Olympiads in Informatics, 2012, Vol.

6, 226228
2012 Vilnius University

226

REVIEWS, COMMENTS

Competitive Programming 2
Author: Steven HALIM, Felix HALIM
Publishing house: Lulu
Country: United Kingdom
Year of edition: 2011
Language: English
Number of pages: 262

There are many ways in which students can be prepared for programming contests and
indeed training more generally in algorithms. Two of the more obvious areas are exposure
to a wide variety of different algorithms and techniques, and practice by solving a good
number of such problems. The former is important, not only at the initial levels where a
student is introduced to the fundamental building blocks (e.g. dynamic programming or
depth-first search) but at the progressive levels where a knowledge of more sophisticated
methods gives a useful, if not essential, toolbox. Algorithmic programming contests have
changed significantly since the 80s and, as this book mentions in passing, what were
once the deciding problems are now basic requirements. The need for practice should be
clear. Not only does it give the student the opportunity to confirm they understand the
algorithms, it develops skills in finding appropriate algorithms, appreciating the different
(test) cases an algorithm needs to solve, seeing ways in which multiple algorithms and
data-structures can be combined and, from a purely contest perspective, helps in increasing a students speed and accuracy.
Competitive Programming 2 is an excellent resource for both exposure and practice.
It is the second edition (the first edition is just Competitive Programming) of this book,
which grew originally from a similarly titled course that has been taught at the National
University of Singapore since 2009. Its page count puts it at over 70% larger than the
first edition. Readers of that edition, considering whether to also buy that edition can find
a detailed comparison of the two editions (and the planned third edition) at the authors
website [1]. The book is available in both A4 and A5 formats and an electronic version
should be available by the time you read this. The A4 copy was review and worked well

Reviews, Comments

227

as a format. This reviewer has heard from a colleague that the A5 edition is too small,
although has not personally seen a copy at that size.
The format throughout the book is generally a brief overview of a given topic, followed by multiple algorithms, exercises, a long list of categorised problems to practise,
and a brief conclusion. Starting with the categorised problems, this is a wonderful resource. Around 1300 problems appear in the book (almost exclusively from the University of Valladolid [2] online judge) and the bulk of these appear in these lists. These
problems, which appear after an algorithm or group of related algorithms, typically have
multiple sections each of which list a good number of problems and also highlights three
must try problems. For example, the graph traversal problem list contains 56 problems (18 must try problems) over 6 sections: Just Graph Traversal; Flood Fill / Finding
Connected Components; Topological Sort; Bipartite Graph Check; Finding Articulation
Points / Bridges; Finding Strongly Connected Components. The ad-hoc section towards
the front of the book contains 160 problems. Superb!
The description of algorithms through the book is a little more variable and varies
from very detailed to incredibly terse. In general there is an okay amount of detail; enough
for a motivated student to get the idea of the algorithm and start to play with it (or find
details elsewhere) but not enough for the book to stand alone as a text for the weaker
students. In fairness, the book does not set out to become another introductory algorithm
textbook the prerequisites suggests that it expects its audience to have passed a basic data structures and algorithms course although contrary to this the very detailed
sections tend to be the more basic material, for example in the dynamic programming
explanation. When it goes to the other extreme it is often because of tendency to give
the algorithm through code rather than explanation or just to drop in the name so that the
algorithm is ticked-off.
The style used for each algorithm varies but the following are typical items that appear. There is often a motivating problem, discussion of specific tasks, examples of using
the algorithm for other algorithmic tasks, exercises (as distinct from tasks, these are of
the type more typical in a normal algorithm text; hints and solutions appear for many of
these exercises), examples, discussions on complexity and short biographies of the people involved. Example code appears frequently in the text, and is available in both C++
and Java forms online. The coverage of the book is good, and focuses on what is used in
programming contests, with the broad chapter headings covering pretty much what you
would expect. The focus on contests works well and enables the book to offer focused
advice on choosing an algorithm; weighing up the ease of implementation against the
bounds specified in the problem or those generated by a specific sub-problem in a given
solution.
The book, while applicable to all programming contests, is focused on the ACM ICPC
and the IOI, and chooses its subject material accordingly. Where appropriate in the text
the authors indicate where material is only relevant to one of these contests. The book is
skewed towards the ICPC and the authors, whilst displaying a very broad knowledge of
its problem set (the authors claim to have solved over 50% of the Valladolid problems),
do not display similar knowledge of the IOI set (only 8 problems are discussed and only

228

Reviews, Comments

one is pre-2008). This is not a problem in terms of presenting the material. In the terser
algorithmic sections where no problems are discussed, and a good IOI task exists, it
appears to be a choice of the authors since not only do good ICPC tasks exist but they are
used by the authors in the corresponding problem lists. Furthermore, the bias means that
the listed problems can be submitted to an online judging system which is invaluable. One
does feel however, on occasion, as though the existing text has been hacked to include
IOI references rather than having had it in mind from the beginning.
The majority of the book discusses algorithms (and data structures) though it does
kick off with an introductory section on competitive programming. It makes a good introduction and one that might usefully be read by all students entering such contests, even
those for whom most of the material in the rest of the book would be too advanced. It is
debatable whether Type Code Faster! should have been the first tip for the vast bulk
of competitors this is not going to be an important factor but the remaining tips are
sensible and strong.
The book has two clear audiences; coaches and students. For the former the book provides a comprehensive set of categorised problems, and a far wider selection of problems
that most are likely to have familiarity with themselves. The latter get the same list, as
well as a good sampling of applicable algorithms. The book could be given as is to
the more talented or motivated students, though for many the guidance of a coach along
with exposure to other material would be advisable. It is not a substitute for an algorithms textbook but it is an excellent accompanying book and one which comes highly
recommended.

References
1. https://sites.google.com/site/stevenhalim/
2. http://uva.onlinejudge.org

Richard Forster

Olympiads in Informatics, 2012, Vol. 6, 229231

229

Teaching Programming
for Secondary School:
A Pedagogical Content
Knowledge Based Approach
Author: Mara SAELI
Publishing house: Eindhoven University of Technology
Country: The Netherlands
Year of edition: 2012
Language: English
Number of pages: 164
ISBN: 978-90-386-3084-7

In this book (a PhD thesis) Mara Saeli reports her research about teaching programming in secondary school with the aim of portraying the Pedagogical Content Knowledge of this subject", as is mentioned on the backside of it. The reason for a review in
this journal is that it contains a meticulous description and analysis of why teaching of
programming in secondary school is important, what should be taught, what problems
and limitations students have while learning programming and how the teaching could be
performed. All these issues have a great relevance for teachers who prepare students for
Olympiads in informatics and for researchers who investigate this Olympiads. Problem
solving, perhaps one of the most important aspects for Olympiads participants turns out
to be a very relevant part of the teaching and learning of programming.
The study of Saeli takes Pedagogical Content Knowledge (PCK) as a starting point.
PCK is defined by Shulman (1986): it is the amalgam or combination of knowledge of a
teacher of the content of a specific topic and the knowledge of the pedagogy of it: giving
insights into educational matters relative to the learning and teaching of a specific topic.
The PCK of a teacher starts during the education as a teacher in a particular field and
grows by enhancing his theoretical background and teaching experiences. Teachers with
good PCK are teachers who can transform their knowledge of a topic into something
accessible for the learners.
In order to operationalize the concept of PCK for programming in secondary school,
the author used the reformulation of PCK by Grossman (1990), namely the answers to the

230

Reviews, Comments

four key questions: why to teach . . ., what to teach . . ., learning difficulties of . . ., and how
to teach . . .? Saeli started to answer these questions by performing a broad literature review with respect to programming in secondary school. She found the following answers.
Why: programming enhances students problem solving skills and offers them a
learning environment which includes aspects of different disciplines, gives opportunities to use modularity and transferability of the knowledge and/or skills, and to
work with a multi-disciplinary subject.
What: a list of concepts/aspects which a programming curriculum should include,
for instance knowledge of data, instructions and syntax of a programming language, but also primitive expressions, means of combination and of abstraction,
strategies, and programming sustainability which refers to the ability to create user
friendly and attractive software that takes care of ethical and privacy issues.
Learning difficulties: difficulty to instruct the machine about the solution of a problem, the tendency to converse with a computer as if it was human, the tendency the
students maintain a local, limited point of view, failing to find a suitable solution.
How: for instance offering a simple programming language so students can focus
on the syntax, carefully choosing a diversity of problems in order to focus the
students on algorithmic thinking.
In the literature these answers are not very well connected to each other. So, it was
necessary to go into more detail with respect to each commonly taught topic. To this purpose Saeli organized six workshops in four different countries (Lithuania, Italy, Belgium
and the Netherlands) with about five experienced informatics teachers in each workshop.
Each workshop started with the question: What are the core concepts (Big Ideas") of
programming? Every participant answered this question individually. During the second
part of the workshop the participants discussed one of more of the formulated Big Ideas
using the following eight questions as a guideline:
1. What do you intend the students to learn about this Big Idea?
2. Why is it important for the students to know this Big Idea?
3. What else do you know about this Big Idea (and you dont intend students to know
yet)?
4. What are the difficulties/limitations connected with the teaching of this Big Idea?
5. What do you think students need to know in order for them to learn this Big Idea?
6. Which factors influence your teaching of this Big Idea?
7. What are your teaching methods (any particular reasons for using these to engage
with this Big Idea?
8. What are your specific ways of assessing students understanding or confusion
about this Big Idea?
Question 1 is about the why, question 2 about the what, questions 3, 4, 5, 8 about the
difficulties, and questions 6, 7 about the how. This method to obtain data as a basis for
portraying the PCK are based on CoRe (Content Representation), developed by Loughran
et al. (2004) to give a narrative account providing an overview of how teachers approach
the teaching of a specific topic in science in secondary school.

Reviews, Comments

231

The results of the workshops turned out to be the following seven Big Ideas: control
structures with focus on loops, data structures, arrays, problem solving skills, decomposition, parameters and algorithms. These results are in line with the research literature.
The next part of this study was to find out if and how the Dutch textbook support
teachers with respect to the teaching of this seven Big Ideas. The reason for performing
this and the following part of study is that the Dutch informatics teachers have almost
all originally another disciplinary background and got a license for teaching informatics
by a re-educating programme focused on informatics Content Knowledge, comparable to
one year of a university Bachelor study in informatics. Using the developed portray of the
PCK of programming as a referential framework the conclusion is, in most general terms,
that the textbooks are helpful as far as the Content Knowledge is involved, but failed for
the Pedagogical Knowledge.
The last part of the study was an investigation among the Dutch secondary informatics
teachers: how do they assess their own PCK of programming? Here again the developed
portray of the PCK was the basis for developing the online questionnaire that served as
the research instrument. Saeli argued convincingly the validity of this instrument. About
a quarter of all the almost 350 Dutch secondary informatics teachers filled in the questionnaire, but only 69 did so completely. (The reader should know that informatics is
an elective course in the curriculum of the upper part of the Dutch secondary schools
preparing to higher vocational and university education). The outcome of this study confirmed that the Dutch secondary informatics teachers have, in general, a poor PCK as
a consequence of the fact that that they are re-educated teachers form other disciplines.
As for their Pedagogical Knowledge, it turns out that this is sufficient, but not for extracurricular topics this is at least problematic for a fast developing subject as informatics.
This conclusion is of course closely related to the fact that their Content Knowledge must
be qualified as low. These findings are even more problematic because of the conclusion
that the Dutch textbooks do not really support the teachers with respect to the Pedagogical
Knowledge.
Saeli is maybe not the first researcher who tried to portray the PCK of programming,
but certainly she is the first who has done so in a systematic way. The four chapters of
her thesis which report her studies are in the meantime all published in peer reviewed
scientific journals. Interesting for the readers of this journal is her recommendation to
use tasks borrowed from Olympiads or designed analogue to Olympiad tasks in order to
improve the teaching of programming in secondary schools.

References
Grossman, P.L. (1990). The Making of a Teacher: Teacher Knowledge and Teacher Education. New York,
Teacher College Press, Columbia University press.
Loughran, J., Mulhall, P., Berry, A. (2004). In search of pedagogical content knowledge in science: developing
ways of articulating and documenting professional practice. Journal of Research in Science Teaching, 41(4),
370391.
Schulman, L. (1986). Those who understand: knowledge growth in teaching. Educational Researcher, 15, 414.

Bert Zwaneveld