEM initiatives already underway to help
improve this trend, we have offered a number of interdisciplinary approaches to
integrating Computer Science into a variety of subject areas within the K-12 curriculum,
the key here being the interdisciplinary aspect of our approach. Nonetheless, this paper
offers only the "tip of the iceberg" and will hopefully motivate others at all educational
levels to increase student exposure to computer science. In addition to this, other future
work includes organizing, cataloging, and publishing these topics in a more uniform
manner for direct use by K-12 educators.

REFERENCES
[1] Alice, http://www.alice.org, accessed November 2010.
[2] Carey, K., Decoding the Value of Computer Science, The Chronicle of Higher
Education,
http://chronicle.com/article/Decoding-the-Value-of-Computer/125266,
November 7, 2010.
[3] Chute, E., Lack of diversity part of equation in STEM fields: Colleges try to
increase numbers of women, minorities in science and engineering, Pittsburgh
Post-Gazette, http://www.post-gazette.com/pg/09041/947952-298.stm, February
10, 2009.

84
 CCSC: Northeastern Conference

[4] CSTA National Secondary Computer Science Survey: Comparison of 2005, 2007
and 2009 Survey Results, http://csta.acm.org/Research/sub/CSTAResearch.html,
accessed January 30, 2011.
[5] Gal-Ezer, J. and Stephenson, C., The Current State of Computer Science in U.S.
High Schools: A Report from Two National Surveys, Journal for Computing
Teachers, Spring 2009, http://csta.acm.org/Research/sub/CSTAResearch.html,
accessed January 30, 2011.
[6] In Pictures: The College Degrees With The Biggest Salaries, Forbes.com,
http://www.forbes.com/2010/08/11/college-degrees-best-salaries-leadership-care
ers-jobs_slide.html, August 11, 2010.
[7] Learning Standards for the Arts,
http://www.emsc.nysed.gov/ciai/arts/pub/artlearn.pdf, Revised Edition, April
1996.
[8] Robelen, E., Schools Fall Behind in Offering Computer Science, Education
Week: Digital Directions,
http://www.edweek.org/dd/articles/2010/10/20/01bits-diged.h04.html, October
15, 2010.
[9] Science & Engineering Degrees: 1966-2006, Detailed Statistical Tables, NSF
08-321, http://www.nsf.gov/statistics/nsf08321/pdf/nsf08321.pdf, October 2008.
[10] Scratch, http://info.scratch.mit.edu/Educators, accessed January 29, 2011.

85
 METHODOLOGIES AND TOOLS FOR THE SOFTWARE

QUALITY ASSURANCE COURSE*

Vladimir V. Riabov
Department of Mathematics and Computer Science
Rivier College
Nashua, NH, 03060
603 897-8613
[email protected]

ABSTRACT
Tutorials, labs, projects, and homework assignments were designed to help
students explore modern techniques of software quality assurance; debugging
C/C++ and Java codes; and developing high-quality computer projects.
Different methods (predicate-logic and topological approaches of graph
theory; metric theory of algorithms, and object-oriented methodology of rapid
prototyping) have been explored by using various tools in analyses of complex
computing code. Applications cover software test strategies, code reusability
issues, and ways to significantly reduce code errors and maintainability effort.
The related course materials provide students with knowledge, instructions and
hands-on experience, and motivate them in their research studies.

1 MOTIVATION
Students pursuing careers in software development should be familiar with various
methods of software quality assurance (SQA). A year ago we launched a new SQA course
that addresses the issue of quality throughout the software development process,
including system analysis and design, rapid prototyping, implementation, testing, and
delivery. Special attention is given to setting quality standards [1], developing quality
measurement techniques, writing test plans, and testing the user interfaces. It becomes a
challenge for an instructor to provide students with the state-of-the-art hands-on
technology-exploration experience in this field.

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

86
 CCSC: Northeastern Conference

Topics covered include the integration of quality activities into the software project
life cycle, CASE tools overview, structured testing methodology, complexity and
object-oriented metrics, configuration management, capability maturity models, software
engineering practices and standards, code re-engineering strategies, and miscellaneous
topics.
Students were encouraged to examine various software-development projects.
Exploring different testing strategies, they analyzed computer code by using various
software-organization metrics including cyclomatic complexity [2], Halstead's [3] and
object-oriented metrics [4], and re-designed the code with lower risk level and errors.
This paper contains an overview of the SQA software tools, tutorials, lab manuals,
homework assignments, project reports, and research papers of students who took the
Software Quality Assurance course. The advantages of using these tools for instruction
in online and hybrid courses are also discussed.

2 TOOLS FOR SOFTWARE QUALITY ANALYSIS

2.1 Industrial SQA Tool Used in Software Engineering
One of the most popular SQA tools, the McCabe™ IQ software package, was
selected for exploring various study cases and projects in the course on software quality
assurance. This industrial tool became available for students under the free-license
agreement with the McCabe™ University Program. The tool allows them to explore
McCabe's structured testing methodology [2] that became a widely used method in code
analyses, unit and integration test planning, and test-coverage estimations. Following
specially designed computer-lab assignments and using the McCabe™ IQ tool, students
study how to apply the theory of graphs for the complexity code analysis, develop test
strategies, and predict possible errors [3, 6] in the code developed by themselves and
companies. Unfortunately, the McCabe™ IQ package could be used only on campus;
therefore, other Open Source free-license SQA packages were evaluated for use by
students at home, specifically when taking courses online or in the hybrid format.

2.2 Open Source Free-License SQA Tools

There are several Open Source free-license SQA tools available for students. The
Java Source Metric™ package [7] has been used to analyze Java source code with quality
metrics like the Inheritance Depth, Lines of Code, and McCabe Complexity Metric suite.
The CCCC™ tool [8] generates a report on various metrics (including the Lines of Code
and McCabe's complexity) of C/C++ code. The freeware program SourceMonitor™ [9]
has been used for code analysis to identify the relative complexity of code modules.
SourceMonitor™ measures metrics for source code written in C++, C, C#, VB.NET,
Java, Delphi, Visual Basic (VB6), and HTML. It operates within a standard Windows
GUI and exports metrics to XML or CSV (comma-separated-value) files for further
processing with other tools. The COCOMO-II™ tool [10] was used by students to
estimate the cost, effort, and schedule associated with their software development
projects.

87
JCSC 26, 6 (June 2011)

3 LAB AND HOMEWORK ASSIGNMENTS

The main goal of labs and homework assignments is to introduce software quality
metrics and help students build their individual skills of code analysis, testing, and
redesign to improve code quality and enable possible reuse in other projects.

3.1 Introducing the Structural Testing Methodology

The first set of lab and homework assignments deals with implementation of the
structured testing methodology offered by McCabe [2]. The approach is based on
graph-theoretical complexity-measuring techniques in code studies and control of
program complexity. Using the experimental results of Miller [5], McCabe suggests that
code modules approach zero defects when the module cyclomatic complexity is less than
10. During lectures, the instructor provided an overview of the graph-based complexity
metrics and the results of his systematic metric analyses of software for two industrial
networking projects [6]. Following the lab assignments, students explored the McCabe
™ IQ tool and used it to perform metric analyses of several codes by applying cyclomatic
complexity (v), essential complexity (ev), module design complexity, system design
complexity, and system integration complexity metrics [2] in order to understand the
level of complexity of a code module's decision logic, the code's unstructured constructs,
a module's design structure, and the amount of interaction between modules in a program,
as well as to estimate the number of unit and integration tests necessary to guard against
errors.

3.2 Estimating the Number of Code Errors and Efforts to Fix the Errors
The second group of the lab and homework assignments was designed to introduce
students to the comparative analyses of algorithm implementations in different languages
(FORTRAN, C, C++, Java and some others). Following Halstead's procedures [3],
students identified all operators and operands, their frequencies, and estimated the
program length, vocabulary size, volume, difficulty and program levels, the effort and
time amounts to implement and understand the program, and the number of delivered
bugs (possible errors), B. They compared their findings with values calculated by using
SQA tools (McCabe™ IQ, Java Source Metric™, CCCC™, SourceMonitor™, and
COCOMO-II™), and found that the results are sensitive to the programming language
type (procedural or object-oriented).
In particular, students found that efforts to implement and understand the program
were higher for procedural languages (FORTRAN and C) than for the object-oriented
language (Java), even for simple algorithms, like Euclid's algorithm for calculating the
Greatest Common Divisor. They also found that large C/C++ source files [6] contain
more actual errors than the number of delivered bugs (B) suggested [3].

3.3 Interpreting Object-Oriented Metrics

The third group of the lab and homework assignments was developed to help
students identify clusters of object-oriented metrics that would better describe the major

88
 CCSC: Northeastern Conference

characteristics of object-oriented systems (properties of classes, polymorphism,

encapsulation, inheritance, coupling, and cohesion) implemented in computer code
written in C++ and Java. Their findings are summarized in Table 1 below.
Metric Class Polymor- Encapsu- Inheri- Coup- Cohe-
phism lation tance ling sion
Weighted Methods Per Class Yes Yes
Response For Class Yes Yes
Percentage of Public Data Yes
Accesses to Public Data Yes
Lack of Cohesion of Methods Yes Yes Yes
Number of Children Yes
Number of Parents Yes
Depth in Inheritance Tree Yes
Coupling Between Objects Yes Yes
Attribute Hiding Factor Yes Yes
Method Hiding Factor Yes Yes
Polymorphism Factor Yes Yes
Table 1. Clusters of Object-Oriented Metrics
Students also identified some specific object-oriented metrics (Weighted Methods
per Class, Response from a Class, Lack of Cohesion between Methods, and Coupling
between Objects) that are the important factors for making a decision about the code
module/class re-usability.

3.4 Comparing Two Releases of the Code

The last, fourth set of the lab and homework assignments was offered to students
to identify major factors that forced programmers to change the code [6] in the project
redesign efforts. After analysis of the project software (about 300,000 lines of C-code)
using the network-protocol approach 271 modules of the old Code Release 1.2 [6] were
recommended for redesign. The re-engineering efforts resulted in the deletion of 16 old
modules and in the addition of 7 new modules for the new Code Release 1.3. Analyzing
the deleted modules, students found that 7 deleted modules were unreliable (v > 10) and
6 deleted modules were unmaintainable (ev > 4). Also, 19% of the deleted code was both
unreliable and unmaintainable. Moreover, all seven new modules are reliable and
maintainable.
After redesign, code changes resulted in the reduction of the code cyclomatic
complexity by 115 units. 70 old modules (41% of the code) were improved, and only 12
modules (about 7% of the code) deteriorated. This analysis demonstrates a robustness of
the structured testing methodology and successful efforts in improving the quality of the
Code Releases. Studying the relationship between software defect corrections [6] and
cyclomatic complexity, students found a positive correlation between the numbers of
possible errors, unreliable functions (with v > 10), and error submissions [6] from the

89
JCSC 26, 6 (June 2011)

Code Releases (see Fig. 1) in the implementation efforts for six network protocols (BGP,
FR, IP, ISIS, OSPF, and RIP).

Figure 1: Correlation between the Number of Error Submits, Number of Unreliable

Functions (v > 10), and the Number of Possible Errors for Six Network Protocols.

4 STUDENTS' PROJECTS AND RESEARCH STUDIES ON SQA

As the main assignment for the SQA course, students were asked to develop their
individual or team projects on the quality analysis of moderate-size computer programs
written in an object-oriented language (C++ or Java) and compare different releases of
the code. The project assignment included the study of the code complexity and quality
based on the analysis of cyclomatic complexity metrics, Halstead's metrics, and
object-oriented design metrics by using available SQA tools (McCabe™ IQ, Java Source
Metric™, CCCC™, SourceMonitor™, and COCOMO-II™). The project reports included
the code structure chart, "Battlemap"; McCabe's Complexity Metrics summary; examples
of flowgraphs with low, moderate, and high complexity; scatter diagrams with
identification of unstructured-unreliable, reliable-structured, unreliable-structured, and
reliable-unstructured modules; examples of flowgraphs with the independent paths for
the unit tests; Halstead's Metrics report with estimation of the total number of the
delivered errors (B); Object-Oriented Metrics report with basic interpretation of the
metrics; and recommendations to improve the code.
Michael Jeffords in his project, "Using SQA Metrics to Guide Refactoring of
Medium to Large Scale Projects", developed a plan to reduce the overall complexity of
the code while adding important functionality to the system that provided visualization
of geophysical data. His code refactoring efforts took approximately 20 hours, but have
reduced the number of tests by about 200 unit tests. Summarizing these efforts, he offered
two effective methods of refactoring. In the first method, he used weighting factors to
"boil" McCabe's and Halstead's metrics down using a formula M = v/10 + ev/4 + V/3000.
This approach gives equal weights to v-cyclomatic and ev-essential complexity metrics.
Combined they have double the weight of Halstead's V-metric of the program volume [3].
In the second method, three thresholds for cyclomatic complexities were used to chart
improvement over time. His goal was to achieve 100% of methods below the cyclomatic
complexity of 20 and less than 97% of methods below the cyclomatic complexity of 10.
He achieved 100% for v < 20; 99.67% for v < 15, and 98.01% for v < 10. Michael
described three stages of refactoring: 1) He started by refactoring classes to remove

90
 CCSC: Northeastern Conference

public variables and adding accessors; 2) Next, he chose any of the methods that ended
up on the "hot" list that he could understand the basic methodology, and then tried to
tighten up the algorithms; 3) Finally, he worked on the hardest methods once he felt that
his plan had taken shape and he could see a direction for fixing the architecture of the
worst classes.
In the other project, Timothy Houle and Douglas Selent analyzed the complexity of
the Light-Up Puzzle program [11], its maintainability, testability and metrics related to
the quality of the software program's design. The application's vulnerabilities were
identified by using the McCabe™ IQ tool. This approach helped them to identify
vulnerable code areas, reduce error rates, shorten testing cycles, improve maintainability,
and maximize reusability. In order to verify the effectiveness of the McCabe™ IQ tool,
they re-factored the program in the areas reported to be highly complex and error-prone.
After this, they compared the McCabe's Metrics reports on the initial analysis to the
reports on the re-factored analysis and charted results to clearly indicate the
improvements made to decrease complexity. In addition to the McCabe metrics, the
students also added various UML diagrams [1], which helped them understand the
concept of the program structure and identify areas where object-oriented design
principles could be applied to increase code reusability and maintainability.

5 EFFECTIVENESS OF THE COURSE AND STUDENTS' RESPONSE

All 16 computer-science graduate students that took the SQA course in the spring
of 2010 expressed in their course evaluations full satisfaction with course organization,
content, and material delivery. The overall course-evaluation score was high (4.71 out of
maximum possible 5.0). In their anonymous comments, students shared mostly favorable
observations, e.g., "This is a practically-oriented course and it has a lot of demand in the
job market…"; "The instructor opened my eyes to the importance and usefulness of SQA
metrics…"; "I learnt a lot of how the SQA is handled in the software development
process…"; "Great experience…", and "I met my expectations more than I thought."
Later these students effectively used the SQA techniques to improve their software
programs in other courses, including the final capstone projects. All the students are
currently employed by local computer companies.

6 CONCLUSIONS
The author has described the challenges and experience of running software quality
assurance courses for undergraduate seniors and graduate students. The knowledge of
graph theory and its applications in software engineering is beneficial for students with
Computer Science majors. Detailed analysis of code complexity reveals areas of code
structures that should be revised. The code revision allows students to find those code
areas with potential errors and to improve code design and testing practices. In particular,
the resulting analysis can be used in identifying error-prone software, measuring optimum
testing efforts, predicting the effort required to maintain the code and break it into
separate modules, reallocating possibly redundant code, and providing a fundamental
basis for unit and integration testing. The complexity code analysis and structured testing

91
JCSC 26, 6 (June 2011)

methodology should become a necessary attribute of software design, implementation,

testing, sustaining, and re-engineering practice and training.

REFERENCES
[1] Galin, D., Software Quality Assurance: From Theory to Implementation, New
York: Pearson Education, 2004.
[2] Watson, A. H., McCabe, T. J., Structured Testing: A Testing Methodology Using
the Cyclomatic Complexity Metric, NIST Special Publication, No. 500-235.
Gaithersburg, MD: National Institute of Standards and Technology, 1996.
[3] Halstead, M. H., Elements of Software Science. New York: North Holland, 1977.
[4] Rosenberg, L.H., Applying and Interpreting Object-Oriented Metrics. In
Proceedings of the Tenth Annual Software Technology Conference, April 18-23,
1998. Salt Lake City, UT, 1998.
[5] Miller, G., The Magical Number of Seven, Plus or Minus Two: Some Limits on
Our Capacity for Processing Information. In Psychological Review, 63 (2),
81-97, 1956.
[6] Riabov, V. V., Networking Software Studies with the Structured Testing
Methodology. In Computer Science and Information Systems. Athens, Greece:
ATINER, 2005, pp. 261-276.
[7] Java Source Metric, SourceForge.Net, 2009,
http://jsourcemetric.sourceforge.net/, retrieved January 25, 2011.
[8] Littlefair, T., CCCC: C and C++ Code Counter, SourceForge.Net, 1998,
http://cccc.sourceforge.net/, retrieved January 25, 2011.
[9] SourceMonitor Freeware Program, Campwood Software, Inc., 2010,
http://www.campwoodsw.com/sourcemonitor.html, retrieved January 25, 2011.
[10] COCOMO II, University of South California, 2010,
http://sunset.usc.edu/csse/research/COCOMOII/cocomo_main.html, retrieved
January 25, 2011.
[11] Light-Up Puzzle Program, Puzzle-Loop.Com, 2009,
http://www.puzzle-light-up.com/, retrieved January 25, 2011.

92
 A SOFTWARE ENGINEERING COURSE WITH A

LARGE-SCALE PROJECT AND DIVERSE ROLES FOR

STUDENTS*

Bonnie K. MacKellar
Division of Computer Science, Mathematics and Science
St. John's University
Queens, NY 11439
718-990-7452
[email protected]

ABSTRACT
Industry software projects tend to be large enough and involve enough people
that no one person can understand the entire project in detail. Teams are
structured into groups that are responsible for different features of a product,
with coordination between groups required. Typical industry software projects
also involve people in many non-programming roles. Undergraduate software
engineering courses, however, tend to be based on small projects that only
require communication within each group of students rather than among
groups. In this paper, we describe a software engineering course which uses
a large-scale class project and diverse student roles to simulate a
heterogeneous development environment.

INTRODUCTION
Software engineering is one of the core knowledge areas listed in the 2008 ACM
Curriculum Update[1], which notes that software engineering is one of the topics that
"receives attention in the dialog with industry". As a result, most computer science
programs offer some type of software engineering course. Students often mention their
software engineering projects in their resumes when job hunting and employers tend to
be highly interested in what students have learned in this course. Our university is no

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

93
JCSC 26, 6 (June 2011)

exception; software engineering is a required course in our computer science major. At

our school, students enroll in the software engineering course with various backgrounds
- some have done little programming and others know quite a bit. Many students have
transferred in from community colleges. A number of our students do not aspire to careers
as software developers and are more interested in testing, user interface design, or project
configuration. These students want to learn about the software project management
process at a fairly high level, and need to know more about non-programming roles - a
topic that is often given short shrift in computer science programs. On the other hand,
some students have very strong programming skills and may be working at software jobs
in industry while taking the course. These students are interested in a software
engineering course that is focused on program design and development. Our core problem
was to meet these diverse needs while also realistically simulating an industry software
development project.
Problems, however, often lead to opportunities. We observed that at many
companies, development teams are formed into cross-functional groups based on role -
for example, a QA group, a technical writing group, or a user interface group. Thus, we
organized this course to mirror a role-based organization. Some students were placed on
a QA team, others on a user interface standards team, and others on subproject
development teams. In order to accomplish their work, students had to communicate not
just within their teams but with other teams as well. Once we had decided to include
non-developer roles in the project structure, we realized that the best way to do this was
with one large project that involved the entire class.
Traditionally, software engineering courses are based on standalone projects suitable
for smaller teams of 4 to 5 students. A literature survey [11] on project group
characteristics stated as one of its findings that 4 to 5 participants is most optimal from
a course management perspective. Many other papers describe small project teams in
software engineering, including [7], [6], and [9]. Anewalt[2] describes a course structure
in which students rotate between groups, but the groups are still small (3 to 4 students)
and are not interdependent in any way. In typical SE courses, students usually have
homogenous roles in the project - all work on requirements together, then design, and
then coding. Communication occurs between individuals in each group rather than among
groups.
This is not typical in real-world software development, where unless one is working
in a very small startup company, people tend to interact with a larger number of people
who have heterogeneous backgrounds and roles. According to Bruegge and Dutoit, "The
development of software brings together participants from different backgrounds, such
as domain experts, analysts, designers, programmers, managers, technical writers, graphic
designers, and users. No single participant can understand or control all aspects of the
system under development…". [3]. It is precisely this experience that is hardest to
duplicate in a typical software engineering course. Often students come away saying they
could have gotten the project done more quickly if they didn't have to work in groups.
This happens because the type of software project that can be finished by 4 students in
a semester is small enough to be understood by one person; hence the feeling that the
overhead of working in a group has merely made the task more difficult.

94
 CCSC: Northeastern Conference

In many software companies, developers may be divided into smaller groups based on
components or project features, but these smaller groups are interdependent. A decision
made in one group about a data representation, for example, may well impact what other
groups can do. It is also very common to have non-development groups, such as
documentation specialists or testing specialists, organized into their own groups, who
then interact with all of the feature-oriented teams. It is especially common for the QA
engineers to be organized into a separate group, in order to ensure independence from the
developers. This type of organization, with its need for cross-group coordination, is the
way in which we organize our software engineering course.
There have been a few large scale software engineering course projects described
in the literature. In a 1991 paper surveying approaches to the software engineering course,
Tomayko and Shaw[13] strongly endorsed the use of whole-class projects. However, this
did not seem to catch on in the wider community, since only a handful of reports on this
type of project have appeared in the literature since that paper appeared. Nita-Rotaru,
Dark and Popescu [8] describe a course that integrates students of varying backgrounds
into one large development team; however, this course involves graduate students as well
as upperclassmen at a large engineering school - a very different demographic from our
students. In addition, the group organization described in that paper is quite different from
ours. Their organization is solely in terms of features, and seems more typical of a large
research effort than a commercial software development effort. Rebelsky and Flynt [10]
describe a whole class project in which students work in groups focused on project
subcomponents; this course did not include non-developer roles such as testing or user
interface design. Tan and Philips[12] describe a course setup that allows for a number
of different development roles; however, the project teams are small (5 students per
team). Finally, Coppit and Haddox-Schatz [5] describe a whole-class project in which
task allocation and grading is tightly tied to the issue tracking system. This is an
interesting approach but it seems very complex and may be difficult for students to
manage.
In the course reported in this paper, students are assigned to groups and tasks by the
instructor who functions as a project manager. The students submit resumes as their first
assignment, and these resumes are used to allocate students to groups based on
background, skills, and interests. Just as in industry, the main criterion for assigning
students to groups is the likelihood of project success. Since the aim of this course is to
teach students how to function in a real-world development environment rather than to
teach them any specific programming skills, we feel this approach to group formation is
justified. The students are informed of the criteria as they write their resumes. In addition,
we use this exercise as an opportunity to discuss resume-writing skills.

COURSE STRUCTURE
The first semester, the course was one semester long and met 3 times a week for an
hour each session. Currently, the course is being taught in a once-a-week, 3 hour format.
The enrollment for the course is capped at 25 students. The class time is structured so that
some time is spent in traditional lecture, presenting topics such as project management,
UML, quality assurance, and requirements gathering. Time is also spent in hands-on lab
activities, learning to use development systems such as Bugzilla and Subversion, as well

95
JCSC 26, 6 (June 2011)

as some of the specific technologies that will be needed for project development. Students
enrolled in this class had, for the most part, never used a version control system or a bug
tracking system before.
The class project was the development of a website for a weekend children's
Chinese school. This project exemplified a typical web-oriented software project - a web
browser as front end, a middle layer of "business logic", and a database backend.
Depending on the component, students used Java, C++, JavaServerFaces, SQL, HTML,
and CSS. The project was divided into a number of subprojects: the student and teacher
management module, a set of teaching applications including pinyin and traditional
character flashcards and a dictionary, a parent volunteer scheduling application, and a set
of database reports. A small student team was assigned to each of these development
projects, based on their interests and backgrounds as stated on their resumes. In addition,
some students were assigned to the following cross functional groups: QA, user interface
standards, technical writing, and build engineering.
Since students now no longer had uniform roles and tasks in the course, a different
set of deliverables and grading policies needed to be created for each group. All students
participated in the initial writing of the requirements document. Once the requirements
document was created, however, tasks and schedules began to diverge. The teams
working on development subprojects followed a traditional software engineering course
timeline of first submitting a detailed design, then submitting milestones as well as
weekly status reports. Once QA testing commenced, the weekly status reports had to
document responses to bug tickets as well. The QA team, which had three members, was
responsible for creating test plans from the requirements document, executing the tests,
tracking results and submitting tickets in Bugzilla, and then working with the subproject
teams who were responsible for tickets. Their final report consisted of all test plans,
showing the status of tests. The user interface group was responsible for designing an
overall "look and feel" for the website, documenting user interface standards, helping the
other groups achieve the user interface standards, and coding the Cascading Style Sheets
and the top level HTML page that tied together all of the subprojects. The user interface
group was responsible for a user interface report which documented the look and feel, and
justified it in terms of user interface standards. The technical writing group built an online
help system. They initially worked from the requirements and detailed design documents
created by the class to create their first milestone effort. Then they modified their initial
help system to reflect what the developers had actually created. The QA group included
the online help system in their test plans, so the user interface group had to respond to
bug tickets just like the other subproject groups. The role of the build engineering group
was to create an automated build and deploy the system onto a Linux server.
Unfortunately, they did not get as far as the other groups, largely because of the
complexity of the build requirements (some subproject groups brought in third party
software which greatly complicated the build) and lack of knowledge about application
server environments.

96
 CCSC: Northeastern Conference

GROUP COMMUNICATION
One of the key goals of the course was to create a project large enough that students
would have to communicate and coordinate between groups as well as between
individuals. With this project, inter-group coordination fell into two categories:
• Software coordination: The groups had to agree on certain common software
functionality as well as database structure, and coordinate changes to the software
and database as they came up. Coordination between the groups to achieve a
consistent look and feel was also necessary; this was the responsibility of the user
interface group.
• Task coordination: The groups also had to coordinate in terms of tasks. For
example, the QA team had to communicate bugs to the relevant development teams,
and the students responsible for fixing bugs had to communicate status back to the
QA team. The user interface group had to communicate interface standards to the
development teams, and work closely with each group in order to achieve a
consistent interface.
A number of communication mechanisms were used to coordinate between the groups,
such as email and discussion boards. One communication mechanism which was very
successful was the use of in-class presentations. Each group had to do an in-class
presentation at about the time that serious development was starting, to introduce the
group and their role. The discussions at these presentations were very animated as each
group suddenly realized their interdependency with other groups.
We also used class time to coordinate between groups. Since almost all students in
the course are commuting students with outside jobs, we recognized that it would be very
difficult for these students to regularly meet outside of class. As it turned out, this time
was frequently used for inter-group coordination, often because students had trouble
identifying which member of a different group to communicate with when in a non
face-to-face setting, or the best means of coordinating with that person. This difficulty in
identifying information sources in other groups was interesting and fits with work done
with professional software teams [4].

ASSESSMENT
In order to measure student perspectives on what they had learned in the course, a
2 part survey was developed and administered at the end of the semester. We will use this
survey again at the end of the current semester. The first part of the survey consisted of
7 questions measuring their beliefs about how well they had learned and understood
various aspects of the course. The survey scale was 5=strongly agree, 4=agree, 3=neutral,
2 =disagree, 1=strongly disagree. Table 1 shows the results of the first part of the survey.
Interestingly, the question asking if the student is likely to pursue a career in
software development had the most "Strongly disagree"/"disagree" results - a total of 4.
In addition, 3 were neutral, meaning that 7 of the 17 students in this software engineering
class were not aiming towards a career in software engineering. From conversations with
the students, it was clear that some of the students came into the course with plans to
pursue a non-development career. It would be interesting to measure this attitude at the

97
JCSC 26, 6 (June 2011)

beginning of the course as well as at the end, to see if there are any changes as a result of
the course.

Table 1. Results of student survey

Question 1 2 3 4 5 Mean
respons
e
I feel I understand what a professional 0 0 4 9 4 4
software developer does on the job.
I am comfortable that I could participate 0 1 8 5 3 3.59
in the planning and development of a real
world software project
I understand the role of testing in a 0 1 1 11 4 4.05
software project.
I feel that I know how to use the kinds of 0 0 5 12 0 3.70
tools that software engineers use on the
job.
I am likely to pursue a career in software 1 3 3 4 6 3.65
development
I feel more comfortable now with the 0 1 4 6 6 4.00
idea of getting a big job done in a large
team.
The tools available to the class helped me 0 1 5 6 5 3.88
get my job done.

The second part of the survey listed 10 class topics and asked students to list the
ones they had learned the best, the ones that were the most difficult, the ones that they felt
were most important to their future career, and the ones that they felt were least important
to their future career. The questions were also more open ended and allowed them to
insert comments. A majority (10) of the students felt that the topic they learned best was
requirements writing, even though only a couple of weeks were spent on this topic in the
course. This is probably because the textbook had a lot of coverage of the topic, and also
because it was the least technical topic. Seven students believed they had best learned
group communications, and also that this was a topic that is very important to their
careers. The fact that communication within and among groups was such a key
component of the course probably led to that result. And not surprisingly, students chose
the most technical topics as the ones they had the most difficulty with - web programming
and automated tools. Because of the lack of strong prerequisites, many students were
very weak in their technical preparation.

98
 CCSC: Northeastern Conference

DISCUSSION AND CONCLUSION

The goals for this course were to realistically simulate industry software
development by using a project big enough to require students to be divided into
interdependent groups and to productively integrate students with varying backgrounds
and skills into the course. Students responded well to the course organization and were
active and engaged throughout the semester. Students with little programming
background were able to make important contributions by developing the online help
system or designing and carrying out test plans. These roles were far from being a
backwater and forced the students to interact extensively with every other group,
following the product through all of its changes. Students with experience developing
web-based interfaces were able to use those skills in the user interface group, and students
with skills in database development could contribute that knowledge. The result was that
students came away with a better understanding of the different interdependent roles
within a software product group. The course has been offered a second time this past fall,
and is currently being offered again this spring. We made some adjustments, using the
experience of the first semester. We used Swing instead of a web-based project to
simplify deployment details. We provided more explicit guidance to the students in
recognizing risk factors in their projects or roles early on. And finally, we are tracking
communication patterns among the students closely, with plans for future research into
modeling these communication patterns.
In this paper we describe a problem - lack of uniform background and skills among
students taking a software engineering course - that was turned into an opportunity to
make the course better reflect the way that software is produced in many companies.
Since real world software development makes use of a relatively large number of people,
all with differing backgrounds, talents, and roles in the project, we structured the course
to reflect that. The project was large enough to require different specialized roles also
involved students in a situation where each individual development team impacted the
work of other development teams. As a result, students learned the importance of
communication and coordination among groups, and the varying roles that project team
members may play in a software project.

REFERENCES
[1] 2008. Computer Science Curricum 2008 :An Interim Revision of CS2001.
Association for Computing Machinery and IEEE Computer Society.
[2] Anewalt, K. 2009. Dynamic group management in a software projects course.
Journal of Computing Sciences in Colleges. 25, 2 (2009), 146-151.
[3] Bruegge, B. and Dutoit, A.H. 2009. Object Oriented Software Engineering Using
UML, Patterns, and Java. Prentice Hall.
[4] Cataldo, M. et al. 2009. Software Dependencies, Work Dependencies, and Their
Impact on Failures. IEEE Transactions on Software Engineering. 35, (2009),
864-878.

99
JCSC 26, 6 (June 2011)

[5] Coppit, D. and Haddox-Schatz, J.M. 2005. Large team projects in software
engineering courses. Proceedings of the 36th SIGCSE technical symposium on
Computer science education (St. Louis, Missouri, USA, 2005), 137-141.
[6] Ellis, H.J.C. 2006. An evaluation of learning in an online project-based web
application design and development course. Journal of Computing Sciences in
Colleges. 21, 6 (2006), 217-227.
[7] Guo, J. 2009. Group projects in software engineering education. Journal of
Computing Sciences in Colleges. 24, 4 (2009), 196-202.
[8] Nita-Rotaru, C. et al. 2007. A multi-expertise application-driven class.
Proceedings of the 38th SIGCSE Technical Symposium on Computer Science
Education (Covington, Kentucky, USA, 2007), 119-123.
[9] Pieterse, V. et al. 2006. Software engineering team diversity and performance.
Proceedings of the 2006 Annual Research Conference of the South African
institute of Computer Scientists and Information Technologists on IT Research in
Developing Countries (Somerset West, South Africa, 2006), 180-186.
[10] Rebelsky, S.A. and Flynt, C. 2000. Real-world program design in CS2: the roles
of a large-scale, multi-group class project. SIGCSE Bull. 32, 1 (2000), 192-196.
[11] Richards, D. 2009. Designing Project-Based Courses with a Focus on Group
Formation and Assessment. Transactions on Computer Education. 9, 1 (2009),
1-40.
[12] Tan, J. and Phillips, J. 2005. Real-world project management in the academic
environment. Journal of Computing Sciences in Colleges. 20, 5 (2005), 200-213.
[13] Tomayko, J. and Shaw, M. 1991. Models for Undergraduate Project Courses in
Software Engineering. Technical Report #CMU/SEI-91-TR-10. Software
Engineering Institute, Carnegie Mellon University.

100
 SOURCE CODE ANALYSIS AND ATTACK TECHNIQUES AS

EDUCATIONAL APPROACHES TO TEACH SECURE

PROGRAMMING*

W. Scott Harrison and Nadine Hanebutte

Department of Mathematical and Computing Sciences
St John Fisher College
3690 East Ave.
Rochester, NY, 14618
585-385-8000
[email protected]

ABSTRACT
We believe that a sound knowledge of security is necessary for students, and
in this paper, we intend to justify this statement and show how this can be
accomplished using a focused set of two courses. We will describe the layout
of the courses in such a way that it would be possible for an instructor to
implement our model. Although the paper makes references to different types
of security issues, such as race conditions or buffer overruns, it is not
necessary to have technical knowledge of what these terms mean. Space
limitations restrict our ability to explain these terms; however, we have
included citations to explanatory materials.

INTRODUCTION
The need for computer security is not debatable. Daily we hear about lost databanks
of social security numbers, private records, as well as nation-states conducting
cyber-warfare on their enemies, as happened between Russia and Georgia in 2008 [1].
However, this does not mean that all computer science students will need to be
security experts. Certainly there is a need for these types of graduates; however, there is
a need in all fields of computer science. We argue, then, that even if a student does not

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

101
JCSC 26, 6 (June 2011)

intend to go into the area of computer security, a knowledge of the field is a benefit. In
fact, in its commentary to the 2008 Computer Science curriculum update, the ACM
suggests the security should be added into computer science core classes, in particular,
programming classes [2]. Graduates in computer science will, logically, be programmers,
managers, software engineers, and the like. All of these groups of people will benefit
from knowledge of basic computer security issues.
It is critical, we feel, that students understand the impact of the choices they make
when writing code or making design choices. We have argued in the past that poor
programming leads to security issues [3]. A buffer overrun, which involves copying data
into a too-small allocated memory space, can cause an attacker to take complete control
of the program, and, possibly spawn a shell giving the attacker access to user or
administrative rights (an excellent, if somewhat dated, discussion of this can be found in
[4]). We have done a survey of introductory programming texts, and discovered that
most, if not all of them, contain serious security flaws in their own examples and
explanations [5]. This leads students to writing software containing security flaws, which
of course can lead to serious repercussions later.
Students should be keenly aware of what actually happens when they do, for
example, a strncpy as opposed to strcpy (the former can restrict the size of data to be
copied, while the latter does not, in C). The question, then, is how do we make them
understand this? We can tell them, but simply stating that it should not be done leaves
something to be desired. The next logical question is why? And instead of simply
answering the question, we reverse it on the students; we ask them why. The method by
which we do this is through a series of two in-depth courses aimed first at identifying
security flaws, and then actually using these security flaws as an attacker would; seeing
exactly, first-hand, what can happen when code is poorly written.
In the next two sections, we will give an outline of both courses. This is intended
to serve as a model for those who may be interested in implementing a similar system.
Finally, the last section will conclude with commentary about impressions and
effectiveness of such a series of courses on students.

SECURE SOFTWARE DEVELOPMENT

This course, offered at the junior level focuses on identification (and subsequent
correction) of potentially vulnerable code in software, as it is being developed. It is
conducted in C on a UNIX operating system. The course addresses potential areas in
which security flaws can be introduced in any stage of the software engineering lifecycle.
It is a traditional lecture-type class supplemented with homework and quizzes.
The course begins with a strong emphasis on the motivations of such a course; why
the students are being expected to think about potential security flaws in code, and what
questions they should be asking themselves during the software engineering process. The
lectures then proceed to address secure coding principles, such as redundant levels of
defense, graceful (and secure) failure, and least privilege. Following this, topics such as
compartmentalization, avoidance of overly complex code and high coupling, privacy
issues (including, for example, clearing a buffer of data before reusing it and encryption),

102
 CCSC: Northeastern Conference

and finally, that input to a program should not be trusted, regardless of the source
(including user input as well as, for example, network packets).
After this overview, specific types of errors and avoidance are discussed. The first
error discussed is a buffer overrun; and specifically, we discuss safe versus unsafe C
functions, as well as how shellcode can be introduced (shellcode is the ability to implant
a command line interface into faulty code; it is a popular exploit, and is discussed in
detail in [5]). After a discussion of what a buffer overrun is, the material then shifts into
how to avoid them. Following on this, format string exploits (where the arguments to
printf() are incorrect, which can allow attackers to take control of a program, discussed
in [4]) are covered, however, this is covered more lightly, as an in-depth discussion takes
place in the next course.
The next topic expands on knowledge gained from our Operating Systems course,
and describes access control models in both Windows XP and UNIX. The importance of
topics such as understanding how setuid and setgid modifies ownership and what this
means to system security is discussed. Many examples are given to illustrate the
complexity of how these bits can change, depending on the operation.
Following this, we discuss race conditions. While this is also discussed normally in
Operating Systems, in this case, we look at how not understanding a race condition exists
can result in security failures of code (a number of exploits are actually based on race
conditions) For example, we discuss what can happen if temporary files are not properly
protected, as well as what should and should not be in a temporary file. We also
emphasize secure file access in this set of lectures. Finally, we discuss various auditing
tools and their strengths and weaknesses.
By the end of this course, we have introduced a number of important security faults,
and, using examples with code, have shown how to identify and mitigate them. However,
this only fulfills part of our objective; in addition to knowing what to avoid, we also wish
to demonstrate what can happen if these are not avoided. This is the goal of the next
course, Applied Security Concepts, which we now discuss.

APPLIED SECURITY CONCEPTS

The purpose of the Applied Security Concepts course, then, is exactly as the name
implies. However, instead of merely identifying vulnerabilities and repairing them, as was
done in Secure Software Development, in this course, the focus is more on the mind of
an attacker; specifically, what sorts of things do attackers do, how do they do it, and how
can it be defended against? This course introduces the students to real-world attacks, they
perform buffer overrun as well as format string attacks, insert shellcode, and analyze
real-world viruses and denial-of-service tools. The course is taught in C, and, as with
Secure Software Development, the prerequisites enforce this. We also require senior
standing to take this course; a certain amount of maturity is desired before students start
writing and examining real attacks.
One advantage of teaching this course at a liberal arts college is that the students
have typically taken ethics courses before taking this course, but we discuss ethical issues
in this course as well. We would also like to take this opportunity to explain why we feel
this is such an important course. While it is certainly possible to learn about

103
JCSC 26, 6 (June 2011)

vulnerabilities and attack techniques in theoretical form without examining any real
source code (as some suggest), this, we feel, hampers the students in a number of ways.
In general, the problem with not showing, in a laboratory setting, how attacks
actually work is twofold. The old adage of "Tell me and I will forget, show me and I may
remember, but involve me and I will understand" is the mantra of this course. Up to this
point, we have told and shown students, but not involved them. The reason we feel it is
so important for students to actually look at attacks and write buffer overruns themselves
is to give a feel for how difficult it is to take advantage of a vulnerability and break into
a system. Keeping track of all the information needed to accomplish a successful attack
is daunting. For example, in a format string attack where shellcode is inserted, one needs
to keep track of the current stack pointer address, know the top of the stack is, know the
offsets of promising addresses, know the address of the shellcode, know the byte-order
of the architecture, and finally, monitor how a %n directive changes as the format string
is built. This is a lot of information to juggle at once. While students can certainly
memorize "what do I need to know and how do I figure it out" (and, of course, some of
this is necessary and enforced by quizzes), it's much more meaningful to stand in front
of the classroom and say "O.K., we want to modify the value of X. What do we need to
do to accomplish this?" At this point, we're looking at actual code and trying to exploit
vulnerabilities within it, as an interactive session.
The other issue with not demonstrating how real attacks work is a secondary result:
the students learn to respect the people that they might be defending themselves against.
Once they understand the complexity of the task of exploiting software in a useful way,
the course starts to examine real-world Denial-of-Service tools, virii, worms, and so forth.
The students are required to analyze the source code and explain how a particular attack
works. And without fail, after this exercise, they are amazed at the skill of those who
write such attacks. They realize that these are not amateurs, but instead, highly skilled
programmers that should not be lightly dismissed. They learn that a single line of poorly
written code can compromise a computer, via examples where this exact situation has
happened.
It should be pointed out at this time that this course obviously needs a special
classroom laboratory. No one could benefit from our college accidentally releasing
viruses into the world. To this end, we have secured a designated laboratory for this
course. This classroom consists of a number of reclaimed PCs that are connected via an
air gapped network.
As a side note, the machines are all installed, configured, and maintained by the
students in the course, as well. When, for example, we get a new set of PCs (donated to
us instead of being destroyed during regular upgrades), the students are required to set
them up in our laboratory configuration (specifically, dual-boot XP and Linux). Students
are usually very eager to work on this project, as they get little opportunity to do this type
of network configuration and setup on such a large scale and level of detail. It also gives
them a nice set of hands-on skills. The machines all have two network cards so that we
can reconfigure the system as a gateway, attacker, or victim. We have two routers, one
devoted to connecting the Windows side of the machines, and the other connecting the
Linux side. We can, of course, easily interconnect as we wish, but this is our standard
configuration. We had students configure the router and determine the network

104
 CCSC: Northeastern Conference

connections to be made. All of this work gives them a sense of connectedness with the
lab; it isn't simply something they step into, but something they are an integral part of.
Further, it gives them a sense of satisfaction to know that the efforts they are making will
be used by future classes.
As for the course's content, we start with a focused refresher on C. Pointers tend to
be problematic for students and this course requires absolute mastery of them, for most
of what is done is manipulating memory. To that end, we spend the first two weeks or so
drilling the students on all aspects of pointers, including homework and quizzes to ensure
their knowledge. After this, we move into actual methods of exploitation, beginning with
buffer overruns (This is not a remote or isolated problem: in 2008, for example, buffer
overruns inside a game allowed wii users to boot the device from an SD card, which
previously had been impossible without modifying the hardware [5].) Our students
learned what a buffer overrun is in Secure Software Development and how to avoid them,
now they see what harmful impact this type of code error can have. The approach is
along the lines of showing the students some odd-looking errors, and speculating about
what they are seeing. Eventually, they are able to understand how variables can be
changed as well as return addresses and how shell code can be inserted via a simple
buffer overrun.
The course then moves into an examination of format string attacks. While not as
common as buffer overruns, these are interesting attacks to examine because they allow
complete manipulation of the source code (whereas a buffer overrun tends to simply
destroy stack data, a format string attack allows selective manipulation of the stack). The
technique of instruction is similar; the students examine what first appears to be
random-looking output, and eventually figure out that none of it is actually random but
instead reveals a great deal of information about the program. This period of instruction
culminates in the students attempting to gain privilege escalation by exploiting a piece
of software written specifically for this purpose and this course.
The class then moves into a second phase, which involves examining a number of
real-world exploits which use exactly the techniques that the students have learned. We
examine the attacks from several angles: history (what the attack did), mechanism (how
it works, including demonstrations), and defenses (how it can be detected and/or
stopped). The course then ends with the students themselves presenting the same type of
information within groups of two or three students. During this period, they have
complete access to the laboratory, and are free to experiment with different attacks and
defense methodologies on their own.

CONCLUSION
We feel that, with the continued expanding reliance on computing technology, it is
of paramount importance that not only do Computer Science students learn to program,
but learn to program correctly. As technology gets more pervasive, it is important to
realize that not only does a program have to work correctly, it has to resist attempts to
exploit it.
To that end, we feel every Computer Science student needs a background on basic
computer security. Although it is the case that not all of our graduates go into security

105
JCSC 26, 6 (June 2011)

fields, it is the case that all of our graduates will need to have this background knowledge
as the demand for programmers with these skills increases.
We have accomplished this at our institution by creating two new courses, which we
have described in some detail. It should be pointed out that it does not take an expert in
security to offer such a course, the references we have given, as well as some additional
outside reading, can allow any Computer Science instructor to teach such a course. At our
institution, these courses are now part of the required coursework. However, we initiated
these courses as electives, to get a feel for the difficulty and to solicit student responses.
Our students enjoyed the courses and have benefitted from them. It has happened
more than once that students (and former students) drop by and tell us how important this
knowledge is with respect to their jobs. While the introduction of the courses was a great
success, making them a required part of the curriculum has greatly benefitted our
students.

REFERENCES
[1] Nazario, J., Georgia DDOS attacks - a quick summary of observations. Arbor
Networks. August 12, 2008
http://asert.arbornetworks.com/2008/08/georgia-ddos-attacks-a-quick-summary-
of-observations/, retrieved November 19, 2010
[2] ACM, An interim revision of CS 2001: Report from the Interim Review Task
Force, 2008, www.acm.org/education/curricula/ComputerScience2008.pdf,
retrieved November 19, 2010
[3] Hanebutte, N., Oman, P., Software vulnerability mitigation as a proper subset of
software maintenance, Journal of Software Maintenance and Evolution,
Research and Practice, Vol 17., Issue 6, 379-400, 2005.
[4] Aleph One., Smashing the stack for fun and profit. Phrack Magazine 7, 49
(1996)
[5] Harrison, W. S., Hanebutte, N., Alves-Foss, J., Programming education in the era
of the Internet: a paradigm shift. Proceedings of the 39th Annual Hawaii
International Conference on System Sciences (HICSS '06), 2006
[6] Seacord, R., Secure Coding in C and C++, Upper Saddle River, NJ:
Addison-Wesley Publishing, 2006
[7] Murph, D., Wii Twilight hack could enable homebrew booting from SD cards,
2008,
http://www.engadget.com/2008/02/11/wii-twilight-hack-could-enable-homebrew
-booting-from-sd-cards/, retrieved November 19, 2010

106
 LEARN HOW TO USE EXECUTABLE FLOWCHARTS TO

ENHANCE LEARNING IN GENERAL EDUCATION, CS0, and

CS1 COURSES*

TUTORIAL PRESENTATION

Dee Gudmundsen Lisa Olivieri Namita Sarawagi

The College of Saint Rose Chestnut Hill College Rhode Island College
Albany, NY 12203 Philadelphia, PA 19118 Providence, RI 02908
518-454-2105 215-248-7092 401-456-9865
[email protected] [email protected] [email protected]

Visual Logic © (www.visuallogic.org) is an interactive, graphical software tool that

enables a novice programmer to develop interactive, executable flowcharts. Visual Logic
is extremely user-friendly, requires minimal syntax, and is very simple for a student to
learn. In this tutorial, we will present the ease, power and versatility of using Visual Logic
to teach programming concepts in three different courses at three different colleges. The
first course is a General Education CS course introducing algorithmic thinking. The
second is a CS0 course which uses Visual Logic for six weeks and then transitions to
Python. The third course is a CS1 course which uses Visual Logic as a tool to teach
programming concepts and to brainstorm solutions. Visual Logic is not taught separately,
but integrated with the various Java control structures.
The goal of this tutorial is to highlight the ways Visual Logic can be used to improve
student understanding of programming concepts and to enable the participants to easily
adopt Visual Logic in introductory courses. Participants will explore and experiment with
Visual Logic first-hand through supplied exercises. The presenters will provide
deliverables, such as sample programming projects and comparisons of Visual Logic
syntax to Python and Java syntax.

TUTORIAL OUTLINE
1. (10 minutes) Introduction to Visual Logic and the rationale for using it.
2. (20 minutes) Brief overview of how Visual Logic is used:

___________________________________________

*
Copyright is held by the author/owner.

107
JCSC 26, 6 (June 2011)

• in a General Education course with a varied student clientele. Students taking this
course lack prior programming experience. Visual Logic is the appropriate
programming language tool for these students to develop their algorithms and test
their logic due to its user-friendliness and (almost) syntax-free nature.
• in a CS0 course that teaches programming concepts using Visual Logic during the
first half of the course and transitions to Python. Similarities between Visual Logic
and Python such as lack of variable declarations, use of pre-defined functions, and
the interpreted nature of Python and Visual Logic will be examined.
• in an objects-late CS1 (Java) course that uses Visual Logic as a tool to illustrate
concepts and brainstorm solutions. Examples of lecture notes with Visual
Logic-created flowcharts and Visual Logic demonstrations of concepts before
introducing Java syntax will illustrate how the visual approach helps students grasp
concepts and reduces the frustration associated with syntax errors. The presenter
will also show the successful strategy of having students incrementally develop a
solution using the Visual Logic tool and then code the solution in Java by
associating standard building blocks of Java code with flowchart steps.
3. (45 minutes) Hands-on exercises illustrating major features of Visual Logic.

DELIVERABLES
Presenters will make the following available:
• A list of exercises from various disciplines correlated with programming concepts
in a General Education course.
• A tutorial PDF, Introduction to Programming Logic: Introducing Computer
Programming Concepts through Interactive Flowchart Development.
• A textbook appendix covering traditional flowcharting symbols and their meaning.
• Comparison of Visual Logic Syntax to Java Syntax.
• Materials that support student transition from Visual Logic to Python.
• Sample Visual Logic programs illustrating major concepts.

REQUIREMENTS
This tutorial will be offered in a computer classroom with Windows-based computers and
teacher station with large-screen projection capabilities. Participants should bring a USB
flash memory drive for installation of (non-registry- altering) software.

BIOGRAPHICAL INFORMATION ABOUT PRESENTERS

Dee Gudmundsen is an Assistant Professor of Computer Science. Dee has been teaching
CS1, using an objects-late approach, at The College of Saint Rose for 10 years. Since Fall
2009, Dee has been using Visual Logic in CS1 to introduce programming concepts before
introducing Java syntax and as a tool to brainstorm solutions to programming problems.
Lisa M. Olivieri, SSJ, PhD is an Associate Professor of Computer Science and
Technology and chair of the Computer Science and Information Technology Department.

108
 CCSC: Northeastern Conference

Having taught the introductory programming course for sixteen years, she has used
Visual Logic in conjunction with C++ or Python to teach this course for the past four
years.
Namita Sarawagi is an Assistant Professor in the Mathematics and Computer Science
Department. She has taught introductory programming courses for several years. She was
involved in the development of a new CS General Education course: "Introduction to
Algorithmic Thinking". She has presented a poster at CCSC-NE 2010 illustrating the
effectiveness of this tool in a course targeted for students of all majors.

109
`

STUDENT INTEREST AND CHOICE IN PROGRAMMING

ASSIGNMENTS*

Lisa Torrey
Computer Science Department
St. Lawrence University
Canton, NY 13617
315 229-5446
[email protected]

ABSTRACT
This paper describes a study, conducted in an introductory programming
course, on the factors that make students interested in programming
assignments. These factors include whether the assignment is perceived as
easy or difficult, and the paper analyzes these in detail. There are also
significant factors involving the end product of the assignment. The study also
looks at the impact of these interest factors on choice, when students must
actually choose a program to write. The same factors are involved, but some
become more important while others become correspondingly less important.
These observations may be useful for instructors considering the design of
assignments or the role of student choice in introductory programming courses.

INTRODUCTION
Introductory programming courses typically see a wide range of student aptitude.
Faced with this diversity, computer science educators may struggle to design
programming exercises. Easier assignments are trivial for high-aptitude students, but
harder assignments may not be effective learning experiences for low-aptitude students.
Ideally, each student would perform exercises at an individually appropriate level,
working up to more challenging levels over time. It is interesting to note that this is
exactly what modern video games get players to do. Education researcher James Gee

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

110
 CCSC: Northeastern Conference

suggests that these games encourage learning because they "operate at the outer edge of
a player's growing competence" [4].
Perhaps we could create a similar learning environment by designing exercises with
a range of difficulty and encouraging students to choose the ones that interest them. If
challenges just at the edge of our comfort zone are the most interesting, this framework
could be very effective. However, it is worth asking whether student interest in programs
really would follow this pattern.
I have conducted a small study on this topic at my undergraduate liberal arts
institution. This paper presents the results of that study, and attempts to answer the
following questions:
• Are my introductory programming students truly most interested in mild challenges?
• Other than the level of challenge, what other factors affect my students' interest in
programs?
• If I did give my students a choice of programs, what factors would most affect their
choice?

BACKGROUND
The impact of interest on learning is well established in psychological research.
Studies in this area most frequently focus on reading comprehension as the learning task,
but their findings may apply more generally. In one early example, Hidi [6] found that
interest in subject matter is correlated with motivation, and leads to higher attention,
persistence, and memory. Ainley et al. [1] suggest an explanation: interest increases the
emotional value of the reading, which encourages persistence. Edelson and Joseph [3]
also found that interest leads to increased effort and a higher mastery of skills.
One study on mathematical word problems may be more directly applicable to the
programming setting. Renninger et al. [7] found that interest in problem topics leads to
higher focus and tolerance for frustration. However, they also had a cautionary finding:
familiarity with a topic can lead to overconfidence in a problem solution.
Some educational research has also addressed the factors that promote student
interest in learning activities. Tobias [8] argues that prior knowledge about the topic of
a learning task is an important factor. Edelson and Joseph [3] suggest that the perceived
usefulness of an activity is also relevant.
There are a few studies on student interest specifically in the programming setting.
Hansen and Eddy [5] had CS2 students rank assignments by level of engagement and
frustration, and found that the two were correlated. They proposed a measure that they
call niftiness, which increases with engagement and decreases with frustration.
Cliburn and Miller [2] allowed their students to choose between three types of
assignments: traditional, game-based, and story-based. A large majority chose the game
assignments because they found games interesting. Story assignments were the least
popular, but only because students found them to be too open-ended.

111
JCSC 26, 6 (June 2011)

DATA COLLECTION
To answer the questions posed above, I conducted survey activities with students
in my introductory programming course. Participation was voluntary and earned students
a small amount of extra credit in the course. I conducted three different surveys, spread
evenly throughout the semester, to increase the opportunities for participation.
I was able to collect data from 14 students, exactly half of my class of 28. Of these,
8 participated once, 3 participated twice, and 3 participated all three times. Based on
their final grades, these students represented a good cross-section of the course: one
ranked first in the class, one ranked 22nd, and the rest fell in between.
The surveys each presented a list of 7 program descriptions. Each description
specified a program to be written in Python, using concepts recently covered in the
course. Table 1 shows some examples. All of the programs were somewhat short, to
keep the activity brief, since it was being conducted outside of class on a volunteer basis.
None of the programs were games, since preferences for those are already known [2], but
some did involve graphical displays. Graphics programs used a course library, a simple
wrapper for Tkinter that allowed drawing, animation, and mouse interaction.
Given this list of programs, each survey asked students to do the following:
• Give each program an interest rating: boring, neutral, or interesting.
• Give each program a difficulty rating: easy, moderate, or hard.
• Choose one program to write.
The surveys asked students to provide written explanations for their ratings and their
program choice. These questions were entirely open-ended, so that students were free to
mention any relevant factor.

Table 1: A few examples of program descriptions from surveys.

Write a program that inputs people's weights and prints whether the
total weight entered is safe for a 1000-pound capacity elevator.
Write a program that inputs a number and prints whether it is a
power of two. A power of two can be divided by two repeatedly, and
the result is always an even number, until it hits 1.
Write a program that draws animated circles moving randomly
across the screen.

DATA ANALYSIS
Student responses explaining their interest ratings contained several strong themes.
I extracted these six factors from their responses:
• The program involves graphics.
• The program has a useful practical application.
• The program is entertaining to use.
• The program connects to my major or hobby.
• The program looks challenging to write.
• The program looks easy to write.

112
 CCSC: Northeastern Conference

Several of these factors focus on the end product of the assignment. Many students
found programs that involved graphical displays to be more interesting than console
programs. Others appreciated when they could see "a point" to a program, by which they
meant either practical use or entertainment value. Connections to a major or hobby were
important for some students.
The other factors focus on the perceived difficulty of the program. Two distinct
preferences appeared here. Some students were more interested in programs that looked
challenging, while others favored programs that looked easy.
Table 2 shows how many students mentioned each of these factors at some point
during one of their surveys. Overall, end-product factors had substantially more impact
on student interest than challenge factors did. Those who did mention challenge factors
were approximately evenly split on whether they preferred easiness or difficulty.
When actually choosing a program to write, students explained their choices with
the same factors. However, they cited some factors more frequently than before, and
others less. Table 3 shows how many students mentioned each factor to explain a choice.

Table 2: How many of the 14 students mentioned each factor to explain an interest.
Interest factor Number of students
Graphics 10
Useful 9
Entertaining 8
Challenging 5
Easy 4
Connects 4

Table 3: How many of the 14 students mentioned each factor to explain a choice.
Interest factor Number of students
Easy 8
Useful 4
Entertaining 4
Challenging 2
Connects 2
Graphics 1

Figure 1 illustrates how factors influenced choice differently than interest. I

estimated the importance of a factor as its percentage of the total weight in Table 2 or 3.
In the shift from interest to choice, easiness went from being one of the least important
factors to one of the most, and graphics went from being one of the most important to one
of the least. The other factors remained relatively stable.

113
JCSC 26, 6 (June 2011)

Tables 4 and 5 summarize the difficulty and interest ratings that students gave to
programs. Overall, across all the programs, the most common difficulty rating was
moderate and the most common interest rating was interesting. However, the programs
students actually chose to write were most commonly rated easy for difficulty level and
interesting for interest level.

Table 4: How many times students assigned each difficulty rating to a program.
Difficulty rating All programs Chosen programs
Easy 29 11
Moderate 40 10
Hard 20 2

Table 5: How many times students assigned each interest rating to a program.
Interest rating All programs Chosen programs
Interesting 37 15
Neutral 32 6
Boring 20 2

It is more informative to look at difficulty and interest ratings jointly. Table 6

summarizes the rating pairs that students gave to programs. Easy programs were most
likely to be labeled boring, moderate programs neutral, and hard programs interesting.
Table 7 shows the rating pairs only on programs that students chose to write. Programs
labeled moderate-interesting were plentiful overall, so it is not surprising that students
chose them frequently. However, easy-interesting programs were less frequent, and
students appear to have chosen them disproportionately.

114
 CCSC: Northeastern Conference

Table 6: How many times students assigned each rating pair to a program.
Easy Moderate Hard
Interesting 7 18 12
Neutral 7 19 6
Boring 15 3 2

Table 7: How many times students assigned each rating pair to a chosen program.
Easy Moderate Hard
Interesting 6 7 2
Neutral 4 2 0
Boring 1 1 0

DISCUSSION
Some of the interest factors my students cited resonate with previous studies.
Edelson and Johnson found that usefulness is important in learning activities [3], and one
of the top interest factors for my students was the usefulness of a program. Tobias claims
that prior knowledge about a topic enhances interest [8], and some of my students were
interested in programs because of connections with their majors or hobbies. Cliburn and
Miller noted student preferences for games [2], and while none of the programs in this
study were games, my students cited two interest factors - graphics and entertainment
value - that are surely related.
The graphics factor had somewhat surprising effects: a strong impact on student
interest, but a weak impact on student choice. The easiness factor made the opposite
shift. This may indicate that students perceived graphics programs to be difficult, or that
graphics were a surface attraction that they were willing to forego in exchange for
expedience. Note that they were less willing to forego other factors, such as usefulness
and entertainment value.
Overall, my students rated harder programs more interesting than easier ones. This
lends some support to my conjecture that challenges just at the edge of our comfort zone
are the most interesting. However, when offered a choice, my students disproportionately
chose to write less challenging programs than their interest patterns had suggested. They
did not typically choose boring programs - just ones that were interesting for reasons
other than their challenge level. Note that these programs would score highly on Hansen
and Eddy's niftiness scale [5].
At the beginning of this study, I was imagining a learning environment in which
students chose among exercises of varying difficulty. Unfortunately, the study has
negative implications for this type of learning environment. Rather than choosing
programs at the optimal level of challenge, many students would choose easy programs
that happen to contain other interest factors.
Instead, I plan to ensure that my homework assignments focus on the three most
important interest factors for my students: graphics, usefulness, and entertainment value.

115
JCSC 26, 6 (June 2011)

Every program can include at least one of these factors, and most can contain more than
one. It may always be challenging to cope with a wide range of student aptitude, but by
focusing on these factors, I hope to increase motivation and persistence in all of my
students.

REFERENCES
[1] Ainley, M., Hidi, S., Berndorff, D., Interest, learning, and the psychological
processes that mediate their relationship, Journal of Educational Psychology,
94(3): 545-561, 2002.
[2] Cliburn, D., Miller, S., Games, stories, or something more traditional: the types
of assignments college students prefer, Proceedings of the Thirty-Ninth ACM
Technical Symposium on Computer Science Education (SIGSCE), 2008.
[3] Edelson, D., Joseph, D., The interest-driven learning design framework:
motivating learning through usefulness, Proceedings of the Sixth International
Conference of the Learning Sciences (ICLS), 2004.
[4] Gee, J., Good games, good teaching, Campus Connections, University of
Wisconsin-Madison School of Education, Winter 2004.
[5] Hansen, S., Eddy, E., Engagement and frustration in programming projects,
Proceedings of the Thirty-Eighth ACM Technical Symposium on Computer
Science Education (SIGSCE), 2007.
[6] Hidi, S., Interest and its contribution as a mental resource for learning, Review of
Educational Research, 60(4): 549-571, 1990.
[7] Renninger, K., Ewen, L., Lasher, A., Individual interests as context in expository
text and mathematical word problems, Learning and Instruction, 12(4): 467-490,
2002.
[8] Tobias, S., Interest, prior knowledge, and learning, Review of Educational
Research, 64(1): 37-54, 1994.

116
 A CONTENT-BASED IMAGE RETRIEVAL PROGRAMMING

ASSIGNMENTS FOR

INTRODUCTORY COMPUTER SCIENCE COURSES*

Michael Eckmann
Skidmore College
Mathematics and Computer Science Department
815 N Broadway
Saratoga Springs, NY 12866
518 580-5294
[email protected]

ABSTRACT
Introductory computer science courses cover certain content like arrays, loops,
sorting, etc. There is evidence that students enjoy working with media (e.g.
images, video, sound) and it keeps them interested in the course material [1].
This paper introduces ways in which core content covered in introductory
computer science courses is used in Content-Based Image Retrieval (CBIR)
systems. We also provide specific programming assignments that are intended
to be engaging to the students while requiring them to show their mastery of
the course content. This CBIR content and related programming assignments
are intended to achieve two goals. One goal is to increase student interest and
another goal is to introduce them to an active research area.

INTRODUCTION
Content-Based Image Retrieval (CBIR) systems allow users to search for images in
databases, often using an image (query) as input. The content of that query image is then
used to search the database to find similar images (results). This is in contrast to
searching images using word tags. CBIR systems often use the color content, the texture

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

117
JCSC 26, 6 (June 2011)

content, shapes, etc. to search for images similar to the query image. This paper focuses
only on the color content of the images.
Color images are stored as a grid of pixels, where each pixel is represented by a
number or set of numbers. Grayscale images (colloquially and less accurately
black-andwhite images) have each pixel represented by one number, usually of size 1
byte (range 0..255), where 0 represents black and 255 represents white and all the
numbers in between different levels of gray. Color images often store 3 bytes per pixel,
each byte representing one of the red, green and blue (RGB) components of the color. For
example a pixel whose 3 RGB byte values are 255, 0, 0 would represent a pure red color,
whereas 0, 128, 128 would represent teal and 135, 235, 135 represents a light green.
We taught an introductory computer science course to non-majors (a CS0-style
course) using Guzdial's "Introduction to Computing and Programming in Python, A
Multimedia Approach" [2]. It was useful in a CS0 course in our opinion. However, the
drawbacks to using this textbook or the Java version [3] of the textbook in a CS1 course
is that the approach uses classes that are not in the Java API. Instead, they are written by
and provided by the authors to make it easier on the students to utilize sound, images and
video in their programs. While it may make writing multimedia programs easier, having
students use non-standard libraries is undesirable in the author's opinion. The approach
described in this paper, at least for utilizing images, is a simple, intuitive approach using
only the standard classes in the Java API (without even using the Java Image I/O API)
that would typically be taught in a CS1 course. Other motivation for us was to find
interesting topics to which the students can apply the core introductory computer science
content (arrays, bit manipulation, sorting, etc.) There is much research in the literature
that students enjoy working with media (images in particular) ([1], preface of [2].)
Students also often have their own personal image collections that they can use as the
database of the CBIR system.
We introduce how to use a simple text based image file format and the steps
involved in creating a CBIR system that searches images based on color content. We also
divide the system into multiple programming assignments.

IMAGE FORMATS
There exist a variety of image formats. The one most often used today is jpeg. Jpeg
and other popular image formats (such as gif or tiff), store compressed images to save
space. PPM [5] is an easy to use, intuitive to understand image format which is part of the
Netpbm [4] open source software package. Utilizing this image format as opposed to
popular formats like jpeg is good for introductory computer science students due to (I)
a text editor can be used to view/create/edit ppm files where students can see the pixel
values in plain text, (ii) ppm files can be read/written into programs using the common
Java text file I/O classes (e.g. Scanner, BufferedReader, BufferedWriter) which are often
covered in introductory CS courses, (iii) jpeg is a less intuitive, compressed, often lossy
image format that cannot be viewed/created/edited using a simple text editors, nor
read/written in programs without using a Java jpeg package. Note: if file I/O is not
covered in the introductory courses, the professor can easily provide the code for reading
ppm files into a two dimensional array to the students to use in their programs.

118
 CCSC: Northeastern Conference

A simple PPM file is shown below. P3 specifies that this file is a text ppm file. The
# is a comment. The 6 6 line says it is 6 pixels by 6 pixels. The 255 is the maximum value
of a color component. 210 220 5 represents yellow and 0 0 200 is blue. These are
alternated to create a 6 by 6 checkerboard of alternating yellow and blue.
P3
# yellowbluecheckerboard.
ppm
6 6
255
210 220 5 0 0 200 210 220 5 0 0 200 210 220 5 0 0 200
0 0 200 210 220 5 0 0 200 210 220 5 0 0 200 210 220 5
210 220 5 0 0 200 210 220 5 0 0 200 210 220 5 0 0 200
0 0 200 210 220 5 0 0 200 210 220 5 0 0 200 210 220 5
210 220 5 0 0 200 210 220 5 0 0 200 210 220 5 0 0 200
0 0 200 210 220 5 0 0 200 210 220 5 0 0 200 210 220 5

COURSE CONTENT AND CBIR PROGRAMMING ASSIGNMENTS

Content-based Image Retrieval typically consists of the following steps.
1. Decide on an image representation (e.g. a color histogram, textures, shapes in the
image, etc.) and compute that representation for each of the images in a database.
2. Allow the user to provide an image (query) of which s/he wishes to find similar
images
3. Compute the query image's representation
4. Compare the query image representation to some/all of the representations in the
database
5. The results of the comparisons are a score for each compared image in the database.
Sort the scores associated with the images and provide the top scoring images from
the database to the user.
In this paper, we focus on color histogram as the image representation that is easy
to explain to students and can be implemented using the skills the students have been
taught in the introductory course. A color histogram is an image representation that does
not take into account any spatial information (it ignores where the various colors in the
image are) but instead only represents the count of how many pixels are of certain colors.
There are 256*256*256=16777216 different colors that can be represented using 3 bytes
for RGB color pixels. Since colors that have close RGB values look similar (e.g. 100,
211, 40 would look similar to 95, 215, 38) we do not want to keep track of so many
colors. Each bin will represent a different color. One decision for the bins would be to
separate the red values into 4 groups, the green values into 4 groups and the blue values
into 2 groups. This leads to 4*4*2 = 32 bins. Since each color component is represented
by one byte, to divide red into 4 groups we could choose 0..63, 64..127, 128..191,
192..255. We would divide green in the same way and we would divide blue into 0..127
vs. 128..255. Given the 3 byte pixel color, to determine the bin number associated with
it, one could use the bit shift operators on each of the color components ( >> 6 for the red
and green components and >> 7 for the blue component.)
For example, this would group all pixels that have a Red component of some
number between 0..63 AND have a Green component of some number between 128..191
AND have a Blue component of some number between 0..127 to be in the same bin and

119
JCSC 26, 6 (June 2011)

hence treated as the same color. For example, a pixel of color 5,129,120 would be
considered the same color (put in the same bin) as a pixel with 60,190,4.
bin 0: Red:0..63, Green:0..63, Blue:0..127
bin 1: Red:0..63, Green:0..63, Blue:128..255
bin 2: Red:0..63, Green:64..127, Blue:0..127
bin 3: Red:0..63, Green:64..127, Blue:128..255
...
bin 31: Red:192..255, Green:192..255, Blue:128..255
For each image, compute the number of pixels that fall into each bin and the color
histogram representation of the image is simply an array of those 32 numbers. The array
index represents the bin number, which in turn represents a color. The value stored in
each element of the array is the frequency of that color in the image. To be able to handle
images of different sizes, one should normalize the histogram to become a relative
histogram. That is, simply divide the number stored in each bin by the total number of
pixels (after this process, the sum of all the values in the 32 bins will be 1.) Then,
comparison between the histograms of images of different sizes is more meaningful.
One has many options in how to compare two histograms. One simple way is
Euclidean distance between the histograms. It is a simple like-bin distance measure that
results in a low number when the histograms are similar and a high number when the
histograms are dissimilar. The Euclidean distance is easily calculated with the following
code.
// assumes hist1 and hist2 are two arrays of doubles
double d, sum=0, euclideanDistance;
for (int i=0; i<hist1.length; i++)
{
d = hist1[i] hist2[i];
d *= d;
sum += d;
}
euclideanDistance = Math.sqrt(sum);
Other options students can explore would be to convert the RGB color values into
a different color space such as HSV or CIELab and experiment with dividing the colors
into different numbers of bins. Other options for comparing histograms include other
distance measures such as L1, or cross-bin types like Mahalanobis distance or Earth
Mover's distance. The cross-bin types allow one to specify how close or distant a color
in one bin is from a color in another bin (e.g. the color represented by bin 0 vs. the color
represented by bin 31 are very different, whereas the color represented by bin 0 and bin
2 are not as different).
The distances between the query image histogram and each of the database images'
histograms could be stored in a list of two columns in a text file, one column being the
database image file name and the other being the distance. Since those images whose
distance from the query image is small are expected to be those images that are most
similar (in color content) to the query image, one should sort the list from low to high
distance. The images whose file names associated with the lowest distances should then
be displayed to the user as the results of her/his query.

120
 CCSC: Northeastern Conference

The topics that the students will have been expected to be introduced to prior to
being given programming assignments for CBIR are as follows.
• Arrays, (e.g one dimensional arrays of doubles (for histogram representation), two
dimensional arrays of bytes (for grayscale image storage), two dimensional arrays
of a class type RGBColor (for color image storage).) Note: RGBColor should be a
class created either by the professor or students that store 3 bytes (one for red, one
for green and one for blue.)
• Loops to process the array data.
• Bit manipulation (bit shift operators) for determining bin numbers.
• Text file I/O for reading in a simple image format like ppm and writing/reading a
file of image names and distances.
• Sorting algorithm(s) to sort the distances

PROGRAMMING ASSIGNMENTS
Besides the core content that would be introduced ordinarily in an introductory
course, the amount of extra class time that is needed to prepare the students for these
programming projects is minimal. It includes:
• how images are stored as a grid of pixels and that each pixel contains a number (for
grayscale) or three numbers (for color RGB)
• the ppm file format
• what a histogram is (or whatever kind of representation you prefer to teach)
The time spent on this preparatory work, though, we believe is well spent as it will
convince the students of the usage of these topics in real world applications, introduce
them to an active research area (CBIR) and allow them to utilize their own personal
image collection for programming projects. Once they are introduced to this extra image
and CBIR related material, some programming projects that can be assigned are:
• Write a program that takes an image file name (ppm format) as input and computes
a color histogram and outputs it to a text file (adhering to a specified format.)
Deliverables: a ppm image file, source code for program and the text file output
containing the color histogram of the image.
• Write a program that reads in a histogram file name (representing the query image)
and compares it to all the histograms in a directory and outputs a text file list of
score/image file name pairs (adhering to a specified format.) Deliverables: a
histogram file, source code for program and the text file output containing the
score/image file name pairs.
• Write a program to read a text file containing score/image file name pairs and sort
it by score. Deliverables: a score/image name file, source code for program and the
sorted score/image name file.
Students who seem to take to this well may explore more than just "RGB color
histograms" as representations of the images. Different color spaces, such as CIE Lab and

121
JCSC 26, 6 (June 2011)

HSV are more perceptually relevant to humans than RGB. Besides color, they could try
to implement some interesting texture descriptions.
The students will typically have jpeg images that can be batch converted very
simply to ppm files using freely available software. The professor could provide a script
to do that as well as a script to run the image representation computation on a database
of images. If the students are not familiar with the code used to display images, the
professor could also provide a program that displays the results of the query — show
several images, given their file names.
Pedagogical materials such as lecture notes on images, the ppm file format, color
histograms and CBIR in general can be found at
www.skidmore.edu/~meckmann/CBIRfor-intro.html . Scripts to batch convert jpeg
images to ppms, as well as code to display multiple result images also can be found there.

CONCLUSIONS
In this paper, we outlined how CBIR can be used in an introduction to computer
science course to (i) introduce the students to an interesting application they can perform
on their own images, (ii) reinforce the usage of the core concepts taught including arrays,
loops, sorting, bit manipulation and text file I/O and (iii) introduce the students to an
active research area. We explained the general outline of most CBIR systems and
suggested how this could be assigned as several programming projects.
The students in an introductory course that utilizes this material will have learned
how and when to use arrays, loops, file I/O and sorting via interesting subject matter. The
additional topics that the students will have learned (PPM files, how images store colors
as numbers, RGB color space, etc.) will be a byproduct and provide them with new
knowledge of the inner workings of things they see every day (digital images.) The ideas
presented in this paper provide a way for students to not only learn the subject matter
(introductory computer science topics) but also learn the basics about an active research
area (there are current conferences on CBIR, and many CBIR papers are continually
published in the computer vision journals.)

REFERENCES
[1] Guzdial, M.and Forte, A., Design process for a non-majors computing course,
Proceedings of the 36th SIGCSE technical symposium on Computer science
education, 361-365, 2005.
[2] Guzdial, M. J., Introduction to Computing and Programming with Python: A
Multimedia Approach, Upper Saddle River, NJ, Prentice Hall, 2004.
[3] Guzdial, M. J., Ericson, B., Introduction to Computing and Programming with Java:
A Multimedia Approach, Upper Saddle River, NJ, Prentice Hall, 2006.
[4] NetPbm open source software, http://netpbm.sourceforge.net/doc/index.html,
February 22, 2009.

122
 CCSC: Northeastern Conference

[5] PPM Image format specification, http://netpbm.sourceforge.net/doc/ppm.html,

October 3, 2003.

123
 LANGUAGE CONSIDERATIONS IN THE FIRST YEAR CS

CURRICULUM*

Jonathan D. Blake
Department of Mathematics and Computer Science
CL-AC3, Quinnipiac University
275 Mt. Carmel Ave.
Hamden, CT 06518
Phone: 203-582-8539
Email: [email protected]

ABSTRACT
Computer science (CS) degree programs privilege programming as a
fundamental tool to teach algorithmic thinking. CS curricula start with an
introductory course sequence that focuses on learning how to program, and
often delay introducing material with a significant ontological component until
later. The importance placed on programming as a skill, coupled with its
position in the course sequence and the vacuum left by postponing identity
development has understandably resulted in a rousing debate about how to
teach these introductory courses. As such, the literature is filled with
passionate arguments for and against particular languages, processes, and
paradigms. This wealth of writing should not come as a surprise. The choices
made in the teaching of this sequence are very personal ones. I believe that we,
as faculty each support the teaching of programming in ways that support our
own identities as computer scientists, and our own definitions of the discipline.
Computer science is not programming, however, and computer scientists are
not programmers. I say this not to attempt to sever the ties between the two,
but to disagree categorically with the notion that the two can be directly
equated. I believe that we need to consider what we want students to learn
beyond the syntax of a programming language, and this paper is an attempt to
start that discussion.

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

124
 CCSC: Northeastern Conference

BACKGROUND
As a discipline, we have struggled (and ultimately failed) to agree on just what the
best way is to teach programming to novices [5, 6]. There are a variety of reasons why
we don't come to a consensus ranging from the influence of outsiders on the CS
curriculum to our own internal preferences and biases. Most CS faculty will agree that
learning to program is integral to the process of becoming a computer scientist [17, 19].
Interestingly, we tend to be equally in agreement that computer scientist is not equivalent
to programmer [2, 15].
Choosing an appropriate first language to teach then becomes a question of at best
attempting to reconcile the two (trying to create an identity as a computer scientist by
learning how to program), and at worst simply teaching the Next Big Thing™ that
industry-in the guise of the current job market for our students-is demanding of us.
For a long time, there was little controversy about the appropriate languages for
introductory teaching. Languages are designed to be effective in particular contexts, and
several were developed specifically to provide teachers with appropriate teaching
environments [10, 14, 16]. The notion of a "marketable" language, however, changed the
landscape. Introductory programming in CS went from PASCAL to C as systems
languages became more popular. When the notion of appropriate programming paradigms
entered the discussion [4, 5, 18] we rapidly moved to C++, and then Java. Throughout
this, functional programming (FP) adherents calmly continued to offer introductory
programming courses. They are a small but strong minority, and courses taught using FP
offer their own ontological component not addressed in this paper. Instead, this discussion
reflects on introductory programming courses in the imperative/object-oriented tradition.
I have spent a great deal of time over the past 12 months thinking about the
processes we employ when teaching introductory programming. My thinking has been
heavily influenced by several things, most notably work on Threshold Concepts [12] and
an invited editorial in SIGCSE Inroads about enrollments and programming [13]. In this
second paper, when Manaris speaks of "nonobvious ideas," he is speaking about student
engagement with, and understanding of, the material in an introductory programming
course, but in a very real sense he is also talking about the ability of a programming
language to "get out of the way" of student learning and understanding of core concepts
in the discipline (or at the very least, core concepts in programming and problem solving).
As such, I am convinced of two things: first, that we weren't wrong before when we
considered future course work-and even industry requirements-when we went through
the language selection process for the introductory programming sequence; and second,
that we weren't exactly right, either! We cannot lose sight of the learner in the teaching
process, and some of our choices, at least at the beginnings, need to take into account
student roles. Berglund and Lister [3] published an excellent paper on Kansanen's didactic
triangle [9] and the teaching of introductory programming that addresses just this need.
I struggled, however, with reconciling these two points. It seemed to me that they were
relatively orthogonal, and it would prove difficult to find a way to maximize both.

125
JCSC 26, 6 (June 2011)

DISCUSSION
The idea that students must master "nonobvious" ideas when learning to program
is a profound one. It is no coincidence that the examples in Manaris' editorial [13] are for
Java and Python. Java is a powerful but demanding programming language with a myriad
of (at times finicky) syntax requirements. Python is much smaller and more agile.
Additionally, it is interpreted (with an interactive shell), which lowers the barrier to first
success in programming for all students. Research has shown that assignments that
provide early feedback and early success (and that are compelling and visual) are
important in improving not just retention, but also gender equity [1, 11].
We taught our introductory sequence in Java for many years. Research has shown
no significant difference in performance when students are asked to switch languages
between CS1 and CS2 [7, 8], so starting in the Spring of 2008 we decided to investigate
various other languages for use in the first course in our programming sequence. I first
switched the language focus of our CS1 course to Ruby (we are retaining Java for CS2).
Since then I have taught Ruby twice more in CS1, followed by Groovy and now Scala.
What follows is an attempt to identify why the particular languages were chosen, and
provide anecdotal commentary on the process.
Ruby: When I first considered switching languages in CS1, I was struck with the
fact that scripting languages require the mastery of far fewer "nonobvious" topics, and
interpreted languages that provide us with an interactive shell lower the barrier to early
programming success. I figured (unsurprisingly) that I would be better off selecting a
language that I had more than a passing familiarity with. As such, I chose to offer the
course in Ruby. Ruby follows the principle of least astonishment, and I believed that this
would make it easier for students to develop problem-solving skills. Numerous interesting
powerful frameworks have been developed in Ruby. The language would clearly remain
relevant for students after the course, and student interest should remain high because the
language has recently seen a strong increase in its use, and has developed a lot of positive
buzz.
There was a definite downside to the choice of Ruby as a CS1 language. There is
no appropriate introductory textbook. While books and resources about programming-and
even learning to program-in Ruby are plentiful, they are not written to support the
concepts and processes covered in a CS1 course. The lack of appropriate materials,
however, was not the biggest reason why the choice of Ruby was not ideal. The students
were switched to Java in CS2, and this transition proved troublesome. Students reported
a great deal of dissatisfaction with Java because Ruby was easier for them to think and
program in. While expected, the depth of difficulty was a bit surprising. The students
persevered, but we should provide more to them than the opportunity to persevere. Three
iterations of the course did not significantly improve this transition, and measurable gains
in student retention were not realized. Even attempts at "pre-teaching" elements of Java
programming at the end of CS1 proved to be ineffective.
Groovy: A colleague taught CS1/CS2 during the 2009/2010 academic year, and he
reported similar problems when he used Python. When I next taught the course (during
the spring semester of 2010), I switched the language to Groovy in an attempt to
overcome these difficulties. I chose Groovy specifically because it was both a scripting
language promising to provide a similarly low barrier to programming, and a JVM

126
 CCSC: Northeastern Conference

language. As hoped, the switch to Groovy provided a considerably easier path to Java.
Groovy scripts can seamlessly interact with Java, and the students accepted the transition
to Java much more readily than with Ruby. Groovy was not without additional problems,
however. Specifically:
Groovy syntax is very different from Java syntax. Groovy idioms do not look like
Java idioms. Novice programmers think about problems and their solutions in relation to
simple idioms in the language they are learning. While not a significant boundary, the
switch to Java was made a bit more difficult because of this idiomatic shift.
Java is statically typed and Groovy is dynamically typed. Static typing turns out to
be very helpful when teaching programming. I have found that novices can grasp notions
of type faster when they must explicitly set a type that is enforced by the compiler.
Groovy has a wealth of control structures. Much like Ruby, Groovy has quite a few
control structures provided in an effort to support a variety of programmers' styles. While
in theory this could provide instructors with an opportunity to specifically target control
structures for schema development, this turned out to not be the case. Scala has very few
control structure and students were relatively quick to grasp the concepts.
Groovy does not respect private access modifiers. It is difficult to teach
object-oriented principles when the language doesn't respect access modifiers.
Interestingly, Groovy allows you to set access modifiers on entities. It just doesn't care.
I hesitate to call this issue a true problem with the language, however, because it provided
an ideal reason to shift the course focus to Java in CS1. Once the students recognized the
importance of hiding data, they were much more open for the switch to a language that
supports it.
Scala: This fall I am teaching CS1 again, and I made a conscious decision to attempt
to address all of the issues I have mentioned. I wanted a language that supported
programming idioms that were not clearly dissimilar to Java. I was looking for an
interpreted language that had an interactive shell for rapid prototyping and development.
I wanted a statically typed language. And finally, I was looking for a JVM language that
privileged Java interoperability as a way to facilitate the transition to Java in CS2.
In recent years a number of programming languages have been developed that
produce code that executes on the JVM. With a more nuanced list of requirements, one
language immediately moved to the top of the list. Scala has been steadily developed
since 2001, and represents a multi-paradigm language, combining functional and
object-oriented features. Scala supports an interactive shell, and Scala scripts can be
compiled into bytecode to seamlessly interoperate with Java. Additionally, the language
has very few control structures (although the list is complete).
While I will admit to some trepidation about the presence of functional
programming idioms-I was concerned that the students would see the presence of both
functional and object-oriented idioms as confusing-I was pleasantly surprised that they
were readily able to grasp functional techniques as they came up in the course. As they
mature as programmers, they should have no problem thinking in-and ultimately
programming in-the functional idiom.
Finally, the transition to Java at the end of the semester progressed smoothly.
Because of the seamless integration with Java, it was trivial to use Java libraries when

127
JCSC 26, 6 (June 2011)

building Scala applications (including the FANG game engine library), and due to the
similarities between Java idioms and Scala object-oriented idioms, I was able to
demonstrate programs in Scala and easily transition students to similar examples in Java.

CONCLUSION
The choice of language in teaching CS1 is a somewhat personal one. We each
approach the course differently, and thus teach with very different expectations. In this
paper I am reporting my personal approach to the course, and thus my
requirements/expectation for a programming language. I am compelled by the argument
that the syntax of an introductory language must minimize "nonobvious" topics. In
addition, I feel that introductory languages need to be used in upper-division computer
science courses, or at the very least ease the transition of students. Finally, I feel strongly
that statically typed languages are safer for novice students as they learn about data and
type.
For me, the choice of Scala is turning out to be quite a successful one. The retention
rate between CS1 and CS2 this year (as measured by registration for the spring semester)
is one of the highest we have recorded, and I can report anecdotally that the level of
discourse I developed with my students is much higher than in past semesters. I attribute
much of this apparent success to the switch to the Scala language.
The purpose of this paper is to hopefully foster discussion about programming
languages in CS1 and CS2 as a way of identifying and updating what we want students
to learn about programming. Interestingly, I feel that this exercise has served as a bit of
a Threshold Concept for me. Having gone through the process, I find myself thinking
quite differently about teaching and learning in introductory programming courses.

REFERENCES
[1] Barker, L.J., Garvein-Doxas, K. and Roberts, E. 2005. What can computer
science learn from a fine arts approach to teaching? SIGCSE '05: Proceedings of
the 36th SIGCSE technical symposium on Computer science education, 421-425.
[2] Beaubouef, T. and McDowell, P. 2008. Computer science: student myths and
misconceptions. Journal of Computing Sciences in Colleges 23, 43-48.
[3] Berglund, A. and Lister, R. 2010. Introductory programming and the didactic
triangle. ACE '10: Proceedings of the Twelfth Australasian Conference on
Computing Education 103, 35-44.
[4] Brilliant, S.S. and Wiseman, T.R. 1996. The first programming paradigm and
language dilemma. SIGCSE '96: Proceedings of the twenty-seventh SIGCSE
technical symposium on Computer science education, 338-342.
[5] Bruce, K.B. 2005. Controversy on how to teach CS 1: a discussion on the
SIGCSE-members mailing list. SIGCSE Bulletin 37, 29-34A.
[6] Dale, N. 2005. Content and emphasis in CS1. SIGCSE Bulletin 37, 69-73.

128
 CCSC: Northeastern Conference

[7] Doube, W. 2000. The impact on student performance of a change of language in

successive introductory computer programming subjects. ACSE '00: Proceedings
of the Australasian conference on Computing education, 71-78.
[8] Enbody, R.J., Punch, W.F. and McCullen, M. 2009. Python CS1 as preparation
for C++ CS2. SIGCSE '09: Proceedings of the 40th ACM technical symposium on
Computer science education, 116-120.
[9] Kansanen, P. and Meri, M. 1999. The didactic relation in the
teaching-studying-learning process. TNTEE Publications 2, 107-116.
[10] Lecarme, O. 1974. Structured programming, programming teaching and the
language Pascal. SIGPLAN Notices 9, 15-21.
[11] Leutenegger, S. and Edgington, J. 2007. A games first approach to teaching
introductory programming. SIGCSE '07: Proceedings of the 38th SIGCSE
technical symposium on Computer science education, 115-118.
[12] Lucas, U. AND Mladenovic, R. 2007. The potential of threshold concepts: an
emerging framework for educational research and practice. London Review of
Education 5, 237-248.
[13] Manaris, B. 2007. Dropping CS enrollments: or the emperor's new clothes?
SIGCSE Bulletin 39, 6-10.
[14] Pagan, F.G. 1980. Nested sublanguages of Algol 68 for teaching purposes.
SIGPLAN Notices 15, 72-81.
[15] Peckham, J., DiPippo, L., Reynolds, J., Paris, J., Monte, P. and Constantinidis, P.
2000. A first course in computer science: the discipline is more than
programming. CCSC '00: Proceedings of the fifth annual CCSC northeastern
conference in the journal of computing in small colleges, 223-230.
[16] Shultz, G. 2007. Using a restricted subset of Java in the first part of CS1. Journal
of Computing Sciences in Colleges 23, 212-218.
[17] Stroustrup, B. 2009. Programming in an undergraduate CS curriculum. WCCCE
'09: Proceedings of the 14th Western Canadian Conference on Computing
Education, 82-89.
[18] Vujoševi?-Jani?i?, M. and Toši?, D. 2008. The role of programming paradigms
in the first programming courses. Teaching of Mathematics 11, 63-83.
[19] Walker, H.M. 2010. The role of programming in introductory computing courses.
Inroads 1, 12-15.

129
 INTRODUCING BIOINFORMATICS INTO THE COMPUTER

SCIENCE CURRICULUM*

PANEL DISCUSSION

Mark LeBlanc, Professor of Computer Science, Wheaton College (Norton, MA)

David Rilett, Assistant Professor Computer Science, Wentworth Institute of
Technology, (Dobbs 303, 550 Huntington Ave., Boston, MA 02115), [email protected]
John Russo, Associate Professor Computer Science, Wentworth Institute of
Technology
Michael Werner, Professor of Computer Science, Wentworth Institute of Technology
Hongsheng Wu, Associate Professor Computer Science, Wentworth Institute of
Technology

With recent developments in genetics and molecular biology, and the impetus of the
human genome project, the analysis and interpretation of large genomic datasets is more
crucial than ever. This requires interdisciplinary expertise and the close collaboration of
biologists, statisticians, software engineers and computer scientists. In response to these
new challenges and opportunities, computer science students need to have a solid
understanding of the fundamentals of high performance computers and networks, basic
bioinformatics and its related algorithms, data mining and computational statistics. Open
source tools like Linux, scripting languages and R often make up the underlying enabling
science and technology.

POSITION STATEMENTS:
Mark D. LeBlanc: Modeling Interdisciplinary Science by Example
Our interdisciplinary course called "DNA" brings together biology, computer science, and
ethics, providing a fast-paced introduction to scripting and offering opportunities for
experimental design in genomics, while exploring the ethical aspects of living in a
post-genomic world and its challenges of "personalized medicine." Activities include: a
showing of the movie "GATTACA" and follow-up discussion, talks and discussions with
a genetic counselor and professor of bioethics, and student-produced YouTube
"commercials" of companies currently promoting genetic-based medical profiles.
See: http://genomics.wheatoncollege.edu

___________________________________________

*
Copyright is held by the author/owner.

130
 CCSC: Northeastern Conference

David Rilett: Introducing Bioinformatics to Undergraduate Programmers

The goal of our introductory course is to investigate bioinformatics methods in Perl and
to use common tools at NCBI to develop and validate projects. Translation and
transcription are coded. Large Fasta files are searched for various patterns after being
fetched by BioPerl and application pipelines are created. The goal is to spark interest in
the programming language of life and the tools of modern informatics.
See: http://biowit.cs.wit.edu
John P. Russo: Biological Data Mining for Computer Science Undergraduates
It can be difficult to teach biological data mining to undergraduates who may only be
required to take one database management course. One way to do this is to develop a
course that introduces data mining topics using bioinformatics data sets. Introductory
bioinformatics and a database management courses are prerequisites.
Michael Werner: Bioinformatics Algorithms
I am now teaching Bioinformatics Algorithms for the third time since 2007 using
procedural programming languages (C++, Java and C#) to implement and benchmark
algorithms in exact, approximate and multiple pattern matching, as well as sequence
alignment, clustering and hidden Markov models. Students learn the algorithms in
sufficient detail to implement them and benchmark them in several environments
including their own laptops, the school's 16-processor Sun server and in virtual clusters
constructed in Amazon's cloud environment. The algorithms are then re-examined to
improve their efficiency, looking for opportunities to parallelize algorithms in a
multi-processor environment and enhance performance.
Hongsheng Wu: Computational Statistics for Computer Science Undergraduates
The biostatistics course introduces students to basic statistical methodological concepts
and to the software packages that are central to computation and database operations.
Students with a background in biostatistics will be at a competitive advantage when
entering the work force or graduate school.
• Students explore several analytical techniques and determine the appropriate
analytical methods for data to be used to design experiments.
• Students learn to use of statistical software packages (such as SAS or R) to solve
biological problems.
• Students develop analytical tools to investigate simple biological systems.
• Students develop skills in presenting data in easily interpretable formats.

PANELIST BIOGRAPHIES:
Mark D. LeBlanc, Meneely Professor of Computer Science at Wheaton College and
co-leader of the Wheaton Genomics Research Group, has been teaching interdisciplinary
bioinformatics since 1998. NSF funding over the years has supported the development
of course materials for integrating bioinformatics in courses across the computer science
program. He co-authored Perl for Exploring DNA (Oxford, 2007) with his research and
teaching collaborator Betsey Dyer, Biology. [email protected]

131
JCSC 26, 6 (June 2011)

David Rilett, Assistant Professor of Computer Science, Wentworth Institute of

Technology, holds a Master's degree in Software Engineering and a bioinformatics
graduate certificate from Brandeis University.
John P. Russo, Associate Professor of Computer Science, Wentworth Institute of
Technology, has been actively involved in computer science education since 1999. He
has been actively involved in bioinformatics research, including collaboration on
modeling autism genetic networks with scientists at Harvard Medical School and Cold
Spring Harbor Laboratory. [email protected]
Michael Werner, Professor of Computer Science at Wentworth Institute of Technology.
His PhD in Computer Science is from Northeastern University. He also holds Master's
degrees from Boston University in Computer Information Systems and from University
of Illinois in Mathematics. [email protected]
Hongsheng Wu, Associate Professor of Computer Science at Wentworth Institute of
Technology, holds a PhD and M.A. in Biostatistics from Boston University, a M.S. in
Computer Science from Ohio University, and B.S. in chemistry from Shandong
University. [email protected]

132
 UNDERSTANDING NSF FUNDING OPPORTUNITIES*

TUTORIAL PRESENTATION

Sue Fitzgerald
National Science Foundation
4201 Wilson Blvd, Suite 835
Arlington, VA 22230
(703) 292-4641
[email protected]

ABSTRACT
This session highlights programs in the National Science Foundation (NSF) of
particular interest to computer science educators. Topics include a description
of program goals and guidelines, the review process and strategies for writing
competitive proposals.

INTRODUCTION
NSF supports projects to improve education in science, technology, engineering, and
mathematics through several programs in its Education and Human Resources (EHR)
directorate, as well as in its research directorates, including Computer and Information
Science and Engineering (CISE). This tutorial presents a description of some
education-related programs in the EHR and CISE directorates, and enables participants
to interact with the presenter concerning specific project ideas that could be appropriate
for the various programs.

SPECIFIC PROGRAMS DISCUSSED

Complete details about each of the following programs can be found on the NSF
websites for the Division of Undergraduate Education (DUE) [1] and the Directorate for
Computer & Information Science & Engineering (CISE) [2].
• Transforming Undergraduate Education in STEM (TUES)
• Federal Cyber Service: Scholarships for Service (SFS)
• Research Experiences for Undergraduates Sites (REU Sites)
• Scholarships in Science, Technology, Engineering and Mathematics (S-STEM)
• STEM Talent Expansion Program (STEP)
• Advanced Technological Education (ATE)

___________________________________________

*
Copyright is held by the author/owner.

133
JCSC 26, 6 (June 2011)

WRITING COMPETITIVE PROPOSALS

NSF programs are quite competitive but there are simple strategies that investigators
should be aware of to improve their chances of success. First and foremost, read the
Program Solicitation carefully. The goal is to help reviewers quickly understand what
you intend to do and that you have given sufficient thought of how you intend to do it.
Organize the proposal to address the essential components described in the solicitation.
Use headings, boldface and bulleted lists to help the reader quickly understand the
organization of the proposals. Address each point thoroughly but succinctly. And finally,
start well before the submission deadline. Include sufficient lead time to allow colleagues
to provide feedback and for your research office to approve the proposal. Specific
directions for completing a proposal can be found online [3].

REFERENCES
[1] NSF Division of Undergraduate Education (DUE),
http://nsf.gov/div/index.jsp?div=DUE, 2009.
[2] NSF Directorate for Computer & Information Science & Engineering (CISE),
http://nsf.gov/dir/index.jsp?org=CISE, 2009.
[3] NSF Funding, How to Prepare Your Proposal, http://nsf.gov/funding/preparing,
2009.

BIOGRAPHY
Sue Fitzgerald is serving a temporary assignment as a Program Director at the
National Science Foundation's Division of Undergraduate Education (DUE). She is a
Professor of Computer Science at Metropolitan State University, a former department
chair and past co-chair of the SIGCSE Technical Symposium. At NSF, she works on the
Transforming Undergraduate Education in STEM (formerly CCLI, now TUES), the
Federal Cyber Service: Scholarship for Service (SFS), and the Scholarships in Science,
Technology, Engineering, and Math (S-STEM) programs.

134
 PROCESS-ORIENTED GUIDED INQUIRY LEARNING (POGIL)

IN COMPUTER SCIENCE*

TUTORIAL PRESENTATION

Clif Kussmaul
Math & Computer Science Dept
Muhlenberg College
2400 Chew St, Allentown, PA 18104
484-664-3352
[email protected]

This tutorial will introduce participants to process-oriented guided inquiry learning

(POGIL) in computer science. POGIL has been developed, and validated over the last 15
years, primarily in chemistry education. In POGIL, teams of learners (typically 3-5) work
on scripted inquiry activities and investigations designed to help them construct their own
knowledge, often by modeling the original processes of discovery and research. Teams
follow processes with specific roles, steps, and reports that encourage individual
responsibility and meta-cognition. Multiple studies have examined the effectiveness of
POGIL, and generally find that POGIL significantly improves student performance,
particularly for average and below-average students.
The tutorial will be organized as follows. First, we will introduce ourselves and
briefly review some relevant background. Second, participants will work through a
sample POGIL activity to understand how it works. Third, we will review POGIL's key
concepts, history, and supporting research. Fourth, participants will begin to draft their
own POGIL activities to better understand the opportunities and challenges. If time
permits, we will review and discuss each other's activities. Fifth, we will conclude with
pointers to additional information and general discussion.

___________________________________________

*
Copyright is held by the author/owner.

135
 MOTIVATING STUDENTS TAKING CS1 BY USING IMAGE

MANIPULATION IN C AND C++*

Paul E. Dickson
Hampshire College
School of Cognitive Science
893 West St.
Amherst, MA 01002, USA
(413) 559-5861
[email protected]

ABSTRACT
Creating a CS1 course that teaches good programming techniques and
computer science concepts while also motivating students is a challenge faced
by all computer science educators. We have all seen that students who are
excited about their assignments make a greater effort and learn the concepts
embodied in those assignments better. Based on this experience we have
designed a CS1 curriculum that motivates students through assignments that
involve image manipulation and creation. For this course we have somewhat
reordered how concepts are typically taught in order to get students quickly
into image manipulation. The choice of which concept to introduce is
motivated by what it enables students to do to images. All concepts are
covered using C for the first three quarters of the semester and then revisited
during the final quarter using C++ to reinforce material and present
object-oriented concepts.

INTRODUCTION
At Hampshire College we have no set majors and a student body that tends towards
the arts. We therefore strive to offer CS1 courses that will provide a solid grounding to
students focusing on computer science while also attracting students who would not
usually take a computer science course. Bearing this in mind and also considering that

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

136
 CCSC: Northeastern Conference

students tend to learn more when excited and motivated by the subject matter, we set out
to design a CS1 curriculum that focuses on image manipulation and creation.
We chose image manipulation and creation because it gives students visual feedback
that lets them check the correctness of their projects by visual inspection and because
gives a tangible feel of accomplishment. The approach to image manipulation includes
both image processing and computer graphics, each used when it is the more appropriate
way to convey a new computer science concept.
The typical order that programming material taught is somewhat changed so that
each new concept enables students to progress with their image manipulation. For
example, students use libraries well before they understand specifics of how libraries
work and learn arrays and for loops before if statements. Reorderings of this type did not
appear to cause students any difficulty despite our concerns while designing the course.
The decision was made to teach this course using the C family of programming
languages for two reasons. First, our computer science students take classes at a number
of different colleges and these colleges do not use a common programming language. We
felt that background in C and C++ would put our students in a good position to quickly
learn any newer language required for the various courses. Second, we believed that by
having material introduced first with C and then with C++, our students would gain a
better appreciation for the differences between procedural and object-oriented
programming. In addition, revisiting programming concepts in the context of a new but
similar language would enable students to more firmly grasp the concepts.

RELATED WORK
Using image manipulation and graphics to help motivate students to learn concepts
in the computer science curriculum and in CS1 courses in particular is not new. It has
been introduced at every level from 9th grade [4] up to college database courses [6]. The
integration varies from a couple of image-related projects [7, 8] to full courses [3, 5].
Common to all of these courses is that the instructors found that integrating image-related
projects into the courses motivated students and helped with students' understanding of
material.
One approach taken was to give students access to prebuilt libraries to give them a
head start on image processing, which enables them to do more complex projects.
McAndrew and Venables [4] used an environment that did not require much
programming and primarily enabled students to run prewritten image processing
techniques that were based on computer vision. This was effective for the 9th and 10th
grade students in their course but is not as applicable to a full-semester CS1 course. Hunt
[3] also used prebuilt image libraries but focused on having students use and expand those
libraries. He had his students use Java's built-in image classes so that they could learn
how to interact with prewritten code. While this proved effective, we wanted to start from
a more rudimentary level and to build the larger framework embodied by Java's image
classes.
Another approach taken was not to base a full course on images but instead to just
have a couple of projects that included images. At Swarthmore College two projects were
added to the CS1 course: a greyscale image processing project and a color image

137
JCSC 26, 6 (June 2011)

processing project [8]. These projects enabled students to learn basic GUI development
while reinforcing arrays and other programming concepts. Students became fully engaged
with each project. At the College of Staten Island computer vision techniques were
introduced as a way of introducing research into a CS1 course [7]. Their results focused
on the effects of adding a research component to CS1. However, these projects used
computer vision techniques to scan brain images and some of the positive results may
have come from students' excitement over getting to use techniques on images.
Clemson University has taken the most aggressive approach to integrating images
into the computer science curriculum by bringing graphics into lower-level computer
science courses. Graphics have been used for both a CS1 course [5] and a database course
[6] as part of their Τέχνη project. For the CS1 course the approach is similar to ours using
C programs to create portable pixel map (PPM) format images. UNIX image tools are
used to convert images to more popular formats like JPEG and GIF so that students can
view what they created. This approach focuses on 4 major image projects and uses only
C; our course has 13 assignments that relate to images and also introduce C++ and a little
bit of web design. Our use of a greater number of assignments enables us to give smaller
assignments that enable a more gradual progression of skills.

COURSE ORGANIZATION
The goal of this course is to teach programming and the concepts included in a
standard CS1 course while keeping students engaged and excited about the material. Part
of this goal is to give students a fundamental understanding of programming that can be
applied to other languages and environments. With this in mind, all course material is
presented using the Emacs text editor within a command line environment. All
programming is demonstrated using the terminal built into OSX; students who used
Windows-based machines are encouraged to install Cygwin [2] so that they can also use
a command line interface. We believe that by learning this baseline style of programming,
students gain a better understanding of programming language vs. programming
environment and are better positioned to learn more supportive programming
environments in the future.
As with all programming courses this one begins with a "hello world" equivalent.
(The exact syllabus used for the course can be found at
http://helios.hampshire.edu/~pedcs/classes/cs110Spring10/syllabus.html.) The course
then quickly covers variables, for loops, and arrays. Immediately after this, on the sixth
day of class, the students are introduced to image libraries that enable them to read and
write PPM images. These libraries are given without explanation of code and students are
only shown how to use them. The students are told that before the end of the semester
they will understand how the libraries work. Students tended to be a little overwhelmed
when they are first asked to use these libraries but quickly move past that as they become
excited by being able to create their first images. The use of Makefiles eases the
introduction of libraries because students do not have to focus on how to include the
libraries in their code.
At the same time students are introduced to the image libraries, they are given a
basic introduction to web design and shown how to create webpages that contain images.

138
 CCSC: Northeastern Conference

This is done so that students can display their images as part of homework assignments
and more easily show the images to friends. Students have been excited by a computer
science course that enabled them to show off their work in ways more common to art
courses. Because students uploaded the web pages to our school's Linux-based student
webspace, creating and posting webpages also reinforced the command line material
covered.
Once students have these image libraries and have seen how to use for loops to
access every pixel in the image array, they are introduced to conditionals and if
statements. If statements are the logical next step as they enable students to select certain
pixels in an image that they wish to change. Conditionals with their Boolean logic follow
as students see how they can be applied to select the pixels they wish to manipulate.
From here the course proceeds to nested for loops. Students find this normally
difficult concept easier to grasp since it is far more logical to traverse an image array by
x and y coordinates than as a single dimensional array as they have been doing. With this
ability to traverse the image in two dimensions, students are then introduced to basic
filtering concepts like blur and haze, which are typically used with photography. Students
have found it exciting to better understand how filters in Photoshop actually work.
From here students are introduced to while loops that are introduced in the context
of applying image compression techniques (which keep running a process to decrease the
number of colors used until a certain number is reached). Students then have structures
explained and the Pixel structure they have been using all semester is put into context.
They are then introduced to functions in the context of how functions can simplify their
code and previous pieces of code are rewritten using functions.
We next address pointers, and this goes more smoothly than would normally be
expected since the images that have been used all semester have been referenced as
pointers in the code. This familiarity makes pointers easier to understand and gives
students a good starting point for seeing where pointers are useful.
The final areas covered in C are libraries, input, and output. The concept of libraries
is introduced and students are brought to understand how the libraries they have been
using all semester function. This leads to the concept of input and output, both user and
from the disk, and again the image libraries work as a map, showing students examples
of how to read and write from image files.
From here the course shifts to C++ and the concepts of object-oriented
programming. Students spend the end of the semester reimplementing their image
manipulation algorithms in C++ as part of a piece of image manipulation software.
Shifting code from procedural to object oriented sheds light on the differences between
the two ways of writing code. Students seem to find this section empowering since it
gives them greater insight into the similarities between syntax in programming languages
that they had heard so much about while also showing them just how much programming
they have learned.
The final assignment given is to build a text-based interface to their image
manipulation software written in C++. This program runs from the command line and
enables a user to load or create images, filter or modify images, and save images. At each
step the user is given a list of options that to perform on the images. All of the options and

139
JCSC 26, 6 (June 2011)

the interface are created by students. This assignment shows students that they had the
knowledge to build an entire piece of stand-alone software that might have application
beyond a single assignment or the class. This assignment gave students a great sense of
accomplishment.

COURSE STRUCTURE
The ordering of the material presented within this course is successful only because
of the structure and support built into the course. The course has 18 homework
assignments given over its 26 meetings. This means that homework assignments are due
twice a week about every other week (the course met twice a week). The constant
assignments mean that students are up to date with material presented in a previous class
and that they have worked on their own with that material before they learn the additional
material that builds on it (e.g., they have done an assignment with for loops before being
exposed to images).
These homework assignments are generally open ended, which encourages students
to experiment. A typical example is the assignment given after students learn if
statements. The assignment specifies that they create two images with code that included
if statements. We discovered that such assignments tended to bring out the best in
students since they tended to play with their assignments until they get interesting images,
spending a lot more time programming than they would have otherwise.
The other piece crucial to this course is the use of undergraduate TAs (UTAs). We
have had three UTAs who each attended class and held evening lab hours. They graded
one assignment per week and generally helped the course go more smoothly. A
description of our UTA program, its benefits, and our justification of it and using UTAs
to help ease grading burdens has been published [1].

CONCLUSIONS
The goal for this course was to build a CS1 course that would keep students excited by
the material throughout the semester and also appeal to students whose primary focus was
not computer science. The course has been a success: students learn how to program and
are engaged with the material. Part of the success is that students are given a chance to
build real software that creates output that they can show to their peers. Very often the
"coolest" thing a student creates during a CS1 course is an ASCII art pine tree. For this
course students created real images that they took pride in showing to others.
The final assignment of the semester was a great success and left students with a
sense of accomplishment. The act of building software that could be used by others and
included hundreds of lines of code left them feeling that they had built something
concrete. By rewriting all of their code so that it became part of this software, they not
only reinforced their understanding of material but also came to better understand why
commenting code is so important as they tried to understand the code they had written a
month before.
By the end of the semester all students had a decent grasp of Unix and enough
knowledge of web design to build basic websites. Seven of the 21 students in the course

140
 CCSC: Northeastern Conference

took computer science courses again the next semester. Because Hampshire College does
not have set majors, we do not have data on numbers of computer science majors.
Using image manipulation and creation as a basis for motivating students to learn
CS1 material appears to be a valid method for teaching CS1. The reordering of material
necessitated by this approach appears to have no ill effects on student understanding. We
highly recommend this curriculum because the levels of student motivation make it a
pleasure to teach.

REFERENCES
[1] Dickson, P., Using undergraduate teaching assistants in a small college
environment, SIGCSE '11: Proceedings of the 42nd SIGCSE Technical
Symposium on Computer Science Education, 2011.
[2] Cygwin developers, Cygwin Information and Installation,
http://www.cygwin.com/, retrieved November 15, 2010.
[3] Hunt, K., Using image processing to teach CS1 and CS2, Inroads-The SIGCSE
Bulletin, 35 (4), 86-89, 2003.
[4] McAndrew, A., Venables, A., A "secondary" look at digital image processing,
SIGCSE '05: Proceedings of the 36th SIGCSE Technical Symposium on Computer
Science Education, 337-341, 2005.
[5] Matzko, S., Davis, T., Teaching CS1 with graphics and C, ITICSE '06:
Proceedings of the 11th Annual SIGCSE Conference on Innovation and
Technology in Computer Science Education, 168-172, 2006.
[6] Matzko, S., Davis, T., A graphics-based approach to data structures, Inroads-The
SIGCSE Bulletin, 40 (3), 109-113, 2008.
[7] Sturm, D., Zelikovitz, S., Integrating research projects in CS1, Journal of
Computing in Small Colleges, 25 (6), 55-59, 2010.
[8] Wicentowski, R., Newhall, T., Using image processing projects to teach CS1
topics, SIGCSE '05: Proceedings of the 36th SIGCSE Technical Symposium on
Computer Science Education, 287-291, 2005.

141
 THE I-PHONE/I-PAD COURSE: A SMALL COLLEGE

PERSPECTIVE*

Lubomir Ivanov
Department of Computer Science, Iona College, New Rochelle, NY 10801
Tel.: 914-633-2342
Email: [email protected]

ABSTRACT
The paper details the experience of teaching an iPhone Application
Development course at a small, liberal-arts college. The course offers students
the opportunity to integrate the knowledge and skills acquired in other
Computer Science courses while working on exciting, real-world applications.
The paper discusses the benefits and challenges of introducing an iPhone
course, outlines the material covered in the course, describes some student
projects, and shares a few instructor observations.

1. INTRODUCTION
Apple's revolutionary iPhone device was introduced in 2007, and has since
established itself as one of the most remarkable developments in the field of personal
electronics. In 2008 Apple released the first version of the iPhone SDK, which allows
third-party developers to create applications for the iPhone. Soon after, several
universities began offering courses on iPhone applications development. Stanford
University was first, and its iPhone course (also available for free on iTunesU) became
widely popular. Other universities quickly followed with their own courses - MIT,
University of Maryland, University of Michigan, NYU. Additionally, the iPhone and the
newer iPad have found application in many traditional courses across the curriculum [1,
2]. Some interesting and unorthodox classroom uses of the iPhone/iPad devices have been
reported as well [3, 4].

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

142
 CCSC: Northeastern Conference

This paper presents the experience of offering an iPhone Applications Development

course at a small, liberal arts college. We describe the material covered in the course and
the projects assigned to students, share some observations arising from preparing and
teaching the course, and discuss some benefits and challenges that such a course offers.

2. THE iPhone COURSE AT IONA COLLEGE, NY

The iPhone course at Iona College, NY was offered for the first time during the
Spring'10 semester. It was taught as an upper-level undergraduate/mid-level graduate
course, meeting once a week for fifteen weeks. Each lecture was two hours long, and was
conducted in a lab setting. During class sessions each student had access to an iMac with
Xcode installed, and could experiment in real time with the software being presented.
The course objectives set forth for the iPhone Applications Development course
were:
- Focus on problem solving & software design principles while working on real
projects.
- Reinforce students' knowledge acquired in other CS courses by integrating elements
of database design, networking, computer graphics into the course, and
demonstrating how app performance is affected by the hardware and the iOS
operating system.
- Introduce the students to yet another programming language, Objective-C.
- Reinforce students' skills of using software modules and libraries.
The first three lectures familiarized students with the capabilities and limitations of
the iPhone device and the iPhone OS (later iOS) operating system. The Xcode
development environment was introduced, followed by an introduction to object-oriented
programming in Objective-C. Unlike other iPhone courses, we decided to avoid using
Apple's Interface Builder (IB) tool, and teach students how to create the necessary
program elements in code. There are multiple reasons for this: Using IB obscures some
significant aspects of software developments - we would like our students to understand
how and why things work, not simply learn to drag objects around the screen. We find
some aspects of IB to be poorly implemented and potentially confusing to students. Most
importantly, while IB is somewhat useful in static applications, it is of little use if
elements have to be dynamically created and rearranged as the program executes. As
students gained some familiarity with Objective-C, we introduced more advanced issues
- inheritance, composition, introspection, categories, protocols, etc. We also introduced
Apple's Foundations Framework, and discussed its fundamental structure and the way it
impacts iPhone/iPad software design. By lecture four, most students were relatively
comfortable with Objective-C and the Foundation Framework, so we introduced the
application design process. We discussed Apple's Model-View-Controller application
design approach, considered the basics of event handling and delegation, and introduced
the first widgets - labels, buttons, text fields, switches, sliders, and pickers. For their first
app project, students were asked to develop a scientific calculator using the design
principles and program elements discussed in class. Testing their apps for the first time
on actual iPhone devices was a significant point of excitement and pride for students.

143
JCSC 26, 6 (June 2011)

The next project built further on the excitement of the first apps - designing a simple
game. The lecture material introduced all the necessary elements: single- and multi-touch
handling, simple animation, and the NSTimer class. Students were given complete
freedom to choose what type of game to design. The only condition was that they employ
(multi-)touch handling, animation, and timers. The submissions included several shooting
gallery type games, a moon-lander game (Fig. 1), billiards (Fig. 2), and even a "Beavis
and Butthead" role playing game (Fig. 3).

The next lecture elaborated on more advanced aspects of game design - multi-view
handling and device accelerometers use. The lecture also provided an introduction to
Apple's Core Graphics library, used for creating 2D drawings and animation. Students
were asked to design a drawing using Core Graphics. Once again, student creativity was
revealed in the ambitious work most of them created. Below are just two examples - a
3D spheres composition (Fig. 4) and a Picasso-style abstract drawing (Fig. 5).

To complete the game design experience, students were introduced to sound and
video processing. The use the AVAudioPlayer, AVAudioRecorder, and MPMoviePlayer
classes was illustrated in class by the design of a sound player/recorder app and a video
player app. By the midterm, students had accumulated sufficient knowledge in different
aspects of game design to be able to create a more immersive game experience.
After the midterm, the focus of the course shifted to more traditional topics. An
introduction to tables and file processing was followed by a discussion of database
implementation on the iPhone. Apple provides a relatively simple framework, Core Data,
which transparently uses an SQLite backend for storing data. The use of Core Data was
illustrated with a Restaurant Menu example, which involved the use of view controllers
and a multilevel table. For their next project, the students were asked to implement a
database accessible through a multi-level table interface. Once again, a number of

144
 CCSC: Northeastern Conference

creative designs were submitted: Fig. 6 shows a "Birthdays" database. Fig.7 is a flight
database containing log information and procedures for operating a multi-engine airplane.

The remaining lectures focused on topics such as Locations Services and MapKit,
which allow the user's location to be determined through GPS, cell-tower triangulation,
or wi-fi, and plotted on a map. We also illustrated the use of the GameKit framework for
voice chat and text messaging. The last project for the semester asked students to develop
a "Find Me!" app which had to employ Bluetooth to determine the location of another
user running "Find Me!" on a different iPhone, plot the other user's location on a map,
and allow the users to communicate using voice chat and/or text messaging.
Throughout the semester, design correctness and efficiency were emphasized. We
considered the iPhone and iOS limitations - the small amount of application memory, the
limited multithreading support (prior to iPhone 4), the peculiar implementation of virtual
memory due to the flash-based secondary storage. We also focused on memory leaks -
students were required to screen their apps for leaks and eliminate any leaks found. In the
process, students became familiar with the Xcode-integrated Leaks tool. The last lecture
of the semester summarized everything that had been discussed about app performance,
and introduced students to a few additional instruments such as Activity Monitor and
Allocations. We also considered the process of submitting an application for distribution
through Apple's App Store. The final projects students were asked to submit had to
adhere closely to Apple's submission guidelines. Student were also expected to make a
public presentation of their work. The actual projects were left to student imagination,
and, once again, the students did not disappoint: As expected, most final projects were
games - a space shoot'm up game (Fig. 8), a "Super Mario Brothers" adventure game (Fig.
9), and a "Whack-a-Mole" game, complete with a sci-fi storyline.

145
JCSC 26, 6 (June 2011)

Other students chose different topics for their final projects: an NFL database (Fig.
10), a flight calculator (Fig.11), and a collaborative wireless/Bluetooth drawing app (Fig.
12).

Each student had ten minutes to present his work before his classmates, faculty, and
other students. The presentations gave students the opportunity to showcase their work
and practice public speaking on a technical topic - an important skill for a successful
career. Students were asked to evaluate each other's performance based on a number of
criteria: technical merits, project complexity, originality, user interface look-and-feel, and
presentation quality. The attending faculty were also asked to evaluate students according
to the same criteria. The results revealed that students were much more critical of each
other's designs (average score 29.8/50) compared to faculty (41.5/50). Faculty, however,
were more critical of student presentations (average 6.4/10 vs. 8.6/10 for students). This
last disparity is due, most likely, to students' lack of experience with public speaking.

3. BENEFITS AND CHALLENGES - AN INSTRUCTOR'S PERSPECTIVE

An iPhone course offers numerous benefits - for students, faculty, the CS
department, and the College. The level of student enthusiasm is unprecedented: For
many, this course represents everything they chose Computer Science for in the first
place. The main appeal, of course, is the opportunity to design iPhone-based games. More
mature students, however, move past games, and focus on other creative projects. In
either case, the ability to develop applications and see them ported to their own iPhones
is exciting for students. They know that learning iPhone app programming is a way to
earn extra income, so they readily devote significant time to the projects, unleashing their
creativity and imagination. For the instructor, seeing students put in such effort into their
work is a pleasant surprise. Since the dynamics of the classroom are quite different from
other CS courses, the instructor may opt to make changes in the way he/she presents the
course material and assigns projects, allowing students more freedom to choose projects
which are exciting to them and stimulate their original thinking. The instructor may
assign some projects for group work, which helps students learn to operate in a team
environment, teaching them leadership skills and collaborative strategies. Public
presentations of final projects helps develop students' public speaking and presentation
skills. Another aspect of the iPhone course which the instructor can take advantage of is
the breadth of topics related to the course subject matter. The material covered in the

146
 CCSC: Northeastern Conference

course relates to just about every other Computer Science course - Programming
Languages, Operating Systems, Architecture, Databases, Data Structures, Computer
Graphics, Networking, and Software Engineering. This allows the instructor to relate the
course topics to students' prior experience, revealing essential links between the
Computer Science sub-disciplines, while reinforcing students' knowledge in a hands-on
fashion. There also appears to be evidence that the iPhone course has been helpful to
students taking other courses: Some students enrolled in Database Systems concurrently
with the iPhone course improved their performance in the Database System course and
implemented some of their database projects on iPhone devices. In the Operating Systems
course, while discussing virtual memory, students brought up the iOS implementation as
an example of the dependence of the operating system design on the hardware.
For the CS department, the iPhone course expands the list of offerings, adding
cutting-edge material to the list of traditional courses. Student enthusiasm for the course
can lead to an increase in the number of students interested in Computer Science. At Iona,
a number of students have inquired as to when the course will be offered again, while
those who took the course have requested that a follow-up course be offered. Student
projects can be used at open houses to showcase student achievements and attract new
students. At a recent open house, a large number of prospective students expressed vivid
interest in the program as soon as they learned about the iPhone/iPad course and saw
some of student final projects. The addition of the iPhone course to the list of
departmental offerings can also benefit departments seeking ABET (re-)accreditation. It
demonstrates commitment to staying current and offering innovating courses on new
technologies.
Introducing an iPhone course at a small liberal arts college, however, presents a
number of challenges. Foremost is the investment in technology: The CS department
needs to provide a lab with Intel-based Macs on which students can carry out software
development. Testing the completed app on actual devices requires that the department
purchases several iPad, iPhone, or iPod Touch devices. The College needs to join Apple's
University program, which requires interaction between school officials and
representatives of Apple Corp.: One individual must be selected as the official College
contact person, and must be able to enter into binding agreements with Apple Corp.
Based on the experiences of other schools, we anticipated that joining the University
program may be a lengthy process. Our approval, however, was quick - less than three
weeks. Once the University program has been joined, an IT representative or a member
of the CS department needs to install and configure the Xcode environment and related
software on all machines to be used in the classroom. Access protocols are required by
Apple to ensure that only students enrolled in the iPhone course will be granted access
to the development systems. For us, this presented a significant challenge since the CS
department has access to three computer labs, but two of them - the ones with iMacs -
are public and open to any student on campus. The third lab, controlled by the CS
department, was equipped only with high-end PCs. Fortunately, the problem was resolved
before the beginning of the semester by the upgrade of the computers in the public labs,
which allowed several of the iMacs to be moved to the CS Department-controlled
advanced lab. Out of its own budget, the Department purchased six iPod Touch (gen.2)
devices which were made available to students enrolled in the course in the lab or on a
sign-out basis. The department intends to purchase several iPad devices before the next

147
JCSC 26, 6 (June 2011)

offering of the iPhone course. Another important issue which needs to be seriously
considered prior to the start of the course is that of intellectual property rights.
Accordingly, students must be informed at the beginning of the course what their rights
are with respect to any software developed as part of the course. For the instructor, the
main challenge is selecting an appropriate list of topics to provide a broad, solid
foundation for students to become familiar with software development on mobile devices.
The instructor needs to familiarize himself/herself with the intricacies of iPhone/iPad
software development, and be aware of potential pitfalls and issues that may arise during
the course. The instructor also has to contend with new developments in the
course-related material. For example, while the course was being offered, Apple
introduced the iPad devices, the iOS 4.0 operating system, and made significant changes
to Xcode. As a result, a number of topics had to be revised mid-way through the semester,
projects had to be adapted to the new hardware and software specifications, and the
instructor had to familiarize himself with the new features introduced in the latest
releases. Challenging as this is, it demonstrates most convincingly to students the
dynamic nature of the discipline of Computer Science and the necessity to remain
up-to-date in all the latest technologies.

4. CONCLUSION
In this paper we outlined the material presented to students in the iPhone
Applications Development course at Iona College, NY. We discussed several student
projects, considered a number of benefits offered by the introduction of the course into
the curriculum, and outlined some of the challenges faced by the instructor and the CS
department. The course offering has been perceived as a tremendous success by students
and faculty alike. We expect to offer the iPhone/iPad course again next semester. We are
also are preparing a follow-up course on Advanced Mobile Applications Development.

REFERENCES
[1] Grissom, S., iPhone application development across the curriculum, Journal of
Computing Sciences in Colleges, Volume 24 Issue 1, 10/08.
[2] McGill, M., Creating an iPhone game in a project course, CCSC-Eastern Conf.,
Villanova, PA, 10/09.
[3] Rogers, M., Honey, I shrunk the cluster! Parallel computing and Monte Carlo
simulations on iPod touches, iPhones and iPads, CCSC-Midwest, Franklin, IN,
9/10.
[4] O'Rourke, J. et al, Learning computer science concepts using iPhone
applications, Journal of Computing Sciences in Colleges, Volume 25 Issue 6,
6/10.

148
 SOUNDLIB: A MUSIC LIBRARY FOR A NOVICE JAVA

PROGRAMMER*

Viera K. Proulx
College of Computer and Information Science
Northeastern University
Boston, MA 02115
617-373-2225
[email protected]

ABSTRACT
We describe the design, pedagogy, and student's experiences with a library that
allows a novice Java programmer to design sound and musical accompaniment
for interactive graphics-based games, as well as explore the programming of
simple musical compositions, sound recordings, or visual representations of
music and sound. The library has been used for three semesters in our classes
and is publicly available at our website.
We wish to highlight two aspects of the library. First, the library explicitly
supports our test-first design approach to teaching object-oriented
programming. Second, the context of musical sounds: notes, pitches, duration,
instruments, how they create a melody, how they can be represented in a
number of different ways, presents a unique design playground for practicing
class-based design.

INTRODUCTION
Introductory programming is hard to teach. Our goal is to give the student an
opportunity to design a class-based system of non-trivial complexity, yet simple enough
to be manageable with only the basic programming skills the student has mastered. To
support this type of design exploration we have used for a number of years libraries
(named draw with variants idraw and adraw) that create a Canvas for simple drawing of
shapes, and a World class that handles time events and key events (the View and the

___________________________________________
*
Copyright © 2011 by the Consortium for Computing Sciences in Colleges. Permission to copy
without fee all or part of this material is granted provided that the copies are not made or
distributed for direct commercial advantage, the CCSC copyright notice and the title of the
publication and its date appear, and notice is given that copying is by permission of the
Consortium for Computing Sciences in Colleges. To copy otherwise, or to republish, requires a
fee and/or specific permission.

149
JCSC 26, 6 (June 2011)

Control), with students designing the game behavior (the Model). Student's game code
extends the World class by providing the onKeyEvent, onTick and draw methods and
it starts the game play by invoking the bigBang method.
Our curriculum enforces test-first design. Students are taught to design unit tests for
every method as soon as its signature and purpose are defined. The programming of the
game behavior supports this pedagogy. Students can define tests that check whether the
state of the world after a key event or a timer tick corresponds to the expected one.
This simple environment for designing complex games has served us well for a
while, but then it began to lose its lustre. In our first course students use a series of
functional languages with similar game libraries. (Indeed, these libraries have served as
models for our Java libraries.) As the supporting software and the first course evolved,
the games became more complex, the game libraries added sophisticated image
manipulation as well as a support for programming distributed client-server based games.
Coming into the second course that introduces class-based design in object-oriented
language (Java) the excitement for designing games with just a simple graphics waned.
The first game, designed in a mutation-free sub-language of Java was still a challenge.
When asked to design yet another game in a mutable style students would recycle old
games and retrofit them into the new style adding little to their class design repertoire.
Inspired by the work of Erich Neuwirth who uses programmatic music composition
(in a mini language for spreadsheets, and in Logo) for introducing computer science
concepts to beginners, we designed a sound library for Java. This library provides fresh
design opportunities, challenges, and possibilities — very different from those the
students have already seen. Additionally, the design of the library itself includes several
interesting case studies in class-based design that we plan to leverage in our course.
We present this tool and our experiences with it as follows. The next section
describes the design considerations for both the tunes package and the isworld package
and highlights the key features, including a deliberate design for testability. We then
illustrate the use of the library through our demo program, as well as through several
student projects, and conclude with acknowledgements.

LIBRARY DESIGN
The SoundLib library consists of the tunes and the isworld packages. The tunes
package implements the music/sound component and allows the programmer to com-
pose programmatically musical sequences (melodies, sounds) and play them. The isdraw
package extends the functionality of the original imperative idraw library by allowing the
programmer to play notes and other sounds on each tick or in response to a specific key
event. It also adds the methods for responding to mouse events.

The tunes package: Building the orchestra

The tunes package defines a MusicBox class that initializes a MIDI Synthesizer
with the default Soundbank and defines the initial MIDI program change selecting 16
instruments that provide a variety of options for the student. There are methods that allow
the programmer to see what is the current MIDI program (assignment of instruments to

150
 CCSC: Northeastern Conference

channels) and that allow the programmer to change the instrument assignments (the MIDI
program). Additional methods allow the programmer to start and stop playing one or
more Tunes, where a Tune represents a channel choice and a Chord (a collection of
Notes to be played or stopped). A simple sleepSome method that allows the given time
to pass before resuming completes this class. The programmer can play tunes by starting
and stopping a sequence of tunes with pauses in between.
To make the MIDI musical notation accessible to the programmer and to allow
students with minimal musical background to use the library, the interface
SoundConstants defines names for all MIDI instruments (e.g. PIANO, TUBA, or
BIRDTWEET), provides simple names to represent the notes in the middle of the piano
keyboard e.g. noteC, or noteDownG), and contains a mapping of MIDI instrument
numbers to their names. This makes it possible to define a Tune as simply as:
Tune pianoA = new Tune(PIANO, new Note(NoteA));
Once a synthesizer has been initialized and the assignment of the instruments to the
sixteen MIDI channel has been completed, the tunes to be played need to specify only the
note and the instrument on which the note should play.
When designing a musical component for a game, all that student needs to do is to
decide which notes or tunes should be played on each tick, or in response to the key
event. Adding the chosen notes or tunes to the appropriate TuneBucket plays the
selected melody. We carry the tunes in a bucket, so even the most musically challenged
programmer could carry a tune :).
The Note class allows the programmer to define any of the 128 MIDI pitches with
a selected duration in a number of different ways. The class is designed to accept a
number of different formats for defining the note, supporting different types of users in
a comprehensive manner. The programmer can specify just the pitch. The default duration
is one tick. Of course, students may not want to remember that 60 represents the middle
C, and so the note names defined in the SoundConstants interface provide an easy
way out. So, the middle C note of duration 1 may be defined in any of the following
ways:
Note midC = new Note(60);
Note midC = new Note(noteC);
Note midC = new Note(60, 1);
Note midC = new Note(noteC, 1);
Note midC = new Note("C4n1");
The last variant uses a String representation of the note, given by the note name
C, the octave 4, the modifier (one of n natural, s sharp, or f flat), and duration (1 tick).
The design of the constructors in this class provides a case study for designing classes for
user's convenience and for assuring data integrity. The class provides additional methods:
nextBeat and skipBeat that either decrease or increase the note duration. These
methods are used when a note is playing for the duration of several beats.
A Chord is a collection of Notes that are to be played at the same time. It can be
initialized in a constructor to a given sequence of Notes, or ints (pitches) or Strings
(note names). It can also be modified later by adding notes to the chord. This class also
includes the methods nextBeat and skipBeat that either decrease or increase the note

151
JCSC 26, 6 (June 2011)

duration for all notes in the Chord. But here we come with an interesting challenge.
When the programmer decides that a given Chord should be played by adding it to the
TuneBucket, she does not expect the Chord to change as the program plays the tune.
So, we need to make a deep copy of the Chord before it starts playing, to assure that the
mutation of the state as we progress through the beats has no ill effects. This provides a
nice example for learning about the meaning of deep copy and illustrating the need for
it.
The class Tune represents a chord that is to be played on the selected instrument.
Each tune specifies the instrument number from the MIDI program and the Chord that
should play. The class TuneBucket then contains 16 Tunes, one for each of the 16
instruments in the current MIDI program. The programmer can add to the TuneBucket
one note, one Chord, a Tune, a collection of Tunes, and also clear the contents of the
TuneBucket that stops playing all notes and removes all Tunes from the TuneBucket.
The methods nextBeat and skipBeat work in the same way as for a single Chord.
To represent a melody, student starts with a sequence of notes. For example, the
sequence of notes (noteC,0,noteD,0,noteE,0,noteC,0) represents the first
phrase of the Frere Jacques tune. The pitch 0 represents a silent note, a pause. The next
note to be played can be generated by a circular iterator, creating an infinite melody loop.
We can combine two traversals over the melody with the appropriate delay in the second
traversal to play a musical canon or we can play several instruments in parallel as an
orchestra would do.

The isworld package: Making games

The isdraw library provides a Canvas for drawing simple shapes (rectangles, lines,
circles, disks, and text) in any color and size. It defines an abstract class World that
handles the creation of the frame with the Canvas, the drawing of the representation of
the current world scene, the handling of the timer events, the key events, and mouse
events. The programmer designs a class that extends the World and implements the
methods that produce the changes in the World in response to the clock tick (onTick
method), in response to a key press and release (onKeyEvent, onKeyReleased), and a
method that draws the current state of the World. Optionally, the programmer can
override the stubs of methods that respond to mouse events.
To start playing a tune at a given time, the onTick method includes a command that
adds the selected Tunes to the tickTunes TuneBucket. The selected notes will play on the
given instrument with duration measured in clock ticks. The Tunes added to the keyTunes
TuneBucket play as long as the key is held down, regardless of the given duration. The
onKeyReleased method does not affect the notes played --- it is provided so the
programmer can take other actions when the key is released.

152
 CCSC: Northeastern Conference

Testing support
The SoundLib library has been deliberately designed to support the test-first
pedagogy. Every class in the tunes and isworld packages comes with methods that allow
the programmer to check the effects or outcomes of methods defined in that class. The
MusicBox class allows the programmer to check the current channel assignments to
instruments (getProgram(int channel) method), and to check which tunes are
currently playing provided by the nowPlaying method. The Note class includes a
method sameNote that checks whether this note represents the same note as the given
note. It verifies that the notes have matching pitch and duration (for example, the two
notes represented by the Strings "G4s2" and "A4f2" are considered the same). The
Tune class and the Chord class both allow us to check their size, and whether they
contain the given Note. The TuneBucket class allows us to check whether it contains
a given note played on the given instrument, and it reports the size of the TuneBucket.

THE DEMO PROGRAM AND STUDENT PROJECTS

To illustrate some of the ways the library can be used, and to help the students to
understand how to program music we have designed a demo program. The goal is to
show the multiple ways of how the musical ideas can be represented, and how the user
can observe and control the music that is played.

The right panel shows the current MIDI program: the assignment of instruments to
the 16 MIDI channels. The user can choose any one of the instruments via mouse click.
This way the students can hear how the different instruments sound. The displayed
keyboard on the top left provides a guide to the user who wants to play a piano. The
labels on the keys show the note that is played and the key press that plays the note. The
note that is currently played is highlighted. The bottom left panel shows a piano roll. The

153
JCSC 26, 6 (June 2011)

musical staff shows the notes that will be played on each tick. The arrow keys stop,
pause, reverse, and stop the playing of the selected tune. As the piano roll plays, the
currently played notes are shown in red.

Student experiences and what they taught us

The first time we used the library we had no duration for the notes, no chords, and
the instrument-note request had to be added to the TuneBucket one at a time. Yet
students came up with truly engaging jazzy tunes, using silence between the notes to
create the right timing. Before the second semester we improved the design with note
duration and a more complex structures for creating musical compositions (almost
everything described here except the mouse interactions and key press duration).
The final projects during the second semester were quite creative. In the frogger
game a jazzy tune plays in the background, each collision and when the frog reached the
other bank of the river produces sound effects, and a great final tune plays when the game
ends. A memory game asked the player to remember the sequence of square choices and
music that went with it. One student sonified the Game of Life. Another project was a
tool for composing music with playback (see below). The user could select one of four
instruments, and pick the pitch to play at the given time on a graphical board that
resembled the piano roll: the vertical dimension represented the pitch, the horizontal
direction represented the timeline. Pressing the start key played the composition
representing the current time by a thin horizontal line. This project made it clear that we
need to add mouse events to the library, as all pitch selections were done by just moving
the cursor using the arrow keys.

154
 CCSC: Northeastern Conference

Another student tried to capture the melody played by playing the keys like a piano
keyboard and playing it back on demand. His project motivated the re-design of the way
the key press and release is handled in the current version of the library. Overall, the
complexity and creativity exhibited in student projects confirmed the expected benefits
of providing this library.

ACKNOWLEDGMENTS
The author would like to thank Erich Neuwirth for his inspiration and
encouragement of exploration of programmatically generated music. His work on using
music with spreadsheets and Logo have influenced the library design and its pedagogy.

REFERENCES
[1] http://www.ccs.neu.edu/javalib/SoundLib.
[2] Neuwirth, E., http://sunsite.univie.ac.at/musicfun/MidiCSD/.
[3] Proulx, V.K.., Test-Driven Design for Introductory OO Programming, SIGCSE
Bulletin, 41(1), 2009.
[4] Sendova, E. (2001) Modelling Creative Processes in Abstract Art and Music,
Eurologo 2001, Proceedings of the 8th European Logo Conference 21-25 August,
Linz, Austria.

155
 UNTANGLING THE MAZE OF PEDAGOGICAL STRATEGIES

FOR INTRODUCTORY PROGRAMMING*

PANEL DISCUSSION

Frances Bailie 1
Associate Professor of Computer Science
Iona College
New Rochelle, NY 10801
(914)633-2335
[email protected]

Adel Abunawass Smiljana Petrovic Deborah Whitfield

Professor of Computer Assistant Professor of Professor of Computer
Science Computer Science Science
University of West Georgia Iona College Slippery Rock University
[email protected] [email protected] [email protected]

Many Computer Science Departments are continually searching for the ideal way
to introduce programming to novices: C++, Java, Python, Alice, Greenfoot, BlueJ,
objects first, objects last, objects never, and so on. By varying the language and the
approach, faculty seek the right combination to teach problem solving skills using
techniques that capture the students' attention, form the basis for future instruction and
encourage continuation in the major. The purpose of this panel will be to explore
strategies that have been implemented at three institutions in introductory programming
courses. Panelists will discuss the pros/cons of different techniques, offer their advice
and assessment of the techniques, and present a comparison of the different strategies.
Panelists will encourage participation from the audience, thus providing a forum for
sharing ideas and experiences in introductory programming.

1
Contact person

___________________________________________

*
Copyright is held by the author/owner.

156
 CCSC: Northeastern Conference

PANELISTS' POSITION STATEMENTS

Adel Abunawass
In our efforts to improve our retention, progression, and graduation rates we have
continually revised and evolved our introductory courses. In response to low retention
rates in CS1 and CS2, we drastically overhauled these two courses and added a new CS0
course. Our goal in creating the CS0 course was to prepare students to meet the
challenges of CS1 and CS2 and to provide non-majors with a gentler introduction to
computer science. In the process, we have utilized a number of languages and
environments including Alice, Java, Ruby, C#, Greenfoot, Eclipse, BlueJ, and Visual
Studio. Over the past eight years, our retention rates have improved in our introductory
classes, with variance over time. Currently we use an objects-first approach with Java
in our introductory courses with Greenfoot in CS0 and BlueJ in CS1 and CS2. Although
we have seen a definite improvement in our retention rates over the years, our efforts are
ongoing and we do not anticipate finding a "silver bullet." As various federal and state
STEM education initiatives bear fruit, we hope to see students who are better prepared
for the study of computer science as well as more women and minorities choosing
computer science as a field of study.

Frances Bailie
Faculty in the Department of Computer Science at Iona College have struggled to
attract students to the major, as have our colleagues in many other institutions. Of those
students who do enroll in our introductory course, many believe that the demands of the
course are too rigorous and eventually drop out. We started teaching Java over ten years
ago, both with an objects first approach (using BlueJ) and with an objects last approach.
Assessment results indicated that students did not truly understand objects and classes
under either approach, and so we kept Java but switched to a completely procedural
approach and postponed objects until the following course. This experiment showed
marginal improvement in retention and comprehension but assessment results indicated
that students were having difficulty not only with data structure concepts such as arrays
but also with the syntax of the language. In Fall 2010, we switched to Python in the
introductory course. We also moved to Python in the programming unit of our core
Computer Science course (having moved from Visual Basic to JavaScript previously).
Preliminary results of this last change are encouraging, especially from faculty. Students
appear to be more engaged and so we are hopeful that this move will improve student
comprehension and retention. Time will tell.

Smiljana Petrovic
Dr. Bailie will present here the departmental experience of Iona's Computer Science
Program; I can offer some observations from my personal experience. While Java was the
language of choice for the introductory programming courses at Iona College, one
attempt was teaching it using a multimedia approach. The motivation was to use a setting
that students can easily relate to and where their solutions are applicable to real life.
Image manipulation problems allow for some programming concepts to be naturally

157
JCSC 26, 6 (June 2011)

introduced. However, to avoid a narrow association of programming concepts with a

particular type of problem, and to provide upper level students from other science
departments with examples that relate to their majors, equal attention must be given to
non-multimedia problems. DrJava programming interface with its Interactions Pane
allows using Java as an interactive language, and works well with a multimedia approach.
Currently, our department is using Python in its introductory programming courses.
Python's clear syntax allows directing more attention to problem solving skills. I further
emphasize it by demanding a full solution of a problem that includes a list of relevant
test-cases, an algorithm, a working, commented code and runs on test cases that match
expected outputs. This procedure raises students' independence and confidence in their
solutions, which I find particularly important.

Deborah Whitfield
The Computer Science Department at Slippery Rock University found that student
retention was low in CS0 (introductory programming) and CS1 (advanced programming
with introduction to objects). Furthermore, assessment data showed that as soon as
conditionals were introduced in CS1 critical thinking and problem solving skills
decreased. In 2003, the department began experimenting with using Alice in CS0 and
Java in CS1. Experimentation included: 1) CS0 taught entirely in Alice, 2) ten of fifteen
weeks using Alice and then introducing Java, and 3) three to five weeks of Alice as an
introduction to Java. Although students taking CS0 for general education requirements
enjoyed Alice, the students majoring in a computing, math, or science disciplines found
Alice to be too childish. Using Alice attracted more non-majors to CS0 and increased
retention in CS0, but had no discernable impact on retention in CS1. For Fall 2010, the
department is taking a step backwards to C++ in CS0 and CS1, and Java in CS2. At the
same time, a course using Alice is being proposed for general education.

ABOUT THE PANELISTS

Dr. Adel Abunawass is a Professor and Chair of Computer Science at the University
of West Georgia. The Department of Computer Science offers a Bachelor of Science (BS)
in Computer Science (CS) and a Master of Science in Applied Computer Science
programs, as well as four graduate certificates in computing. The BS in CS program
received its initial accreditation by ABET in 2002. Dr. Abunawass has led the department
through the program's initial accreditation in 2002, and re-accreditation in 2008. He has
been instrumental in leading the assessment efforts in the department as well as in
designing the assessment system which uses open-source software tools to support the
development and analysis of course portfolios. Dr. Abunawass has published papers on
accreditation and assessment. He also serves as a program evaluator for ABET.
Dr. Frances Bailie is an Associate Professor, Assistant Chair and Chair of the
Assessment Committee in the Department of Computer Science at Iona College. She has
been teaching introductory programming for over thirty years using a wide variety of
languages. One of her research interests has been novice programmers and how they
learn. She has been a major contributor to the department's assessment efforts from our
initial ABET accreditation to our recent application for reaccreditation, taking

158
 CCSC: Northeastern Conference

responsibility for the collection and analysis of data. The use of assessment results for
program improvement has resulted in numerous changes in our approach to introductory
programming.
Dr. Smiljana Petrovic is an Assistant Professor at Iona College in the Department
of Computer Science, where she has been teaching a variety of courses, including
introductory programming courses in Java and Python. She taught introductory
programming courses in very different environments: at a college in Serbia, as an Adjunct
at CUNY, and AP and IB courses at the high school level.
Dr. Deborah Whitfield is a Professor and Chair of the Computer Science Department
at Slippery Rock University. She spearheaded the collection and analysis of the
assessment data that was used to justify the modifications to CS0 and CS1. She has taught
both CS0 and CS1 numerous times in her 20 years at Slippery Rock and has published
and presented at regional and national levels regarding assessment, accreditation, and
pedagogical topics. As a faculty member, she has assisted the department with creating
and revising student learning outcomes, developing and revising assessment methods,
analyzing data, and implementing a feedback loop that includes constituent feedback. As
a result of assessment, the department has made numerous changes to the degree program
including increasing the credit hours of CS1.

159
 SHARING YOUR INSTRUCTIONAL MATERIALS VIA

ENSEMBLE*

TUTORIAL PRESENTATION

Gregory W. Hislop Lillian N. Cassel

Drexel University Villanova University
[email protected] [email protected]

Lois M. L. Delcambre Edward A. Fox Richard Furuta

Portland State University Virginia Tech Texas A&M University
[email protected] [email protected] [email protected]

Peter Brusilovsky Daniel D. Garcia

University of Pittsburgh University of California, Berkley
[email protected] [email protected]

ABSTRACT
Ensemble is a Pathways project of the NSF NSDL (National Science Digital
Library) dedicated to supporting computing education. This tutorial will
introduce Ensemble and enable faculty members to take advantage of the
Ensemble facilities. The session will also provide an opportunity for the
CCSC community to provide feedback about the Ensemble facilities and
suggestions about the on-going development of the portal.

OVERVIEW
Ensemble is a portal for connecting computing educators. As a Pathways project
of the NSF NSDL (National Science Digital Library) program, Ensemble provides
collections, communities, and tools for computing educators. The collections consist of
freely available computing education resources stored within Ensemble and also provide
access to instructional materials stored at other locations. Ensemble provides federated
search, indexing, annotation, reviews, and other services to make these resources
accessible, visible, and more useful to the community. Ensemble also provides

___________________________________________

*
Copyright is held by the author/owner.

160
 CCSC: Northeastern Conference

information streams such as news, notices, and blogs of interest to computing educators.
Ensemble supports the full range of computing disciplines and also programs that blend
computing with other STEM areas (e.g., X-informatics and Computing + X).
Ensemble has been developed as a way to provide access and visibility to collections
of computing education material curated by a broad range of computing faculty and
groups. The success of Ensemble over the long term depends on the owners of these
collections choosing to make them visible through Ensemble.
This tutorial will introduce facilities that Ensemble provides to enable faculty
members to take advantage of the Ensemble facilities. The session will also provide an
opportunity for the CCSC community to provide feedback about the Ensemble facilities
and suggestions about the on-going development of the portal.

TUTORIAL CONTENT
The topics covered in the tutorial will include:
Overview - The session will begin with a short discussion about the history of NSDL and
Ensemble, other NSDL collections of interest to computing educators, and existing
collections of computing education resources.
Ensemble Collections - The coverage of Ensemble collections will highlight some
examples of collections of computing education resources within Ensemble and
collections that are not part of Ensemble but that can be accessed via the portal. The
emphasis for this discussion will be for participants to become familiar with Ensemble
services and facilities from the following perspectives
• Access to instructional materials - Ensemble provides facilities for locating and
evaluating useful instructional material. This includes a federated search across
collections, and ability to comment on resources, provide ratings, and tag items of
interest.
• Existing collections - Collections that are and will remain external to Ensemble can
become visible within the portal by connecting with Ensemble. The connections
may include being part of the federated search by using the Open Archives Initiative
Protocol for Harvesting Metadata. In addition, a collection may make use of
Ensemble features such as the instructional resource commenting, rating, and
tagging. This is accomplished by exposing these Ensemble facilities within the site
that hosts the collection. The tutorial will explain what is needed in order for an
existing collection to take advantage of these Ensemble facilities.
• New collections - The Ensemble team is also reaching out to faculty interested in
creating new collections of computing education instructional materials and sharing
them via the Web. The intention is to lower the barrier to entry so that faculty can
share small targeted collections that they create and maintain.
Communities - Ensemble provides facilities to create online communities to support
interaction among computing educators via facilities such as discussion forums, posting
of working papers, and connections to venues such as Twitter and Facebook. At present,
these services support open collaborations such as a CS1 community site and also hosts

161
JCSC 26, 6 (June 2011)

closed working spaces for groups like the ACM Education Board and the Future of
Computing Education Summit working groups. The tutorial will discuss some of the
open communities already started and explain facilities that may be of use to other
computing education efforts that would find this venue useful.
Tools - Ensemble also contains several tools that provide access to more advanced
facilities to help instructors access and organize materials relevant to computing
education. An example is Visual Knowledge Builder, which provides a workspace for
collecting and organizing computing education resources. The tutorial will briefly
introduce these tools.
Future direction - The Ensemble Computing Portal is still evolving, and the team is
interested in input from the CCSC community about future direction. The tutorial will
include time for participants to provide input and suggestions about features they would
find useful to support the community of computing educators.

ACKNOWLEDGEMENT
This material is based upon work supported by the National Science Foundation under
Grant Numbers 0534762, DUE-0840713, 0840719, 0840721, 0840668, 0840597,
0840715, 0511050, 0836940 and 0937863.

PRESENTER BIOGRAPHIES
Gregory W. Hislop holds a joint appointment in Information Science and Technology
and Computer Science at Drexel University. He has broad involvement in computing
education and online education. In addition to working on Ensemble, Dr. Hislop currently
directs NSF projects to re-design first year computing courses, to involve students in
humanitarian open-source projects, and to explore vertically integrated student teams.
Lillian (Boots) Cassel is a faculty member in the Department of Computer Science at
Villanova University. She is lead PI of the Ensemble project. Dr. Cassel is extensively
involved in computing education including being a past chair of ACM SIGCSE and
current member of the ACM Education Board.

162
 SHORT MOBILE GAME DEVELOPMENT PROJECTS FOR

CS1/2*

TUTORIAL PRESENTATION

Stan Kurkovsky (contact person) Delvin Defoe

Department of Computer Science Department of Computer Science &
Central Connecticut State University Software Engineering
1615 Stanley Street, New Britain, CT Rose-Hulman Institute of Technology
06050, USA 5500 Wabash Avenue, Terre Haute, IN
Tel: 1-860-832-2720 47803, USA
Fax: 1-860-832-2712 Tel: 1-812-877-8815
[email protected] Fax: 1-812-872-6060
[email protected]

Game development and mobile computing have been successfully used to increase
student motivation. However, instructors with no background in mobile computing,
computer graphics, and/or game development may find it difficult to develop or adopt
course materials on these topics. This workshop is designed to address these concerns.
Using Java Micro Edition, we have developed several project-based course modules
focused on mobile game development and designed to study fundamental programming
principles (e.g. loops) while also exposing students to more advanced concepts (e.g.
databases). Using a mobile phone emulator, participants will test-drive one of our
modules and develop a simple game, which can then be transferred to and played on a
mobile device. A Windows or Mac laptop is recommended.
Today, most college students have mobile phones and spend a considerable amount
of time using them for browsing the web, texting, or playing games. Introducing students
to mobile application development allows students to better relate to the course material
and make a stronger connection to real-world applications they see every day. Mobile
game development is by far less complex than traditional game development due to its
smaller scale and simpler graphics. Students are always interested in computer game
development; the course model presented in this workshop brings this topic to students
very early in the curriculum and serves as a good tool to increase student retention.
This workshop is intended for faculty members teaching college-level and high
school Java-based courses in CS who are looking to improve student motivation. All

___________________________________________

*
Copyright is held by the author/owner.

163
JCSC 26, 6 (June 2011)

other faculty members interested in integrating game development as a motivational tool

in their courses will also benefit from this workshop.
A number of free software packages will be used during the workshop. These
include Java Platform Micro Edition Software Development Kit 3.0, J2SE SDK version
1.6 or later, and Mappy Editor for tiled maps. The workshop participants will be provided
with with handouts describing how this software can be downloaded, installed, and used.
Participants will also have access to the source code, data, and relevant course materials
for the currently existing mobile game development modules. Workshop attendees will
also be invited to register on the project website that hosts all current versions of the
modules, along with a discussion board for instructors interested in or those who have
already adopted our course materials.

PRESENTERS' BIOGRAPHY:
Stan Kurkovsky is a Professor of Computer Science at Central Connecticut State
University (CCSU). His research interests are in the areas of mobile and pervasive
computing, software engineering, and computer science education. He has written over
60 peer-reviewed conference papers and journal articles, several of which discuss using
mobile game development in CS classrooms (including a SIGCSE-2009 paper). Dr.
Kurkovsky is currently serving as a PI on an NSF STEM scholarship grant providing
financial support for academically talented students majoring in Computer Science,
Mathematics, and Physics. Dr. Kurkovsky teaches undergraduate and graduate courses
in Software Engineering, Computer Networks, and Mobile Computing. Dr. Kurkovsky
has successfully taught the course in the focus of this workshop; students in this course
reported overwhelmingly positive experiences. Most recently, he participated in an
ITiCSE working group, which conducted a large-scale multinational study that addressed
many issues of attracting, engaging, and retaining CS students.
Delvin Defoe is an Assistant Professor of Computer Science and Software
Engineering at Rose-Hulman Institute of Technology. His research interests include
dynamic memory management, multicore computing, mobile game development, and
computer science education. He has published several articles on memory management,
a Nifty Assignment on modeling software engineering techniques to CS 1 students, and
looks forward to making substantial contribution in the other areas. Dr. Defoe is
currently involved in a program to motivate high school students to pursue undergraduate
studies in STEM disciplines. Dr. Defoe teaches undergraduate courses in Software
Development, Operating Systems, Computer Architecture, and the Theory and Practice
of Garbage Collection. He recently participated in a Summer School on Multicore
Programming in an effort to introduce parallel computing in the CSSE curriculum and
served on an NSF TUES 1 proposal review panel.
Drs. Kurkovsky and Defoe are currently serving as the PIs on a collaborative NSF
CCLI grant "Using Mobile Game Development to Improve Student Learning and
Satisfaction in Introductory Computer Science Courses." This workshop presents some
of the work stemming from this grant project, which has also been applied in CCSU and
Rose-Hulman courses.

164
 MOTION CAPTURE IN A CS CURRICULUM*

FACULTY POSTER

Bridget Baird and Ozgur Izmirli

Department of Computer Science, Center for Arts & Technology
Connecticut College
New London, CT 06320
860 439-2008
[email protected] and [email protected]

Motion capture is a field that presents many opportunities in a CS curriculum, both

for coursework and for research. As motion capture has developed and also become more
accessible, placing it in a CS curriculum has become more advantageous. Some of these
advantages include the possibility of writing interesting algorithms for manipulation, the
wide range of applications, the interdisciplinary nature of the field, the introduction of 3D
techniques and the "pop" culture relevance. Applications include character animation,
dance analysis and synthesis, avatar animation, movement analysis for athletes, gestural
control and study, and tools for persons with disabilities. In this abstract we talk about
some of the ways motion capture can be incorporated into courses and then detail some
of our own faculty/student research with dance.
Optical motion capture uses multiple cameras to record movement, which is then
transformed by software into 3D animation of the movement. Our own (fairly typical)
set-up has eight cameras placed in four corners of a space; each corner has two cameras
at different heights. A person wearing black clothing has approximately 32 sensors placed
on joints and various parts of the body. As the person moves the cameras record the
positions of all visible sensors. Upon completion of the movement the software produces
a 3D animation of a stick figure performing these same motions. Oftentimes this
animation has to be cleaned; for example, if there is much twisting and crossing of limbs
that contain sensors, the software can become confused about which sensor goes with
which location. Once the animation is complete, it can be manipulated, edited, placed
onto characters, and used in a variety of circumstances and for a variety of purposes.
There are several places where we have incorporated motion capture into our
courses. We teach a course in graphics and 3D environments (intermediate level) and we
now include a section on motion capture. Students are intrigued by using this technology
and are able to perform and to capture the movements. They learn how to manipulate the
motion sequences and then, with an introduction to programs such as Autodesk Maya or
3Ds Max, they can bring the motion capture sequences into these programs for further

___________________________________________

*
Copyright is held by the author/owner.

165
JCSC 26, 6 (June 2011)

manipulation and eventual use in 3D worlds. This adds an intriguing dimension of realism
to these worlds and forms part of the discussion of interface design and navigation
choices in virtual worlds. Another place where we are using motion capture is in our
animation workshops. Connecticut College has an interdisciplinary Center for Arts &
Technology and we teach an intensive workshop in 3D modeling and animation using
Maya. Often we also offer a more advanced workshop on the topic of character
animation. We are now incorporating motion capture into these workshops. A third
coursework area that we are exploring (but have not yet implemented) is in the
introductory CS course. Our introductory course uses Python and has a problem-solving,
interdisciplinary bent to it. The software that is used for manipulation of motion capture
sequences (Motion Builder) allows for the writing of algorithms in Python. We are
looking for some good examples where the students can do some interesting
programming that will manipulate the animations; this would be an intriguing application
of object-oriented programming and algorithm development.
In addition to inclusion in courses, we are doing some faculty/student research in
this area. Once again, for all the reasons stated above, this is an attractive area for student
research. It is also a field where extremely interesting questions can be examined. We are
currently working with members of the dance department to first develop a pedagogical
tool for some of their classes and then to provide a creative way to conceive of and study
choreography. Motion capture has been used in this field; for example see [1][2]. One of
the reasons that it is becoming so important is that the field of dance is so visual. Thus it
is difficult to describe with text either what one is seeing or what one would like to
produce. One method to document and explain dance has been through video, but that
often lacks the tools of analysis and precision that are needed for more universal
descriptions. Laban notation bridges this gap somewhat, but is a purely descriptive tool
that lacks the visual component. Thus motion capture, and a process that uses small
sequences of motion capture as building blocks, offers a fruitful approach. After extensive
consultation and collaboration with our colleagues in the dance department we decided
on a process that uses descriptive and visual components. Our motion capture research
begins with distilled and essential movements that take their roots in Laban notation. We
have captured dozens of these snippets using motion capture on dance students who
perform very specific kinds of gestures. The short sequences are then tagged with Laban
notation descriptors. The goal is for dance students in the classroom to be able to combine
these sequences in order to study, understand and choreograph longer dances that
illustrate specific concepts in dance. The algorithmic, computer science end of this
collaboration involves interface design and, most importantly, writing code that will
combine discrete sequences of movements in a smooth manner that also respects the
underlying dance concepts, producing a very valuable pedagogical tool. Upon completion
of this part of the project (during this spring semester) we intend to then move to an
analysis and synthesis of dance using motion capture data. All phases of this research
have involved students and it has been a truly cross-disciplinary experience that
demonstrates the relevancy and versatility of computer science in the undergraduate
curriculum.

166
 CCSC: Northeastern Conference

The authors would like to acknowledge our computer science student Ajjen Joshi, our
colleagues David Dorfman and Lisa Race in the dance department and many dance
students for their participation in this research.

REFERENCES
[1] Nagata, N., Okumoto, K., Iwai, D., Toro, F. and Inokuchi, S. "Analysis and
Synthesis of Latin Dance using Motion Capture Data," In K. Aizawa, et al.
(Eds.), Advances in Multimedia Information Processing-PCM2004, Lect. Notes
Comput. Sci. 3333, 39--44 2004.
[2] Zhao, L. "Synthesis and Acquisition of Laban Movement Analysis Qualitative
Parameters for Communicative Gestures," Ph.D. Thesis, University of
Pennsylvania, 2001.

167
 COMPUTER SCIENCE EDUCATION IN SECONDARY

SCHOOLS IN WESTERN NEW YORK*

FACULTY POSTER

Sarbani Banerjee
Buffalo State College
Buffalo, NY 14222
(716) 878 4912
[email protected]

This research investigates the status of computer science (CS) education in the
secondary schools of Western New York (WNY). Nationally, enrollment in computer
related majors is declining in secondary schools. Prior studies show that (1) most state
education departments do not certify (CS) as an academic area in K-12, (2) K-12
teachers who teach computer courses are most commonly trained in other disciplines, and
(3) too few opportunities exist for professional development for such teachers. The
primary purpose of the current study is to design professional development opportunities
for secondary teachers teaching CS or related courses. Thus, secondary school teachers
teaching CS or computer related courses will constitute the target sample for this
research. Data about the current status of CS education in high schools will be collected
from the teachers through an online survey. The survey has 3 sections: (1) demographic
questions related to the teachers and their schools; (2) types of CS or related courses
taught in the schools; and (3) self-identified areas of need and interest for professional
development among the teachers. The survey is comprised of a combination of items we
deem relevant to the study and questions derived from the pertinent literature [1,2] on this
topic. An initial survey will serve as a pilot study to test receptiveness of teacher to
participate in the online survey, as well as the suitability of the questions. Soon thereafter,
the survey will be administered to teachers at all high schools in the WNY area. Results
from analysis of this data will be available for and featured at the conference. The
primary significance of the results of this study lies in the analysis of the current status
of Computing Education in WNY High Schools; result will help inform the design and
development of professional development opportunities to benefit WNY high school
computing teachers and provide information to help broaden participation and diversity
in computing fields. Future research will include a more comprehensive survey of
computer teachers across the entire state of New York.

___________________________________________

*
Copyright is held by the author/owner.

168
 CCSC: Northeastern Conference

REFERENCES
[1] Blum, L., & Cortina, T. CS4HS: An Outreach Program for High School CS
Teachers. SIGCSE '07, March 7-10, Covington, Kentucky, USA, 19-23, 2007.
[2] CSTA. CSTA National Secondary Computer Science Survey: Comparison of 2005,
2007 and 2009 National Secondary Computer Science Surveys,
http://csta.acm.org/Research/sub/Projects/ResearchFiles/CSTASurvey05-07_09C
omp_DCarter.pdf, retrieved December, 2010.

169
 HOW COMPUTER SCIENCE DEVELOPS MATHEMATICAL

SKILLS*

FACULTY POSTER

Reneta P. Barneva Boris Brimkov

Department of Computer and Information Sciences Department of Mathematics
State University of New York at Fredonia SUNY University at Buffalo
Fredonia, NY, 14063 Buffalo, NY 14260
(716) 673 4750 [email protected] [email protected]

Many colleges and universities have programs of general education - where courses
are grouped into different categories and students have to select at least one course from
each category in order to get a well-rounded education. These categories often include
subjects like language studies, art, history, and mathematics, but very rarely is there a
category in which computer science courses are included. In this work, we argue that
computer science courses develop mathematical skills no worse than mathematics
courses, and can therefore be included in the mathematics category of general education.
We have assessment data which shows that certain computer science courses developed
mathematical skills to an even higher extent than introductory mathematical courses such
as statistics and pre-calculus. Our investigations can be used when computer science
courses are proposed for the mathematics category of the general education requirements;
we have not found any previous research or papers supporting this fact.
Computer science not only requires a mathematical background, but actually
employs and develops mathematical models, methods, tools, denotations, reasoning, and
writing. It is listed as a mathematical category by the American Mathematical Society.
Computer science is interrelated with branches of mathematics such as number theory,
numerical analysis, logic, geometry, discrete mathematics, statistics, and many others.
We should note that the notion "computer science" is deceiving; it is perceived that
computers play a central role in the field. However, there is much more to computer
science than just computers; as one of the founders of computer science, Donald Knuth
says "If I had a chance to vote for the name of my own discipline, I would choose to call
it algorithmics." [1] Informally speaking, an algorithm is an effective method for solving
a problem expressed as a finite sequence of steps. Examples of algorithms studied in
general college mathematics up to Calculus II and linear algebra are the algorithms for

___________________________________________

*
Copyright is held by the author/owner.

170
 CCSC: Northeastern Conference

addition, subtraction, multiplication and division of multi-digit numbers, solving linear

and quadratic equations and inequalities, calculating determinants, finding the derivative
of a function, integrating, finding the local maxima and minima of a function, performing
arithmetic operations on complex numbers, finding the transpose and inverse matrix, etc.
In computer science courses, problem solving requires more creativity than in
mathematics courses, because the instructor does not formally describe the algorithms,
but rather leaves this task to the students. Thus, students first have to formulate an
algorithm, then describe it, and finally plug in several sets of numbers to see if it gives
the desired output. In this process, special attention is paid to the limitations of the
algorithm and all "special cases" are carefully elaborated - something that in mathematics
is usually skipped under the term of "exceptions." While a computer quickly calculates
the result, the student would solve the problem on paper according to the devised
algorithm and compare that answer with the computer's answer to see whether the
computer solved the problem correctly. When the answers differ, the student would revise
the algorithm and repeat these steps. Therefore, computer science courses generally
develop problem-solving skills and creativity much better than mathematics courses. The
student actually acts as an instructor for the computer, and teaches it how to solve
problems. This process not only requires a mathematical background and implementation
of quantitative methods from arithmetic, algebra, or geometry, but actually entails the
formulation of original mathematical models and rules, representing them symbolically,
and drawing inferences. Most importantly, the limits of mathematical methods are
identified - something that is usually omitted in traditional mathematics courses.
In 2010, at SUNY Fredonia, certain introductory computer science courses were
assessed together with other courses in the Mathematics category. Out of 1045 students,
a random sample of 209 was selected. For each student, 3-5 assignments were examined.
The courses were split into 4 groups: Pre-calculus, Calculus, Statistics, and Computer
Science, and the results were generalized based on expected learning outcomes for that
category of general education. The results show that all the learning outcomes were met
to a higher extent in computer science courses than in the other courses in this category.
Specifically, the difference for the outcome of recognizing the limits of mathematical and
statistical methods is 17.3%; furthermore, the outcome of estimating and checking
mathematical results for reasonableness is met completely (100%) [2].
Computer science courses can be included in the mathematics categories of general
education program, because they not only satisfy the learning outcomes of mathematics
courses, but in some cases meet them to a higher extent. As further work, we are planning
to accumulate more assessment data from other colleges and institutions, and perform a
thorough review of the subject.
We would like to thank Dr. Mary Carney from SUNY Fredonia for providing us
with assessment data. This work is supported in part by the Cooperative Research Project
of the Research Institute of Electronics, RIE, Shizuoka University, Japan and by NSF
Grant No 0802994.

171
JCSC 26, 6 (June 2011)

REFERENCES
[1] Donald E. Knuth. Algorithmic Thinking and Mathematical Thinking. In The
American Mathematical Monthly, Vol. 92, No. 3, pages 170-181, March 1985.
[2] Nancy Boynton. Spring 2010 Assessment Results and Recommendations.
September 30, 2010, SUNY Fredonia.

172
 ROBOTICS, AI AND VISION IN JAVA WITH THE DLIFE

LIBRARY*

FACULTY POSTER

Grant Braught
Department of Mathematics and Computer Science
Dickinson College
Carlisle, PA 17013
(717) 245-1401
[email protected]

dLife is a Java library developed for teaching and research in robotics, artificial
intelligence and computer vision. While a significant number of platforms now exist for
using robotics in the CS classroom (e.g. Lego, Pyro, Myro, MS Robotics Studio, Player,
Tekkotsu, Webots, and others…) dLife presents a unique combination of features that
will be attractive to some educators. dLife allows students to program robots in Java, is
free and open-source, runs on Mac OX and Linux and supports a variety of popular
educational robots both physically and in simulation.
The centerpiece of dLife is the ControlCenter, a GUI application designed to
facilitate the use of dLife in the classroom. The ControlCenter allows students to observe
robot sensor values, manipulate robot effectors and experiment with the effects of image
filters all without writing any code. When it is time to write code, dLife provides access
to all of its features through an API that has been designed to be simple enough for use
in introductory CS while also being powerful enough for use in upper level courses and
research projects. In typical classroom use, students use the ControlCenter's integrated
Java editor/compiler to create Controller objects (Java classes that use the dLife API to
interact with the robot). With just a few mouse clicks the ControlCenter will compile,
load and execute a Controller. The ControlCenter manages all of the details of connecting
to the robot, sending commands and retrieving sensor values, allowing students to focus
on programming the control logic. For advanced applications, the dLife API can be
imported into any Java project and can be used more flexibly without the aid of the
ControlCenter.
The dLife API includes packages for robot control, computer vision, neural
networks, genetic algorithms and grid computing. The robotics packages currently
provide support for the following robots: MobileRobots Pioneer 3DX; K-Team

___________________________________________

*
Copyright is held by the author/owner.

173
JCSC 26, 6 (June 2011)

Hemisson, Khepera 2 and Khepera 3; and Sony Aibo. Robot simulations are supported
through the free open-source Player/Stage system. dLife's computer vision package
provides a collection of basic image filters: color matching; blobification; blurring; edge
detection; etc. Images for processing can come from a variety of sources: image file (jpg,
gif, etc.); sequence of image files (e.g. img1.jpg, img2.jpg, …); streaming video from the
Aibo; and streaming video from any V4L device (Linux server required). The neural
network package includes feedforward, backpropagation, Elman and CMAC neural
networks. The genetic algorithms package provides a generational genetic algorithm with
a variety of selection mechanisms and reproduction operators. Grid computing is
supported through an interface to Apple's xgrid technology. All of dLife's packages have
been designed with extensibility in mind facilitating the creation of new types of neural
networks, genetic algorithms, robots, robot devices, image filters and image sources.
dLife currently runs on Mac OS X and Linux. It is available for free under the GNU
General Public License (GPLv3). For additional information or to download dLife please
visit: http://www.dickinson.edu/~braught/dlife.
Acknowledgements: I would like to acknowledge the contributions that many of my
students at Dickinson have made to dLife. Russell Toris developed the Khepera 3
support and the interface to the Player/Stage simulation system. Kent Carmine, Nate
Mitchell, Alex Diehl, Luke Maffey and Mike Olasin all contributed to the Aibo support.
Lewis Flanagan developed the Elman Neural Network.

174
 AN UPDATED TAXONOMY OF WRITING IN COMPUTER

SCIENCE EDUCATION *

FACULTY POSTER

Mark E Hoffman
Department of Mathematics and Computer Science
Quinnipiac University
Hamden, CT 06518
203-582-8449
[email protected]

Analyses of the computing education literature reported by Hoffman et al [4] and

Dugan and Polanski [1] agree on a taxonomy of writing in Computer Science education
where writing can be categorized as Writing to Learn, Academic, or Professional.
Writing to Learn (WTL) writing is used to develop or reflect on understanding that is
informal and frequently ungraded. Examples of WTL writing include reading response
questions or short informal explanations. Academic writing is used to demonstrate
knowledge that is formal and typically graded. Examples of academic writing include
article summaries, research papers (e.g., CS pioneer biography), and formal explanations.
Professional writing is used to apply knowledge in a professional context that is formal
and typically graded. Examples of professional writing include technical papers and
software development documents.
Dugan and Polanski use the taxonomy to guide Computer Science educators in
overcoming reluctance to use writing in their teaching. They provide several examples
of writing for each category and offer advice to mitigate Computer Science educators'
anxiety and resistance. Hoffman et al. use the taxonomy to offer a strategy to "bridge"
writing across the taxonomy from WTL to Academic and Professional writing. They
present three writing assignments based on the strategy. More recently, Giangrande [3]
reviewed faculty objections to including writing in Computer Science courses and offered
strategies to integrate communication skills including a variety of writing types across
lower- and upper-level courses.
Curriculum standards, accrediting agencies, and universities acknowledge and
support the importance of writing for Computer Science students. Dansdill et al. [2]
found support for the importance of writing among Computer Science educators.

___________________________________________

*
Copyright is held by the author/owner.

175
JCSC 26, 6 (June 2011)

However, the papers cited above, and many others like them indicate that there is still
general resistance. Other strategies need to be explored to overcome the resistance.
An interview study with Computer Science educators at a variety of schools provide
new insights into the types of writing that are used in Computer Science education and
provide an opportunity to update the taxonomy developed by Dugan and Polanski [1] and
Hoffman et al. [4]. In addition to the Type dimension, the updated taxonomy includes an
Audience dimension. Audience emerged from the interviews as an important
characteristic of writing in Computer Science education. Computer scientists write for
audiences ranging from a machine (programming) to non-technical (customers). Table
1 shows the updated taxonomy with examples for each Type-Audience category.

Table 1. Computer Science Writing Taxonomy

Audience

Type Machine Technical Non-Technical

Writing to Learn - Code to test Develop Short explanation
Develop and concept (i.e., pseudocode of test code to
reflect on learn) Reading response customer
understanding journal Creative exercise
Short explanation

Academic - Code for a course Research paper Final project

Demonstrate assignment Proof writing report
knowledge Code comments

Professional - Code for a Technical paper Customer specs

Apply knowledge software product Design specs User manual
Bug report
Code comments

The updated taxonomy includes types of writing that are native to Computer Science
education and that may not be widely accepted as writing outside the discipline. For
example, program code and program comments are considered writing by many interview
participants. In this case, the audience is a machine. The updated taxonomy retains WTL
type writing that is used to developing understanding. WTL writing for a machine
captures short programming assignments where students learn by developing and testing
short sections of code. The updated taxonomy captures more traditional forms of writing
that are typically used to fulfill writing requirements.

176
 CCSC: Northeastern Conference

The updated taxonomy offers new opportunities for Computer Science educators to
use writing that is native to the discipline and link it to other forms of writing. For
example, a linked WTL assignment could start with testing a concept by coding
(WTL-Machine) that leads to writing pseudocode (WTL-Technical) and culminates in a
short written explanation for grammar school students (WTL-Non-Technical). Each
"translation" reinforces understanding. This type of assignment can overcome resistance
by capitalizing on writing that is native and familiar to Computer Science educators.

REFERENCES
[1] R.F. Dugan and V.G. Polanski, "Writing for Computer Science: A Taxonomy of
Writing Tasks and General Advice," JCSC (21, 6) pp 191-203, June 2006.
[2] T. Dansdill, M.E. Hoffman, and D.S. Herscovici, "Exposing Gaps, Exploring
Legacies: Paradoxes of Writing Use in Computing Education," JCSC (23, 5) pp
24-33, May 2008.
[3] E. Giangrande, "Communication Skills in the CS Curriculum," JCSC (24, 4)
April 2009.
[4] M.E. Hoffman, T. Dansdill and D.S. Herscovici, "Bridging Writing to Learn to
Writing in the Discipline in Computer Science Education," ACM SIGCSE
Bulletin (38, 1) pp 117-121, March 2006.

177
 THE COMPUTER SCIENCE CLUB: BUILDING A STUDENT

COMMUNITY WITH PROJECTS AND ACTIVITIES THAT

EXTEND BEYOND THE CLASSROOM*

FACULTY POSTER

Susan P. Imberman
College of Staten Island
2800 Victory Blvd.
Staten Island, NY 10314
718-982-2850 contact
[email protected]

Growing a community of students in a commuter college is a difficult task. Students

at commuter schools are disadvantaged with respect to resident students in many ways.
Many resident colleges have dormitory floors or sections which group students of various
disciplines together. Meeting after class to work on class projects is easier when your
classmate lives within walking distance. of your dorm room. Resident students don't have
the financial and societal pressures commuter students have that come with the
responsibilities of living at home. Gersting and Young note that student computer
science campus organizations, such as chapters of the ACM or IEEE and even more
informal computer science clubs, offer students a multitude of social, professional, and
learning opportunities. [2] At the College of Staten Island (CSI), we have leveraged our
student organization's Computer Science Club to create an environment that helps
students form a computer science community outside of the classroom.
One of the most important contributions to the creation of a CS community at CSI
is the allocation of a permanent space for the computer science club. The need for space
stemmed from a club project to create and maintain a web server for all student clubs.
Club project participants needed an accessible space to house these servers. The college
required that the servers sit outside the school's network, "just in case..." The first club
room was a faculty office sized space, equipped with web servers, college networked
computers and some desk space. Because of concerns about too many students "hanging
out" in such a small space, the space allocated to the club has grown into a classroom
sized space, that has more student computers, reference books, tutoring space, video game
equipment, Lego® robots, and desk space. Importantly, students have open access to the

___________________________________________

*
Copyright is held by the author/owner.

178
 CCSC: Northeastern Conference

club room. Students use the room as a place to "hang out" between classes. What is also
significant is that the club room is located in the middle of our faculty offices, offering
students many opportunities to interact with faculty outside of the classroom. Many club
members were able to find paid research work because of these interactions.
The club maintains its own web server and creates its own web site. The club's web
site has had many iterations. Each new generation of students feels that they can improve
and enhance and at times redo, the previous generation's work. Past club web masters
have parlayed their experience on the club's web site into "real" web related jobs.
Besides maintaining a web server, the club offers a number of ongoing computer
science projects for members to participate in. One such project is the "Autonomous
Robotic Barbie Jeep". Based on a project first proposed by Zachary Dodds [1], students
build an autonomous robotic vehicle based on a PowerWheels platform. Students are
involved in many different aspects of this project. Some students are more hardware
oriented and became involved in engineering the Jeep to drive autonomously. Others
were involved in programming the Jeep. One student even decided to write a simulator
using "Blender" a graphics program with which he was familiar. What is important to
note is that students were not bounded by any formal class assignment but allowed total
creative freedom. They were free to modify, experiment and enhance as long as those
involved in the project agreed to the changes.
The club meets weekly with different activities planned for various meetings. Guest
speakers, some former club alumni, have given talks. The club runs an annual computer
science film festival which is open to all students on campus. Some members participate
in service learning projects such as tutoring, sponsoring video game competitions for
charity, and creating web pages for other clubs. Meetings serve a social function as well.
Suffice it to say there are many opportunities for students to become involved.
The success of our computer science club has not only been beneficial to building
a student community but has benefitted our department as well. Our feeling is that club
members are more likely to continue and finish the computer science major as opposed
to students who don't participate in the club. We are investigating whether the or not
computer science club contributes to student retention. Our feeling is that this is most
probably so.
The contribution of this poster will be to highlight the importance of building a
student community via a computer science club. It will focus on how it benefits not only
students but faculty as well. The author has been advisor to the computer science club
for the past 11 years and has seen it grow into an integral part of CSI's computer science
experience.

REFERENCES
[1] Dodds, Z., 2008. Leveraging Laptops For Low-Cost, Full-Fledged Outdoor
Robotics. J. Comput. Small Coll. 24, 1 (October 2008), pp 158-166.
[2] Gersting, A.,Young F. 1998, Service Learning Through the Computer Science
Club, ACM SIGCSE Bulletin 30:4, pp 25-26.

179
 CAPSTONE EXPERIENCE: ENGAGING STUDENTS IN

SPEECH PROCESSING TO EXCITE THEM ABOUT STEM*

FACULTY POSTER

Michael Jonas
Department of Computer Information Systems
University of New Hampshire at Manchester
Manchester , NH 03101
857.389.2692
[email protected]

The newly created Capstone course in Computer Information Systems (CIS) at UNH
Manchester starting in spring 2011 offers students a faculty guided project. The course
will expose students to the rigors of speech processing, giving them the opportunity to
engage in solving real world problems, gaining invaluable experience along the way. By
combining the Capstone course with an active research agenda in speech the course will
be used as a pilot program for a National Science Foundation (NFS) Transforming
Undergraduate Education in Science, Technology, Engineering and Mathematics (TUES)
grant proposal.
The purpose of the NFS TUES program is to improve the quality of science,
technology, engineering and mathematics education at the undergraduate level. One way
to excite students about science is to involve them in direct scientific discovery. Speech
provides a rich area which encompasses all those aspects required by the TUES program.
For a wide area of technological fields, speech can be applied to solve real world
problems. Both its solution and applications are engineering problems that rely heavily
on mathematics. To create a viable solution, good scientific principals need to be applied.
As the Capstone course is a one semester program, having only 16 weeks requires
a staged approach. The inaugural project will focus on developing the platform and
processes to enable active research. Specifically, students will learn with the Carnegie
Mellon University Sphinx open source toolkit for speech recognition, building tools to
use on a set of servers running Linux. The goal will be to develop baseline models and
capture the techniques in creating them by crafting tools in the form of scripts and utility
programs to ease the recreation of those models. Each subsequent course will expand on
the previous, enabling more sophisticated research as the program matures.

___________________________________________

*
Copyright is held by the author/owner.

180
 CCSC: Northeastern Conference

Additionally, creating an experiment paradigm along with all tools required to

generate a viable system is an important exercise that will expose students to all facets
of the speech process. Not only will the aim be to capture the techniques used to create
the models, but to track and document performance results of those models for future
comparison. This is a critical element in any experimental research field.
As the Capstone course matures the aim is to partner with small high tech firms to
solve real world speech projects. Think of it as a low cost alternative to an expensive
speech consulting firm. Students would be key resources in solving problems towards
specific project goals with expert guidance. This would allow partners, both external as
well as within UNH a viable and cost effective solution to speech related problems.
Close ties with industry will enable placement opportunities for graduating seniors
with high tech companies specializing in the field of speech. For non-seniors the
possibility for summer internships in industry is also a very likely prospect. Along with
that, a summer program as a follow up on the Capstone course will also be created to
allow interested student a more in-depth and focused experience.

OPTIONAL REFERENCES
[1] Transforming Undergraduate Education in Science, Technology, Engineering
and Mathematics (TUES), NSF 10-544, February 24, 2010

181
 A WEB SERVICE TO CONNECT DATA ANALYSIS AND

VISUALIZATION*

FACULTY POSTER

Thomas Jyh-Cheng Liu

Department of Computer Science
New Jersey City University
Jersey City, NJ 07733
201-200-3140
[email protected]

This poster presents the preliminary results of software that MSI (Minority Serving
Institution) summer faculty-student team developed[1]. The software is a designated web
service to facilitate cyber security research: It connects multiple data sources with
analysis software and visualization tools for the Department of Homeland Security and
the homeland security community. This project has demonstrated the effectiveness of a
faculty-student team throughout the development of this software. The outcome has been
successful thus far (the project is ongoing) as well as inspiring to the participating
students; It has been a precious learning experience in computer science education as well
as a practice in enterprise-level software development.
As an illustration of the preliminary outcome, this paper presents the analysis of a
large dataset: email exchanges between Enron employees. The data visualization software
adopted for this web project is Graphviz[2], an open source graphics project. The
graphical user interface was designed using Adobe Flash Builder (Flex). The second
phase of this study will employ XML, WSDL and SOAP to facilitate inter-application
communication. The design will draw upon Mirage[3], an open source Java-based
visualization tool. This framework will provide a "one-stop shop" to integrate data
analysis and visualization tools for applications of homeland security community.
This project was initiated during summer research at the Command, Control and
Interoperability Center for Advanced Data Analysis (CCICADA) of the Department of
Security at Rutgers University. A faculty-student team was formed and mentored by
researchers at CCICADA.
As of now, only the basic framework of this summer project has been developed and
tested. One of major accomplishments has been the user interface, developed by writing
Flex MXML. MXML is a rich Internet application markup language for building Web

___________________________________________

*
Copyright is held by the author/owner.

182
 CCSC: Northeastern Conference

applications that deploy on all major browsers, desktops and other platforms (it compiles
to Flash actionscript). On the server side, the XAMPP package was downloaded and
installed on both Linux and Windows XP.
The applications installed on the server-side include: igraph [4], GraphViz, and the
industry-standard R [5] programming language for statistical computing and graphics. In
addition, Perl and PHP are adopted as server-side scripting languages for implementation
of the application backend. Adobe Flash player is the only requirement on the client-side.
Once a user provides an input data file and selects relevant parameters using given
front end, the back-end Perl scripts are executed to conduct calculations (via other
programs written in the R statistical language). In addition, data is communicated to
Graphviz, which generates the intended graphs, histograms and relevant outputs.
XMLhttprequest objects, the generated graphs, diagrams and other outputs are then
delivered to the front-end and displayed on the user's browser window.
The second phase of this project will be continued during the summer of 2011. The
expected result includes a fully operational web service to deploy visualization tools like
Mirage (a Java-based open source software developed in Bell Labs), and a model for
similar tools and software to be deployed on the same framework.

ACKNOWLEDGEMENTS
Dr. William Pottenger, Research associate professor of Rutgers University and Director
of transition, CCICADA in Department of Homeland Security.
Dr. Tin Ho, head of the Statistics and Learning Research Department of Bell labs in
Alcatel-Lucent co.
Dr. Tami Carpenter, Associate Director of CCICADA in Department of Homeland
Security and coordinator for MSI (Minority Serving Institution) faculty student team
summer research program.
Monty Bains, a student member of the summer research team and a current graduate
student of NJIT.

REFERENCES
[1] Thomas Jyh-Cheng Liu, "Report for MSI summer research - A web service to
connect Data Analysis and Visualization", CCICADA at Rutgers University, NJ,
2010.
[2] Graphviz project page. http://www.graphviz.org/, 2010
[3] Tin Kam Ho, in M. Graham, M. Fitzpatrick, T. McGlynn (eds.), "Mirage: A
visual tool for Scientific Inquiries into the VO Archives", The National Virtual
Observatory: Tools And Techniques For Astronomical Research, Astronomical
Society of the Pacific, ASP Conference Series Vol. CS-382, pp. 29-36, 2008.
[4] Igraph Library source - http://igraph.sourceforge.net/ , 2010
[5] The R project for statistical computing. http://www.r-project.org, 2010

183
 USING PENNY AUCTIONS TO INTRODUCE NON-CS

MAJORS TO MODELING AND SIMULATION*

FACULTY POSTER

David Toth
Department of Computer Science
Merrimack College
North Andover, MA 01845
978-837-3405
[email protected]

Computation has become an accepted technique for scientists, mathematicians,

economists, and others to learn about complex phenomena that might otherwise be
impractical or impossible to study. Computational science uses modeling to create a
simplified view of the phenomenon being studied and simulation to see how the
phenomenon will behave over time under given conditions [1]. Computational science
is used for pharmaceutical development, weather forecasting, predicting the behavior of
the stock market, studying disasters like the oil spill that occurred in the summer of 2010,
and many other things. A number of colleges and universities have majors in
computational science [2, 3, 4, 5, 6, 7, 8].
We introduce the idea of modeling and simulation in our service course,
Introduction to Information Technology, in order to give non-STEM students an idea of
how computing can be useful for solving complex problems. One desired learning
outcome of introducing modeling and simulation in this course is that the students will
be able to take a new problem and analyze it to determine what they need to know in
order to solve the problem. The required knowledge includes how a particular system
works as well as the input parameters that affect the behavior of the system and the
outcome of the simulation. The other desired learning outcome is that students learn how
to explain how the system works in prose, incorporating the parameters, so that another
person can read their description and understand how the system works. In the
Introduction to Information Technology course, we have the students construct their
simulations using Scratch because of the ease of converting English text into Scratch code
[9]. The introduction to modeling and simulation also provides the transition in that
course to the section on parallel and distributed computing, motivating the need for
multi-core computers, clusters, supercomputers, and networking technologies.

___________________________________________

*
Copyright is held by the author/owner.

184
 CCSC: Northeastern Conference

We are also planning on introducing a new course, Introduction to Computational

Science, this fall for biology, chemistry, and physics majors. For the Introduction to
Computational Science course, in addition to the learning outcomes for the modeling and
simulation activities from our Introduction to Information Technology course, we also
want the students to learn to build complex models and simulations with some of the
standard tools.
We have used several different simple problems to motivate students in our
Introduction to Information Technology course in the past [10]. We had mixed results
and decided to develop a new problem that the students would find more interesting.
Because many of the students taking the Introduction to Information Technology course
are business majors, we decided to develop a problem on penny auctions. Penny auctions
have become popular on the Internet in the last couple of years. These auctions differ
from traditional auctions because it costs some amount of money to make a bid on an
item. In addition to this, bids increase by one cent. Thus, the auctions are able to sell an
item at a fraction of its retail cost and still make a profit, while the winning bidder is able
to purchase the item at a fraction of its cost. We believe that this problem will also work
well for our new Introduction to Computational Science course because of the simplicity
of the problem and the ability to expand the problem and make it much more complex.
In this poster, we present an explanation of how penny auctions work and discuss
how the company running the auction and the participants in the auction view the
financial ramifications of running and participating in the auctions respectively. We also
discuss the parameters that are involved in modeling the auctions so an accurate
simulation can be created and present a simple simulator that we expect our students to
be able to create. We will use this example in our courses next year, gauging interest
with an anonymous survey and assessing the effectiveness with a quiz in Introduction to
Information Technology and a project in Introduction to Computational Science.

REFERENCES
[1] National Computational Science Institute Introduction to Computational
Thinking Workshop, August 2-8, 2009.
[2] The Florida State University, The Florida State University,
http://www.academic-guide.fsu.edu/computational_science.html, retrieved
January 3, 2011.
[3] The College at Brockport, State University of New York, Computational
Science: The College at Brockport, State University of New York,
http://www.brockport.edu/cps/, retrieved January 3, 2011.
[4] Georgetown College, Computer Science,
http://www.georgetowncollege.edu/Departments/mpc/computational_sciences.ht
m, retrieved January 3, 2011.
[5] University of Waterloo, Findoutmore | University of Waterloo,
http://www.findoutmore.uwaterloo.ca/programs/full_description.php?program=C
omputational%20Science, retrieved January 3, 2011.

185
JCSC 26, 6 (June 2011)

[6] University of Wisconsin-La Crosse, COMPUTATIONAL SCIENCE (CMP),

http://www.uwlax.edu/records/UGCat/03-05/UG-Cat/cmp.html, posted August
25, 2008.
[7] The Australian National University, ANU - Bachelor of Computational Science
(BComptlSci), http://bcomptlsci.anu.edu.au/, posted November 28, 2006.
[8] The Richard Stockton College of New Jersey, Academic Affairs - Natural
Sciences and Mathematics - Computational Science Program,
http://intraweb.stockton.edu/eyos/page.cfm?siteID=14&pageID=118, retrieved
January 3, 2011.
[9] Scratch | Home | imagine, program, share, http://scratch.mit.edu/, retrieved
October 26, 2009.
[10] Toth, D., Problems for Introducing Non-Majors to Modeling and Simulation in a
Service Course, Poster at the SC Education Curriculum Resource Fair at the
international Supercomputing conference (SC10), November 13-19, 2010, New
Orleans, LA, USA.

186
 THE DESIGN AND COMPLEXITY ANALYSIS OF THE

LIGHT-UP PUZZLE PROGRAM*

FACULTY POSTER

Douglas Selent
Department of Computer Science
Rivier College
Nashua, NH 01886
978-692-8478
[email protected]

The purpose of this project was to use the McCabe IQ tool to analyze source code
complexity of the Light-up Puzzle program. The McCabe IQ tool is software used in
quality assurance for code analysis. It can allow people to determine if any parts of the
source code are unreliable or unmaintainable without the user having to look at the source
code. This can save both time and money as well as improve the quality of software code
in businesses. Basically what the tool does is convert the source code into a graph, which
it then uses to analyze the source code. The features of the tool that were used in this
project are the Battlemap, System Complexity metrics, method flowgraphs, scatter
diagrams, Halstead metrics, Class metrics, and Object-oriented metrics.
The McCabe IQ tool provides two main benefits to students. The first benefit is that
students can learn to debug large volumes of code that is not their own. This is a task
students rarely do in school, but may have to do in a work environment. The McCabe IQ
tool can be used to quickly identify areas of code that have a high probability of errors.
It is unlikely that the student would have to search through source code to find bugs. The
second benefit is getting a better idea of what good programming is. Although good
programming practices are taught, specific cut-off points for function size and complexity
are usually not mentioned or followed. Several functions that are thought to be
well-coded are really not. Students who use the McCabe IQ tool can get a better idea of
this by programming to the standards of the McCabe IQ software.
The code chosen for analysis is my program of the puzzle "Light-Up". The program
consists of roughly 4,000 lines of code with around 20 functions among six classes. It
has a GUI for users to actually play the puzzle and a heuristic for attempting to
automatically solve any puzzle. The main volumes of code are the GUI setup and event

___________________________________________

*
Copyright is held by the author/owner.

187
JCSC 26, 6 (June 2011)

handlers, the algorithms for manipulating the puzzle, and the heuristic that attempts to
automatically solve any puzzle.
The source code was analyzed by the McCabe IQ tool and the data was interpreted
to identify problems in the code. A new version of the program was coded to fix the
problems found in the old version. Data for both versions were compared to each other
to identify whether there was improvement between the two versions.
The Battlemap of the original version had 22 (63%) green modules, 1 (3%) yellow
module, and 12 (34%) red modules, for a total of 35 modules. The Battlemap of the new
version had 101 (81%) green modules, 1 (1%) yellow module, and 23 (18%) red modules,
for a total of 125 modules. Although the number of red modules almost doubled from the
old version to the new version, the new version was an improvement. The percentage of
green modules increased from 63% in the old version to 81% in the new version and the
percentage of red modules decreased from 34% to 18%. This indicates significant
improvement in the source code between the old and new version. The reason why the
numbers are deceptive is because 90 new less-complex modules were created by breaking
up a few complex modules.
There was also significant improvement in the System Complexity metrics. The old
version had a highest Cyclomatic complexity value of 281, highest Essential complexity
value of 125, and highest Design complexity of 98. All three values are awful. The new
version had a highest Cyclomatic complexity value of 31, highest Essential complexity
value of 13, and highest Design complexity of 18. All three values are slightly bad.
Considering that the worst values in the new version are only slightly high, that means
the rest of the modules and the program as a whole is good. The problem occurred with
the Class Metrics. The original version had no problems at the class level because there
were not many functions. When 90 new functions were created from breaking up big
functions, it introduced problems for the Class metrics. The RFC and WMC values for
the "Dougrithm" class (the class with all the puzzle algorithms) jumped from 12 to 102.
There were two key discoveries made in this project. The first is that blindly using
the McCabe IQ tool is not a good way to analyze the code. Users must know how to
interpret the data correctly in order to gain the benefits of the tool. The Battlemap was
a clear example where the Battlemap of the new version looked much worse than the
Battlemap of the old version. At first glance people notice how there is many more red
modules in the new version than the old version. They also notice the increase in the
number of modules. The percentage of red/green modules is usually unnoticed.
Therefore someone might think that the new version is worse when it is really much
better. The second discovery is that changes to fix one problem could have unforeseen
consequences on other metrics of the program. The changes made to fix the System
Complexity metrics had a negative effect on the Class metrics. When there were only a
few functions there was no problem. When those functions were broken down into
several new functions, the communication among functions dramatically increased. For
future work the functions of the "Dougrithm" class could be grouped together by what
task they carry out. Then all the function groups can be separated into different classes
to fix the problem that was introduced at the class level.
I would like to acknowledge Tim Houle for contributing to the analysis.

188
 CCSC: Northeastern Conference

REFERENCES
[1] McCabe, T., Watson, A., Structured Testing: A Testing Methodology Using the
Cyclomatic Complexity Metric. NIST Special Publication 500-235, August 1996.

189
DIGITAL HOME: A CASE STUDY APPROACH TO TEACHING

IN COMPUTING CURRICULA*

FACULTY POSTER

Massood Towhidnejad, Salamah Salamah, and Thomas Hilburn

Department of Electrical, Computer, Software, and Systems Engineering
College of Engineering, Embry Riddle Aeronautical University
Daytona Beach, Florida, 32114.
386-226-6891.
Towhid, salamahs, [email protected]

Case studies have been a subject of much study and research about their
effectiveness in teaching and learning [1, 2, 3]. Although the use of case study teaching
has become a proven and pervasive method of teaching about professional practice in
such fields as business, law, and medicine, it is yet to be adopted in any significant way
in the computing education. One of the reasons for the paucity in the use of the case-study
approach is the lack of sufficient material for this purpose. For example, among the
hundreds of case studies in science at the SUNY Buffalo web site [4] only three concern
computer science or software engineering.
Digital Home [11, 12] is a comprehensive case study that can be used throughout
a computing curriculum. The case study material include complete software development
artifacts as well as teaching case modules to be used to teach different topics (i.e.,
software inspections, object oriented design and implementation…) throughout a
computer science or software engineering curriculum.
There are many software textbooks that use case studies (as examples to illustrate
concepts and techniques): [5] (several, including an Airline Reservation System and a
Household Alarm System), [6] (Fireworks Factory), [7] (Picadilly Television and
Ariane-5), [8] (SafeHome). Additionally, there has been an increased effort in introducing
case studies in computing courses as reported in the literature [9, 10].
Although these cases seem to serve a specific purpose (e.g., discussing planning,
requirements, design, risk analysis, or construction issues by using simple examples that
are quasi-realistic), they often lack the following:
• Realistic artifacts (often space does not allow providing a complete requirements or
design document)
• Completeness (covers only a portion of the life-cycle, and not an end-to-end), with
a focus on design and implementation
• Ability to decouple from the text and apply in ways not intended by the author
• Techniques for integration into course activities or into the curriculum as a whole

190
 CCSC: Northeastern Conference

• A scenario format that motivate students' engagement in problem

identification/solution.
• Guidance to the instructor on how to use the case study to teach a course topic or
concept.
The DigitalHome case study aims to address these shortcomings by providing a
complete set of artifacts associated with software development as well as providing case
modules that can be used by faculty in teaching different subjects in a computing
curriculum. Each case module represents a mini case study and is associated with a
specific teaching subject (e.g. requirements inspection, object oriented design, testing…)
and learning objectives. Case modules also include instructions and desired class setup
to assist the instructor in teaching the session. A current list of the Digital Home artifacts,
case modules, and exercises can be found at:
http://www.softwarecasestudy.org/index.html
The goal of this work is to advance the understanding of case study based teaching
and to introduce the audience to a comprehensive case study in software development.
The goals of the poster are as follows:
1. Describe the background and effectiveness of case study techniques in professional
education
2. Introduce the DigitalHome software development case study material
3. Advertise the use of case study material (artifacts and case modules) to faculty in
a variety of computing programs and courses.
4. Invite audience of faculty to contribute to the development and assessment of the
quality and utility of the case study material.

ACKNOWLEDGMENT
Initial work on the DigitalHome case study was funded as part of the NSF project:
"The Network Community for Software Engineering Education" (SWENET) (NSF
0080502. The current work on the case study is funded through NSF's (DUE- 0941768)
"Curriculum-wide Software Development Case Study".

REFERENCES
[1] Davis, C. and Wilcock, E., "Teaching Materials Using Case Studies", UK Centre
for Materials Education, (http://www.materials.ac.uk/guides/casestudies.asp,
accessed Jan 2011).
[2] Grupe, Fritz H. Jay, Joelle K., "Incremental Cases", College Teaching,
September 22, 2000.
[3] Tomey, Ann M., "Learning with Cases", Journal of Continuing Education in
Nursing, Vol 34, No 1, January/February 2003
[4] National Center for Case Study Teaching in Science:
http://ublib.buffalo.edu/libraries/projects/cases/case.html

191
JCSC 26, 6 (June 2011)

[5] Lethbridge T, C., Laganière, R., "Object-Oriented Software Engineering:

Practical Software Development using UML and Java" Second Edition, McGraw
Hill, 2001
[6] Metseker, S., "Design Patterns in Java", Second Eddition, 2006
[7] Pfleeger, S, L., "Software Engineering" Pearson, 2005
[8] Pressman, R, S., "Software Engineering: A Practitioner's Approach",
McGraw-Hill, 2007
[9] Linn, M. , Clancy M., "The case for case studies of programming problems",
Communications of the ACM, v.35 n.3, p.121-132, March 1992
[10] Parrish, A., Hale, D., Disxon, B., Hale, J., "A Case Study Approach to Teaching
Component Based Software Engineering in Proceedings of the Thirteenth
Conference on Software Engineering Education and Training. 2000
[11] Hilburn, T., Towhidnejad, M., "A Case for Software Engineering" Proceedings
of the 20th Conference on Software Engineering Education and Training, 2007
[12] Hilburn, T., Towhidnejad, M., Salamah, S., "The DigitalHome Case Study
Material" Proceedings of the 21st Conference on Software Engineering
Education and Training. 2008

192
 MATHDOKU PROJECTS: MONTE-CARLO MEETS

METHODOLOGY*

FACULTY POSTER

Linda Wilkens
Department of Mathematics and Computer Science
Providence College
Providence, RI 02918
401 965-1896
[email protected]

This poster presentation reports on classroom experience using the logic/math

puzzles known by the generic name Mathdoku[1] for student programming projects in a
Data Structures course. Using these puzzles encourages students to employ an
object-oriented approach to the two major components of the project (1) a user or
computer puzzle player component, and (2) a puzzle generation component. This project
affords students the chance to develop a non-trivial application, to re-use data structures
developed during the semester, and to leverage existing data structures from framework
libraries. It also offers ample opportunity for customizations, so that no two projects
should end up the same.
Students were provided with starter code in small sections. The first iteration simply
presented an existing puzzle, read in from a file. The second iteration looked at possible
ways to support user play, and looked ahead to computer play. Project development next
concentrated on puzzle generation, in three separate steps: First, entirely manual creation,
with the user specifying both the numbers for each square as well as the cage layouts and
operations. Second, a semi-automatic creation mode in which the program generates the
numbers for the squares and the user then specifies the cages. The third and final mode
is a fully-automatic mode, in which the program completely generates the puzzle. We
follow [2] in classifying our puzzle generation as implementing a Monte Carlo algorithm,
since "a Monte Carlo algorithm is a randomized algorithm whose running time is
deterministic, but whose output may be incorrect with a certain (typically small)
probability." We allow that a puzzle is incorrect if it has more than one solution.
At this stage, development switched back to the task of computer-generated
solutions. This is where the "methodology" took part, in that the game tree concept was
employed, using either a standard depth-first or breadth-first search though the generated

___________________________________________

*
Copyright is held by the author/owner.

193
JCSC 26, 6 (June 2011)

tree to find a solution. One of the students in the class did a very nice comparison of the
number of nodes explored finding a solution via the two standard search methods.
The last step of the project was to create a third application that would spawn a
process to execute either the puzzle-playing application or the puzzle-generation
application. This gave students experience with the process concept, and using it in a
program.

REFERENCES
[1] http://www.mathdoku.com, retrieved Jan 4, 2011.
[2] Wikipedia: The Free Encyclopedia, Wikimedia Foundation Inc. Encyclopedia
on-line, Fifteen Puzzle, 2010, en.wikipedia.org/wiki/ Monte_Carlo_algorithm,
retrieved Jan 11,

194
 MINING FORUMS FOR ASYNCHRONOUS LEARNING IN

DISTANCE EDUCATION*

FACULTY POSTER

Songmei Yu, Sofya Poger, Alberto LaCava

Department of Computer Information Systems
Felician College
262 South Main Street
Lodi, NJ 07644
201-559-6000
[email protected], [email protected], [email protected]

I. INTRODUCTION TO AWK ALGORITHM

Distance Education is a process of delivery of instructional resources online. It is
a flexible form of education where a student can study from home, work, or wherever else
is convenient. Computer based teaching materials, online conferencing, discussion
boards, e-mail, and other techniques are used in distance education. The information can
be delivered synchronously and asynchronously. The example of synchronous method
of delivery is Web conferencing where all participants are "present" online at the same
time. The discussions boards, e-mail and recorded video are examples of asynchronous
method of delivery, where participants access course material at their own schedule [1].
Online discussions and forums allow students and instructors to exchange the ideas,
research and information. The aim of this research is to apply data mining techniques to
discussions boards and forums to measure the student participation and performance. In
this research, we plan to introduce and implement the AWK (Author Weighted
Keywords) Algorithm as a tool to measure the student performance and learning
capabilities by monitoring their participation in distance learning forums, extract
discussion topics for later comers and help instructors to guide the online discussions.
The AWK algorithm is based on the following Web Mining Techniques:
• Data/Information extraction
• Web information integration and schema matching
• Opinion extraction from online sources
• Knowledge synthesis
• Segmenting Web pages and detecting noise

___________________________________________

*
Copyright is held by the author/owner.

195
JCSC 26, 6 (June 2011)

An alternative to web searching can be applied to students participating in online

courses using a course management system (CMS). A CMS uses a database program to
store the information of the course, including forum entries. As an example, Moodle [2]
frequently uses a MySQL database engine and stores the information in a database file.
This file and the files in the system can then be searched using the database engine, and
the AWK algorithm can be applied directly to the database contents.
By using data mining techniques we extract the keywords from each thread or topic.
The weight for each student is calculated based on number of threads initiated, number
of posted comments and responses, and content of posted comments and responses. The
content of posted comments and responses is defined by the several factors of extracted
keywords, such as the distance from keywords to other words in a post. The student
performance is based on the weight - the higher the weight, the better performance. The
AWK algorithm helps to understand the dynamic behavior of students in the Web systems
and improve the teaching process in more efficient and effective ways, compared to other
Web mining techniques in current distance learning field [3].

II. OUR FUTURE PLAN

• Build source data from the real online materials of Computer Information Systems
courses from a live distance learning forum (e.g. under Moodle)
• Construct tree structure by simulating distance learning forum scheme
• Extract keywords using text mining techniques and natural language processing
• Apply Author Weighted Keywords (AWK) Algorithms
• Compare manual calculations of the student performance with the result obtained
using AWK algorithms.

III. POSTER CONTENT

The poster will provide an overview the Web Mining techniques, challenging issues
in mining distance learning forums, Author Weighted Keywords Algorithm, and our
implementation plan.

REFERENCES
[1] S. Androutsellis-Theotokis, D. Spinellis, A Survey of peer-to-peer content
distribution technologies, ACM Computing Survey, (36) 4, pages 335-371, 2004
[2] A description of Moodle is given in the community page
http://moodle.org/about/
[3] MJ. Callaghan, M.El-Gueddari, J. Harkin, TM. McGinnity, LP. Maguire,
Intelligent Remote Experimentation, Proceedings of the IEEE/WIC/ACM
International Conference on Web Intelligence (WI'04), 2004.

196
 INDEX OF AUTHORS
Abunawass, A . . . . . . . . . . . . . . . 156 Madigan, E . . . . . . . . . . . . . . . . . . . 64
Bailie, F . . . . . . . . . . . . . . . . . . . . 156 Meinke, J . . . . . . . . . . . . . . . . . . . . . x
Baird, B . . . . . . . . . . . . . . . . . . . . 165 Milonovich, B . . . . . . . . . . . . . . . . 78
Banerjee, S . . . . . . . . . . . . . . . . . . 168 Moll, R . . . . . . . . . . . . . . . . . . . . . . 55
Barneva, R . . . . . . . . . . . . . . . . . . 170 Morreale, P . . . . . . . . . . . . . . . . . . . 71
Blake, J . . . . . . . . . . . . . . . . . . . . . 124 O'Rourke, J . . . . . . . . . . . . . . . . . . . 78
Braught, G . . . . . . . . . . . . . . . 51, 173 Olivieri, L . . . . . . . . . . . . . . . . 23, 107
Brimkov, B . . . . . . . . . . . . . . . . . . 170 Peluso, E . . . . . . . . . . . . . . . . . . . . 12
Brusilovsky, P . . . . . . . . . . . . . . . 160 Peters, S . . . . . . . . . . . . . . . . . . . . . . 6
Cassel, L . . . . . . . . . . . . . . . . . . . . 160 Petrovic, S . . . . . . . . . . . . . . . . . . 156
Chua, M . . . . . . . . . . . . . . . . . . . . . 10 Poger, S . . . . . . . . . . . . . . . . . . . . 195
Decker, A . . . . . . . . . . . . . . . . . . . . . 7 Prey, J . . . . . . . . . . . . . . . . . . . . . . . . 5
Defoe, D . . . . . . . . . . . . . . . . . . . . 163 Proulx, V . . . . . . . . . . . . . . . . . . . 149
Delcambre, L . . . . . . . . . . . . . . . . 160 Riabov, V . . . . . . . . . . . . . . . . . . . . 86
Dickson, P . . . . . . . . . . . . . . . . . . 136 Rilett, D . . . . . . . . . . . . . . . . . . . . 130
Dziallas, S . . . . . . . . . . . . . . . . . . . 10 Roberson, C . . . . . . . . . . . . . . . . . . 30
Eckmann, M . . . . . . . . . . . . . . . . . 117 Russo, J . . . . . . . . . . . . . . . . . . . . 130
Fischbach, A . . . . . . . . . . . . . . . . . 53 Salamah, S . . . . . . . . . . . . . . . 13, 190
Fischer, A . . . . . . . . . . . . . . . . . . . . 61 Sarawagi, N . . . . . . . . . . . . . . 23, 107
Fitzgerald, S . . . . . . . . . . . . . . . . . 133 Selent, D . . . . . . . . . . . . . . . . . . . . 187
Fox, E . . . . . . . . . . . . . . . . . . . . . . 160 Stevenson, C . . . . . . . . . . . . . . . . . 37
Furuta, R . . . . . . . . . . . . . . . . . . . 160 Stewart-Gardiner, C . . . . . . . . . . . . 16
Garcia, D . . . . . . . . . . . . . . . . . . . 160 Stolerman, J . . . . . . . . . . . . . . . . . . 44
Goldschmidt, D . . . . . . . . . . . . . . . 78 Stone, J . . . . . . . . . . . . . . . . . . . . . . 64
Grammer, K . . . . . . . . . . . . . . . . . . 44 Teresco, J . . . . . . . . . . . . . . . . . . . . 58
Gudmundsen, D . . . . . . . . . . . 23, 107 Torrey, L . . . . . . . . . . . . . . . . . . . 110
Hanebutte, N . . . . . . . . . . . . . . . . 101 Toth, D . . . . . . . . . . . . . . . . . . . . . 184
Harrison, W . . . . . . . . . . . . . . . . . 101 Towhidnejad, M . . . . . . . . . . . 13, 190
Hilburn, T . . . . . . . . . . . . . . . . 13, 190 Trees, F . . . . . . . . . . . . . . . . . . . . . . 7
Hislop, G . . . . . . . . . . . . . . . . 10, 160 Werner, M . . . . . . . . . . . . . . . . . . 130
Hoffman, M . . . . . . . . . . . . . . . . . 175 Whitfield, D . . . . . . . . . . . . . . . . . 156
Imberman, S . . . . . . . . . . . . . . . . . 178 Wilkens, L . . . . . . . . . . . . . . . . . . 193
Ivanov, L . . . . . . . . . . . . . . . . . . . 142 Wu, H . . . . . . . . . . . . . . . . . . . . . . 130
Izmirli, O . . . . . . . . . . . . . . . . . . . 165 Wurst, K . . . . . . . . . . . . . . . . . . . . . . 1
Jackson, S . . . . . . . . . . . . . . . . . . . . . 1 Yi, B . . . . . . . . . . . . . . . . . . . . . . . . 44
Joiner, D . . . . . . . . . . . . . . . . . . . . . 71 Yu, S . . . . . . . . . . . . . . . . . . . . . . 195
Jonas, M . . . . . . . . . . . . . . . . . . . . 180
Kurkovsky, S . . . . . . . . . . . . . . . . 163
Kussmaul, C . . . . . . . . . . . . . . 10, 135
LaCava , A . . . . . . . . . . . . . . . . . . 195
LeBlanc, M . . . . . . . . . . . . . . . . . 130
Liu, T . . . . . . . . . . . . . . . . . . . . . . 182
MacDonald, I . . . . . . . . . . . . . . . . . 78
MacKellar, B . . . . . . . . . . . . . . . . . 93

197
198
199
200
JCSC Volume 26 Number 6 June 2011

Muhlenberg College NON-PROFIT ORG.

U.S. POSTAGE
2400 Chew Street
Allentown, PA 18104-5586 PAID
ALLENTOWN, PA
PERMIT #759