Journal of Research on Technology in Education, 39(4), 433–453

Perceptions of Open Source Versus

Commercial Software: Is Higher
Education Still on the Fence?
Shahron Williams van Rooij
Datatel, Inc.

Abstract
This exploratory study investigated the perceptions of technology and academic decision-
makers about open source benefits and risks versus commercial software applications. The
study also explored reactions to a concept for outsourcing campus-wide deployment and
maintenance of open source. Data collected from telephone interviews were analyzed,
emergent themes identified, and a model of differentiators of open source versus commercial
software was created, which was then used to evaluate reactions to the outsourcing concept.
Interviews revealed perceived barriers to open source adoption and the extent to which the
outsourcing concept could alleviate risks. Recommendations for overcoming adoption barriers
are offered and future research opportunities identified to ensure that open source software
applications are both technically efficient and supportive of engaged learning. (Keywords:
open source, adoption, outsourcing, benefits vs. risks, software perceptions.)

INTRODUCTION
As institutions of higher education try to reconcile tight funding with the
rising costs of technology, some institutions are turning to open source—
software delivered with its computer program source code—for campus-wide
applications such as course management systems and administrative systems.
With access to source code, developers can modify the software to meet the
needs of the institution, save the license fees charged by commercial vendors,
and provide the institution with the flexibility to build learning environments
that are both pedagogically sound and technically efficient (Pavlicek, 2000;
Weber, 2004; Williams, 2002). The rollout of Sakai, an open source platform
for teaching, learning, and research collaboration (Sakai, n.d.), the endorsement
of the Kuali open source financial management system by the National
Association of College and University Business Officers (Kuali, n.d.), and the
growing number of institutions worldwide adopting the Moodle open source
course management system, have all contributed to the higher education “buzz”
surrounding open source.

Theoretical Framework: A Cross-Disciplinary Perspective

The chief technology officer or chief information officer (CTO/CIO) is
responsible for the information technology infrastructure of the institution and
is the primary decision-maker when deciding how and when to acquire the
hardware and software that will be used across all departments and divisions of
the institution (Green, 2004). This includes the enterprise-wide administrative
software applications that automate the business functions of the institution
and course management systems (CMS), which offer faculty and students

Journal of Research on Technology in Education 433

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
Web tools to complement or replace classroom seat time. Typically, a CMS
will include curriculum design tools and templates, automated testing and
scoring tools, student tracking, calendars, and a variety of tools (discussion
forums, chat, whiteboard, etc.) for faculty-to-student and student-to-student
communication (Gibbons, 2005). The CIO also makes the decision to use open
source versus proprietary vendor products.
Recently, the chief academic officer (CAO) has become an equal partner with
the CIO with respect to the selection of enterprise-wide software applications
for teaching and learning (Green, 2004). As the administrator responsible for
the institution’s instruction and research affairs, the CAO assesses the extent
to which a particular technology meets or does not meet pedagogical needs.
Consequently, any exploration of open source and the learning environment
requires drawing upon two bodies of literature—software engineering
and education—to determine the theoretical framework in which such an
exploration would take place.

The Software Engineering Literature

Open source advocates point to an extensive body of research in the field
of software engineering that explores the benefits and risks of open source in
the context of (a) social movement theory and appeals to the common good
(Coleman, 2004; Franck & Jungwirth, 2003; Kelty, 2004; O’Mahoney, 2002;
Perens, 1999), (b) a new paradigm in software development methodology,
where developers participate without monetary compensation (Evans, 2002;
Gacek, Lawrie, & Arief, n.d.; Raymond, 2001; Scacchi, 2001; Stewart &
Gosain, 2004; Von Krogh, 2003), and (c) security and risk management
(Raymond, 2001; Stallman, 1999; Weber, 2004). Other conceptual frameworks
in the literature include the Diffusion of Innovations theory first developed by
Rogers (1995), applied to the adoption of technology by Moore (1991, 2005),
then applied to the adoption of open source software, with technical skills as a
critical barrier to adoption (Attewell, 1992; Evans, 2002; James, n.d.; Linden
& Fenn, 2003). Organizational know-how and ability to respond to innovation
has also been the basis for framing open source adoption (Au & Kaufmann,
2003; Overby, Bharadwaj, & Bharadwah, n.d.), while public policy think tanks
have published monographs about U.S. government policy toward open source
(Hahn, 2002; PITAC, 2000).
Higher education technologists and financial administrators —particularly
those in the doctoral/research institutions—are vocal advocates of the
efficiencies of open source. For example, in 2004, the National Association
of College and University Business Officers (NACUBO) conducted a survey
of its members about their perceptions of open source software. Nearly half
(46%) see open source as a viable alternative to commercial software systems,
with the primary reasons being “open standards and interoperability with other
application systems” (61%); “freedom to modify the code” (58%); “software
designed by and for the industry” (58%), and; “lower cost of ownership” (55%).
Although the study’s author (Hignite, 2004) acknowledges that these results are
based on the responses of less than 5% (n=257) of the NACUBO membership,

434 Summer 2007: Volume 39 Number 4

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
the buzz surrounding the survey results were enough to spur the organization to
apply for a grant to explore the feasibility of an open source financial system.
In the July 2006 issue of the trade publication Campus Technology, the CIO
of a private masters institution with a higher education publication (HEP)
enrollment of 3,500 purports to have saved 20% in annual maintenance costs
by switching from vendor products to open source software (Villano, 2006).
What is not stated is what type of open source software (infrastructure level
vs. application level) the institution uses or the number and skill set of the
institution’s programmers. In a recent survey of 195 CIOs conducted by the
IMS Global Learning Consortium (Abel, 2006), more than half (57%) of
the respondents purport to have seen cost-of-ownership improvements since
replacing vendor products with open source. However, the applications adopted
include a mix of desktop (e.g., MyOffice), user interface (e.g., uPortal), and
teaching/learning applications (e.g., Moodle), all with adoption rates of ≤ 24%
of the total survey sample.
The software engineering literature is unanimous in deeming open source
software to be for technologists; the overarching assumption is that the
technologist IS the end user. However, this is not the case for open source
applications intended to support teaching and learning. Moreover, the software
engineering literature offers no insights on the extent to which open source
enables the incorporation of sound pedagogy into the construction of the
learning environment.

The Education Literature

International education has been quick to research and adopt open source
software for the development of instructional content and delivery systems.
Researchers in the UK note that education already has a long-established legacy
of shared responsibility for projects, peer review and distributed development,
and cite projects aimed at open source development of educational materials
with a centralized repository of freely-available, searchable quality controlled
materials (Carmichael & Honour, 2000; Hirst, 2001). There are also studies
that provide European examples of open source course management systems
that incorporate sound pedagogy (Leinonen, T., Hakkarainen, K., Appelt, W.,
Dean, P., Gomez-Skarmetav, A., Ligorio, B., et al., 2001; Dunlap & Wilson,
2002; Jasinski, n.d.).
In the U.S., the collaborative development and sharing of instructional
content is not new. The TLT Group (TLT, n.d.) has been offering online
assessment tools, Web page templates and tutorials to institutions on a low-cost
subscription basis since 1992. MERLOT is one of the longest-running Web
sites providing free materials from faculty for faculty in a variety of disciplines
(MERLOT, n.d.). MIT’s Open Course Ware Web site is a more recent example
of free access to instructional materials. The materials are targeted to educators
and learners worldwide, with the biggest market outside the United States and
the primary use being the enhancement of personal knowledge (Carson, 2002).
It is not yet clear to what extent these materials have been adopted by U.S.
learners, institutions, and faculty.

Journal of Research on Technology in Education 435

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
 Incorporating technology into instruction has meant re-examining the nature
of the environment in which students learn, including the setting or “space”
in which learning is fostered and supported (Wilson, 1996). Perkins (1992)
parses a learning environment into five components: (a) information banks that
include information repositories of all types, ranging from textbooks to digital
media to faculty; (b) symbol pads or surfaces for manipulating symbols and
languages and that include notebooks, word processors, database programs, and
the like; (c) construction kits or packaged collections of content components for
assembly and manipulation. Examples would include Legos or digital authoring
tools; (d) phenomenaria or areas for presenting, observing, and manipulating
phenomena, such as the well-known SimCity; and (e) task managers, the
elements of the learning environment that set tasks, provide guidance, feedback,
and changes in direction. This component has traditionally been the faculty
member, but also includes electronic task managers and intelligent tutor
computer-based programs.
The interplay of these components is part of the constructivist perspective
on learning. Constructivism holds that learners build their own personal
interpretation of the world based on their own experiences. Because there are
many ways of structuring the world and its entities—i.e., there are multiple
perspectives—learners make their own meaning rather than having it imposed
from some external, independent reality (Duffy & Jonassen, 1992). Thus, the
pedagogical goals of a learning environment grounded in constructivism must
provide these multiple perspectives and enable learners to make their own
meaning by providing multiple modes of representation (Honebein, 1996).
Technology’s potential lies in the extent to which it enables the creation of
a learning environment grounded in constructivism (Domine, 2006). There
is evidence that commercially developed course management systems can
accommodate a wide variety of learning activities and perspectives. Studies
of individual courses using WebCT or Blackboard have found that those
technologies augmented the learning experience by enabling collaboration, the
development of a strong sense of community, and the inclusion of constructivist
strategies of collaborative learning into the instructional environment (DeNeui
& Dodge, 2006; Gill, 2006; Iyer, 2003). In a recent review of the literature
on the use of course management systems in higher education, Papastergiou
(2006) states that although students and faculty have positive attitudes toward
CMS-based learning and that faculty can apply participatory, constructivist
approaches to learning in CMS environments, increased faculty workload,
limited assessment capabilities, the inability to support subjects that involve
hands-on tasks, and the need for more sophisticated collaborative facilities
beyond the commonly-offered discussion boards are all weaknesses in
commercial course management systems. However, the education literature has
not yet systematically addressed the extent to which open source enables sound
online pedagogy.
A New Approach
Figure 1 illustrates the themes from the software engineering literature and
the education literature. What is clear is the gap between what the technologists

436 Summer 2007: Volume 39 Number 4

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
Figure 1. Open source theoretical framework
need/want and what academics need and want for teaching and for scholarly
work. Consequently, there is a clear need for systematic analysis of the benefits
and risks of open source software from the perspective of the CIo and from
the perspective of the CAo. The more we understand the similarities and
differences in perceptions of these two decision-makers, the more complete
the picture of the impact of open source on both teaching and learning and
technical efficiencies. Moreover, there is a need to better understand the extent
to which commercial organizations can assist institutions interested in adopting
open source, so that alternative cost of ownership models can be developed and
institutions can make informed decisions about where to use their resources.

RESEARCH QUESTIOnS
Two research questions guided this study:
• What are the characteristics that CIos and CAos believe differentiate
open source software from commercially developed software?
• When presented with a concept for outsourcing the integration and
maintenance of campus-wide open source academic systems, how
interested would CIos and CAos be in purchasing this service and at
what price?

METHOD
Participants and Setting
A pool of 45 individuals representing a variety of Carnegie classifications
and institution sizes had volunteered to participate in this exploratory study

Journal of Research on Technology in Education 437

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
 Table 1: Distribution of In-depth Interviews by Carnegie Classification
Carnegie Classification CAOs CIOs
Doctoral/Research 0 3
Masters 4 4
Baccalaureate 2 0
Associates 3 3
Specialty 1 0
Total Number of Completed Interviews: 10 10

by filling in their particulars in a call-for-volunteers question on a survey of

open source adoption practices this author conducted in the winter of 2006.
Of this pool of 45, 10 CIOs and 10 CAOs were interviewed, for a total of 20
completed in-depth interviews. Participants were selected based on Carnegie
classification and institution size, to get as broad a representation as possible.
Table 1 shows the distribution of interviews by Carnegie classification.

Procedure
Each participant was contacted by e-mail and offered a choice of dates and
times for the interview. Once the participant stated his/her availability for an
interview, an e-mail confirmation containing an informed consent form was
e-mailed to the respondent, with a request to reply to the confirmation to
indicate consent. The interviews took place via phone during business hours.
The tape-recorded interviews averaged 20 to 30 minutes in length.
Validity and Reliability
The validity threats associated with qualitative research (Maxwell, 1996),
along with the measures used in this study to address those threats are as
follows:
• Valid description or inaccurate/incomplete data. This threat was
addressed by the audio recording and verbatim transcription of the
interviews.
• Researcher bias or the imposition of the researcher’s own framework or
meaning on those of the participants. This was addressed during the
interview by corroborating what the researcher thought was said with
the participant.
• Theoretical validity threatened by not collecting/considering discrepant
data/explanations/understandings. Documentation of each step of the
analysis using the NVivo software program automatically developed
a log of code constructions and proposition testing. Documentation
is a method of improving the immediate analysis task being carried
out, enhancing the sophistication of later analyses, and strengthening
confidence in the final conclusions (Miles & Huberman, 1994).
This study sought to flesh out general constructs, patterns, and themes to
obtain insights into the mindset of the participants and better understand some
of the ongoing conversations around open source adoption in higher education.

438 Summer 2007: Volume 39 Number 4

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
As such, the outcomes are not representative of the entire population of higher
education institutions.
Data Analysis
After the data were collected, the audio tapes transcribed and uploaded into
the NVivo software program, a preliminary exploratory analysis (Creswell,
2002) was conducted. This analysis involved reading through all of the
transcripts, jotting down ideas and obtaining a general sense of the data. Using
the software’s visual coding capabilities, words, phrases, and activities that
seemed to be similar were grouped into categories or themes. Emergent themes
were identified and given tentative names or labels that served as the foundation
for preliminary analyses. The software enabled the construction of an audit
trail (Hoepfl, 1997) or schema that maps the themes with their speakers and
context, so that multiple perspectives on a given theme could be captured and
identified. Further analysis enabled themes to be layered and interconnected, to
flesh out major and minor themes, and obtain broader as well as deeper levels of
abstraction (Creswell, 2002).

RESULTS
Differentiators of Open Source Versus Commercial Software: Context
As a first step to understanding how CAOs and CIOs compare and contrast
open source versus commercial software, each interview participant was
asked about his or her familiarity with and usage of industry terminology,
specifically, the terms “open source,” “collaborative development software,” and
“community source software.” All 20 participants were most familiar with the
term “open source,” volunteering that “open source” meant use by anyone free
of license fees, with access to the source code via the Internet, and the ability to
share additions/modifications with others.
Participants were then asked about publications, conference events or other
resources used to become knowledgeable about open source. Common to
both the CAO and CIO participants was the reliance on presentations and
publications from the EDUCAUSE organization, with 10 of 20 participants
mentioning EDUCAUSE as their primary knowledge source. Beyond
EDUCAUSE, both CAO and CIO participants relied on individuals with
technical knowledge, including IT faculty, either at their own institutions or at
other institutions, as well as local and/or regional higher education consortia
and professional associations. CAO participants in particular read articles in
The Chronicle of Higher Education and University Business dealing with open
source in higher education. On the whole, however, CAO participants looked
to the technologists to keep them informed about the value of open source for
enhancing teaching and learning. In explaining the reliance on technologists,
a CAO participant at a community college with a HEP enrollment of 2,600
stated succinctly: “I talk to IT. We’re all too busy to be doing each other’s jobs.”
CIO Perceptions
Only three of the 10 CIO participants have already made a conscious decision
not to adopt open source. All three of these non-adopters are from private,

Journal of Research on Technology in Education 439

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
masters institutions with HEP enrollments in the 3,000-6,500 range. Two of
the non-adopters mentioned the skill set of their technical staff as the reason for
non-adoption, while all three non-adopters cited overall satisfaction with their
commercial product as a reason for non-adoption.
IT staff skill set remained the key factor when discussing the benefits and
risks of open source versus commercial products. All three of the non-adopters
voluntarily mentioned the freedom from vendor license fees as the attraction of
open source. However, all three also stated that the need to hire programmers
to supplement their IT staff would raise the total cost of ownership (TCO) of
open source to levels that would probably exceed what they are currently paying
to their commercial vendors. To illustrate his point, one non-adopter disputed
the idea that open source was a savings, stating:
You can tell me that open source is free or you can get access to
it at nominal fees, but what does it take to maintain it in your
environment, to do releases, to do support, to enhance it if you opted
to enhance? When I’ve listened to some of the conference calls and I
hear the resources that are being invested by some of the universities
to operationalize some of the tools, I don’t have any vision that open
source comes and it’s a freebie. (CIO, masters institution, HEP 4,826)
Institutional resource constraints and skepticism about the true cost of
ownership of open source software were not unique to these three non-
adopters. The seven other CIO participants also mentioned maintenance
and support as a risk associated with open source software. However, these
participants remain open to adopting open source software and are currently at
various stages ranging from assessment to pilot testing to deployment of specific
software applications. What differentiates the undecided participants from the
non-adopters are the direct experiences of these participants with vendor price
increases and mergers, creating fear, uncertainty, and doubt about the financial
and human resource costs of staying with vendors. Typical of undecided CIO
comments about the fear, uncertainty and doubt about the financial and human
resource costs associated with vendor products versus open source are these
comments:
WebCT is the platform that we’re on right now for our learning
management or course management system. They’ve been bought by
Blackboard and it seems like that there’s a collapsing of the off-the-
shelf course software applications out there, so we want to have all
options open so we don’t just have to pay what Blackboard asks us to
pay each and every year. (CIO, masters institution, HEP 2,600)
Lack of support, security, and questionable longevity were the common risk
themes. However, CIO participants appeared to view risk management as part
of their job. CIO participants also point to functionality gaps in the vendor
software, suggesting that CIO’s are indeed working collaboratively with their
CAOs when evaluating open source alternatives to a commercial CMS. Typical
CIO comments about commercial software functionality gaps are these:

440 Summer 2007: Volume 39 Number 4

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
 I think budget and instructional flexibility are the two big ones. Over
the years our usage of Blackboard has increased by our faculty and
students. We went from a basic license to an enterprise license and
that escalated our cost by about six times in one fell swoop, and we
anticipate that that’s gonna get worse since Blackboard and WebCT
merged. There are so very few players in the commercial learning
management systems space that they really do have the capacity to
charge whatever they please. From a faculty standpoint, they find that
their ability to develop customized components is greatly enhanced
for those people who are interested, so that if you do have a faculty
member that has some good programming skill and has an interest
in investing their time in doing things that are unique, and then, of
course, sharing with the community at large, that has a very attractive
feature that most commercial applications don’t offer in the same way.
(CIO, Associates institution, HEP 15,655)

CAO Perceptions
When asked about the business issues driving their institution’s consideration
of open source, the CAO participants voluntarily mentioned the financial
and human resource costs associated with commercial products. However, the
CAOs appear to be looking at costs more from a value perspective—i.e., the
functionality received versus the dollars spent—rather than at the total cost of
ownership perspective of their CIO counterparts. Typical CAO comments are
as follows:
Probably the main thing that drives that—there are really two.
One is the cost management. The problems that we’re having with
Blackboard/WebCT right now across the state are very rapidly
escalating problems. We don’t see open source as a cost savings. It just
means that the costs that we put in can be directed to things that we
need to see done with the applications. You know, right now our costs
for Blackboard/WebCT go up and we’re not seeing a whole lot of
attention to what we perceive as needs at my local campus or across the
university system. (CAO, masters institution, HEP 5,027)

It’s primarily the licensing dollars. And there are some functional
reasons that I think we’re interested in looking at some different
kinds of software because as we go more and more into the online
environment, we’re bumping up against restrictions of Blackboard,
and we’re not happy about that. The biggest restriction we have is we
want to do videos and we want do conferencing type things. You run
into restrictions with Blackboard in those regards. (CAO, masters
institution, HEP 4,200)
Perceptions of the pedagogical fit of open source versus commercial systems
are well stated by this CAO of a baccalaureate institution with a HEP
enrollment of 1,591:

Journal of Research on Technology in Education 441

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
 The benefit for small schools is the ability to customize basically to
get what you need and not what you don’t. So what’s happening now,
and whether or not it’ll be successful is another question, is that you
get places like Mellon and other funding agencies interested in the
health of higher education, and also in staying on top of technology
in a cost efficient way, that are promoting ways for segments of higher
education, notably liberal arts colleges, to work together to get what
they need and enhance teaching, and make outcomes demonstrable
internally and externally. And you’re not really going to do that with
Blackboard.
Figure 2 provides a graphic illustration of the themes differentiating open
source software from commercial software and the relationships among those
themes from the CIO and CAO perspectives. For CIOs, vendor experiences
reinforce negative perceptions of the cost of vendor software, which in turn,
contributes to a negative perception of the value of commercial software as
defined by the software’s fit with desired needs/functions. CAOs perceive
open source as the counterpoint to that functionality gap, driven by IT faculty
knowledge, the availability of funding for projects focused on pedagogy and
the opportunity to collaborate with peers. The net benefit is flexibility, with
control over one’s own destiny as the end benefit. The potential risks posed by
security, the lack of support, and quality control contribute to perceptions of
questionable project longevity. The threshold that enables perceived benefits to
outweigh perceived risks is the knowledge base of the institution’s own internal
IT staff.

Reactions to an Onshore Outsourcing Concept

In order to flesh out any existing biases toward using external talent,
participants were asked if they were using any outside companies or consultants
to help with the evaluation or deployment of open source software applications.
Although only one participant was currently using a consultant company (to
deploy the Sakai learning management platform), six of the 20 participants
had experience with consultants helping to deploy their administrative software
systems, with mixed results. In all cases, satisfaction with the consultant’s
work was countered strongly by cost overruns. With reference to using
consultants specifically for open source software, 19 of the 20 participants felt
that they were not far enough along in the evaluation process to bring in an
outside consultant. Instead, they preferred to rely on the knowledge of peers
knowledgeable about the subject. This is consistent with comments earlier in
the interview when participants mentioned the higher education conferences,
presentations and networking contacts used to become knowledgeable about
open source.
During the second half of the interview, participants received an e-mail
message describing a concept for a package of consulting and training services
aimed at assisting institutions in deploying open source software for teaching
and learning (see Table 2, p. 444) and were asked if they had seen or heard
anything like it. Although no one had seen or heard of a set of services exactly

442 Summer 2007: Volume 39 Number 4

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
 RISKS

Figure 2. Model of differentiators of open source software versus commercial vendor software.
BEnEFITS

Journal of Research on Technology in Education 443

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
 Table 2: Open Source Services Concept
Concept Statement
Company X’s open source management services will provide academic deans,
department chairs, and IT staff with a comprehensive suite of services for the
development, deployment, and maintenance of open source software applications
supporting teaching and learning. The services include:
• Academic and administrative system needs assessment
• Current resource allocation and deployment
• Migration from and/or integration with legacy and vendor systems
• Faculty and IT Help Desk planning
• Education and training for faculty and IT support staff

Company X’s open source management services will be differentiated from other
professional services on the market today by incorporating both the pedagogical
needs of the academic units of the institution with technology goals and objectives
of the institution.

The services will be offered for a fixed, predetermined period at a fixed,

predetermined cost.

like the ones described in the concept, one CAO participant and one CIO
participant stated that they were familiar with similar services, primarily for
deploying administrative systems.
When asked whether or not they would be interested in a service like the
one described in the concept, 15 of the 20 participants stated that they would
be interested. However, the interest was cautious at best. Interest was largely
bounded by the option of picking and choosing the services they needed
rather than taking all of the services as a package. This “Chinese menu”
approach would also help keep down costs, a point of concern when hiring any
consulting firm.
The one service desired by all 15 of the interested participants (CAOs and
CIOs) was the migration from and/or integration with legacy and vendor
systems, primarily the institution’s administrative software system. Services
desired specifically by the interested CAOs were the academic needs assessment
and faculty education and training. The five participants (4 CAOs and 1 CIO)
who were not interested in the concept were disinterested for a variety of
unrelated reasons.
Regardless of whether or not they were interested in the services described
in the concept, most of the participants had a fairly realistic idea of the length
of time that would be required for the services. For the CAOs, who viewed
faculty education/training, academic needs assessment and systems integration/
migration as the most interesting components of the concept package
(regardless of whether or not they would actually contract for those services),
the estimated length of the service engagement was 12 months or longer. CIOs
based their estimates on the time it took to implement their administrative
systems, with data migration and systems integration alone taking 6–9 months

444 Summer 2007: Volume 39 Number 4

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
 Table 3: Reactions to an Onshore Outsourcing Concept Among CAO
(N=10) and CIO (N=10) Interview Participants
Reactions CAOs CIOs
Interested in 6 9
Concept
Preferred services Academic and administrative
needs assessment
Current resource allocation
and deployment
Migration from and/or Migration from and/or
integration with legacy and integration with legacy
vendor systems and vendor systems
Education and training for Education and training for
faculty IT support staff
Not interested in 4 1
concept
Reasons for Technology decision/CIOs
disinterest domain
Need to first educate senior
leadership
Philosophically opposed to
outsourcing
No single organization can
do it all
Expected duration ≥ 12 months ≥ 12 months, of which 6-9
of services months is data migration
engagement
Expected • One-time • Annual
investment in • $10-$30K (integration) • $20K per service
services • $100-$500K (full suite)

to complete. Only one participant—the CAO of a small (HEP = 289) specialty

institution—had no idea how long it would take to implement any of the
services described in the concept.
Estimating the cost of the services described in the concept proved to be a bit
more challenging for the participants, particularly for the CAOs. CAOs tended
to draw on the implementation costs paid to course management/learning
management vendors like Blackboard/WebCT as the frame of reference and
placed the data migration/systems integration component of the concept in the
$10,000–$30,000 range. CAO estimates for the full suite of services described
in the concept ranged from $100,000 to $500,000. CIO participants used their
administrative software implementation costs and/or previous experiences with
external consultants as the frame of reference for estimating the costs of the

Journal of Research on Technology in Education 445

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
services described in the concept. Most were fairly consistent in estimating the
costs in terms of an annual investment rather than a one-off charge and most
expected to pay $20,000 a year for each of the services described in the concept.
Table 3 (p. 445) provides a summary of participant reactions to the concept,
including similarities and differences between the CAOs and the CIOs.
Finally, interview participants were given the opportunity to share any
additional thoughts and comments about open source software. All of the
participants acknowledged that the jury is still out—and probably will be for
some time—on the promise of open source in terms of controlling one’s own
destiny. How far they are willing to test the open source waters has depended
thus far on how they view their internal capabilities, both from a technical and
a financial perspective.
Summary
For both CAO and CIO participants, commercial software implementation
experiences and vendor price perceptions have reinforced negative perceptions
about the cost of vendor software which, in turn, has contributed to a negative
perception of the value of commercial software in terms of its fit with desired
needs and functions. Reinforced by IT faculty knowledge, the availability of
funding for projects focused on pedagogy, and the opportunity to collaborate
with peers, open source is perceived as the solution to the functionality gap, the
end benefit of which is control over one’s own destiny. These plusses outweigh
the potential deltas posed by security, the lack of support and questionable
longevity if (and only if ) the knowledge base of the institution’s internal IT staff
is deemed to be able to meet these challenges. Finally, the outsourcing concept
generated a good deal of interest mixed with a healthy dose of skepticism
grounded in previous experiences with external consultants and implementation
cost overruns. The concept achieved greater acceptance if available as a menu
of consulting services rather than as a fixed package. Further, enthusiasm was
slightly less among the CAO participants, some of whom perceived the decision
about open source consulting services as being the domain of their CIO
counterparts. Nevertheless, most of the interview participants were intrigued
with the ideas behind the concept.

RECOMMENDATIONS
The results of this study call for a deeper, more comprehensive look at
the value of open source for enhancing teaching and learning, as well as for
achieving technology efficiencies. One avenue is to maximize the CAO-CIO
partnership. Study findings show that CAO participants rely on individuals
with technical knowledge—either at their own institutions or at other
institutions—as well as presentations and publications from EDUCAUSE, to
learn about open source software applications and receive recommendations
about the value of open source for enhancing teaching and learning. CAO
reliance on technology counterparts is based in the firm belief that it is the
CIOs job to inform and recommend. What is not clear is the point in time at
which CIOs should begin liaising with their academic counterparts. The case

446 Summer 2007: Volume 39 Number 4

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
could be made that CIO evaluation of open source software applications is not
yet far enough along to merit bringing their CAO partners to the table. This
line of thinking recalls the discussion in the software engineering literature
(Behlendorf, 1999; Evans, 2002; Gacek et al., n.d.; Glass, 2003) about the gap
between the technologist, who is the end user of infrastructure-level software
such as computer operating systems, and the non-technologist, who is the end
user of application-level software such as course management systems. In their
evaluation of open source software in higher education, Courant and Griffiths
(2006) affirm the need for mutual understanding between users and developers
about how the software is to be used and what is important for it to accomplish.
If, at the end of the day, it is the technology professional who determines
what instructional tools will (not) be supported, how can the institution ensure
that technology remains in the service of pedagogy, and not the other way
around? To ensure that open source teaching and learning applications support
the institution’s pedagogical needs, CIOs need to liaise with their academic
counterparts long before the deployment phase. Constructivist-based use case
scenarios that describe how students interact with faculty, with course materials,
with external resources, and with their fellow students, need to be laid out
early in the evaluation process. CAOs should have input into the evaluation
process timeline, so that the academic calendar and key institutional dates (e.g.,
registration, final exams, etc.) are taken into account when soliciting CAO,
faculty, and student feedback on software pilot tests. If, as CIOs contend, the
CAO and the CIO are partners in the process of making decisions around
the purchase and deployment of enterprise-wide applications for teaching
and learning (Green, 2004), then CIOs need to evaluate open source software
applications that impact teaching and learning holistically and include their
academic counterparts, rather than making the go/no go decision solely on
technical merits first, then looking at building functionality after an adoption
decision is made.
Another avenue concerns the cost of open source software. Although the
contribution of open source to the student’s learning experience is the primary
driver for CAO consideration of open source, CAOs voluntarily discussed costs
when asked about the business issues underlying consideration of open source
software. However, CAO participants were discussing cost in the context of
functionality received. To better understand the true cost-value proposition
of open source requires institutions that have already fully deployed open
source software applications to document and publicize the level of effort spent
on open source deployment and maintenance at their institution. From a
technology perspective, the number of IT staff supporting each application, the
skill set of that staff, salaries and number of hours spent on open source versus
basic institution operations (hardware maintenance, etc.), as well on faculty
and student training and support, would all have to be reported. This would
enable institutions to get a more complete picture of the budgetary impact of
open source and whether the savings in software license fees is a true savings. As
Villano’s (2006) article notes, with only the top 300 U.S. institutions possessing
the human resources necessary to implement software and possessing the skills

Journal of Research on Technology in Education 447

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
necessary to manipulate source code, institutions that have deployed open
source software applications have had to hire specialists or increase the training
budget for existing programmers to build and maintain open source software
applications. The result is no savings at all and could mean increased costs.
From a teaching and learning perspective, the level of faculty and/or third
party instructional design resource effort versus student learning outcomes
would need to be monitored and documented. Students own perceptions
of their learning gains and experiences would also have to be tracked and
reported. The bottom line: The absence of hard numbers will perpetuate what
Green (2004) terms “affirmative ambivalence” towards open source; namely,
agreement with the underlying concept, but reluctance to actually adopt and to
replace vendor product with open source solutions.
A third avenue concerns the feasibility of using commercial providers to assist
in the deployment and maintenance of campus-wide open source software
applications, particularly in terms of offering services that support sound
pedagogy. Although the study participants liked the idea of being able to pick
and choose the type of support services their institution needs to adopt open
source software solutions, there is still a gap between the CIO desire to control
costs given previous experiences with commercial vendor price increases,
implementation cost overruns, and less than stellar experiences with some
external consultants on the one hand, and on the other hand, CAO uncertainty
as to whether the academic services that they deemed interesting are part of
the CIOs decision domain or whether such services should be outsourced
at all. This suggests that the availability of open source support services may
not provide the ticket to entry that some open source advocates believe.
Commercial companies need a financial incentive to provide support for open
source software solutions, and that would require a critical mass of institutions
contracting for those services (Courant & Griffiths, 2006). For commercial
service providers, there is unlikely to be a rush to provide open source support
services —technical or pedagogical—until it is clear what price institutions
are willing to pay for control over their own destiny or to trade one master for
another, as it were.

OPPORTUNITIES FOR FUTURE RESEARCH

One way to address the limited scope of this study would be would be to
extend the research to a longitudinal study of open source software awareness,
adoption, attitudes, and impact on commercial software usage. Incremental
steps in overcoming the adoption barriers identified in this research need to
be tracked and the predictors of successful open source implementation and
deployment identified, so that institutions can map their own resources, polices,
procedures and budget constraints with the “knowns” associated with successful
adoption. Future research must also address opportunities for enhancing the
CIO–CAO partnership, so that some mechanism(s) for greater and earlier
collaboration evolve, ensuring that the open source software applications are
both technically efficient and supportive of engaged learning.

448 Summer 2007: Volume 39 Number 4

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
 Research needs to be conducted in evaluating the long-term effect of the
higher education-specific collaborative software development communities
like Sakai. The extent to which the volunteer-developers address the quality
and functionality issues of the broad base of community members, not just the
issues of their home institutions or the community founding members, needs to
be studied. Finally, there is an urgent need for research into the documentation
and dissemination of proven cost-value models. With resource limitations a
given, it is important for institutions to know what the true value of open
source solutions will be, so that they can make an informed decision as to where
they will invest their resources.

Contributor
Shahron Williams van Rooij is director of product marketing at Datatel,
Inc. in Fairfax, Virginia, where she facilitates the engineering of new e-learning
technology solutions. She earned a PhD in instructional technology at George
Mason University and has five years of college classroom teaching plus a strong
background in e-learning. As a Certified Distance Education Professional
(C.D.E.P.) credentialed by Texas A&M University, she is a frequent speaker on
the development of Web-based education and has published in journals such
as the WebNet Journal and the Journal of Interactive Learning Research. (Address:
Shahron Williams van Rooij, Datatel, Inc., 5116 Woodfield Drive, Centreville,
VA 20120; 703.502.6745; Fax: 703.968.4573; [email protected].)

References
Abel, R. (2006). The state of open source software in higher education: Time for
a reality check. Retrieved September 25, 2006, from the IMS Global Learning
Consortium Web site: http://www.imsglobal.org/articles/presentations.cfm.
Attewell, P. (1992). Technology diffusion and organizational learning: The
case of business computing. Organization Science, 3(1), 1–19.
Au, Y. A., & Kauffman, R. J. (2003, January 6). Information technology
investment and adoption: A rational expectations perspective. Paper presented
at the 36th Hawaii International Conference on System Sciences. Retrieved
February 7, 2004, from http://www.hicss.hawaii.edu/HICSSpapers/OSSC105.
pdf
Behlendorf, B. (1999). Open source as a business strategy. In C. DiBona, S.
Ockman, & M. Stone (Eds.), Open sources: Voices from the open source revolution
(pp. 149–170). Sebastopol, CA: O’Reilly & Associates.
Carmichael, P., & Honour, L. (2002). Open source as appropriate technology
for global education. International Journal of Educational Development, 22,
47–53.
Carson, S. (2002). MIT OpenCourseWare: A new model for open sharing.
Paper presented at the MERLOT International Conference, Costa Mesa, CA.
Retrieved August 23, 2004, from http://www.merlot.org.
Coleman, G. (2004). The political agnosticism of free and open source
software and the inadvertent politics of contrast. Anthropological Quarterly,
77(3), 507–519.

Journal of Research on Technology in Education 449

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
 Courant, P. N. & Griffiths, R. J. (2006). Software and collaboration in higher
education: A study of open source software. Retrieved September 26, 2006, from
the EDUCAUSE web site: http://www.educause.edu/LibraryDetailPage/
666?ID=CSD4633.
Creswell, J. W. (2002). Educational research: Planning, conducting and
evaluating quantitative and qualitative research. Upper Saddle River, NJ: Pearson
Education, Inc.
DeNeui, D. & Dodge, T. (2006). Asynchronous learning networks and
student outcomes: The utility of online learning components in hybrid courses.
Journal of Instructional Psychology, 33(4), 256–259.
Domine, V. (2006). Online pedagogy: Beyond digital “chalk and talk.”
Academic Exchange Quarterly, 10(1), 48–51.
Duffy, T. M., & Jonassen, D. H. (1992). Constructivism: New implications
for instructional technology. In T. M. Duffy (Ed.), Constructivism and the
technology of instruction: A conversation (pp. 1–16). Hillsdale, NJ: Lawrence
Erlbaum Associates
Dunlap, J. C., Young, D. L., & Wilson, B. (2002, June). Xtreme learning
control: Examples of the open source movement’s impact on our educational
practice in a university setting. Paper presented at the meeting of the ED-
MEDIA 2002 World Conference on Educational Multimedia, Hypermedia &
Telecommunications, Denver, CO. Available: http://www.editlib.org/index.cfm
Evans, D. S. (2002). Politics and programming: Government preferences
for promoting open source Software. In R. W. Hahn (Ed.), Government
policy toward open source software (pp. 34–49). Washington, D.C.: Brookings
Institution Press.
Franck, E., & Jungwirth, C. (2003). Open source software: New roles for the
market economy. Journal of Management and Governance, 7(4), 401–421.
Gacek, C., Lawrie, T., & Arief, B. (n.d.). The many meanings of open source.
Retrieved January 30, 2005, from University of Newcastle Web site: http://
www.cs.newcastle.ac.uk/research/pubs/trs/papers/737.pdf
Gibbons, S. (2005). Course-management systems. Library Technology Reports,
41(3). Retrieved August 17, 2006, from the American Library Association Web
site: https://publications.techsource.ala.org/products/archive.pl?article=2556.
Gill, T. (2006). The memory grid: A glass box view of data representation.
Journal of Information Systems Education, 17(2), 119–129.
Glass, R. L. (2003). A sociopolitical look at open source. Communications of
the ACM, 46(11), 21–23.
Green, K. C. (2004). Campus computing 2004: The 15th national survey of
computing and information technology in American higher education. Encino, CA:
The Campus Computing Project.
Hahn, R. (2002). Government policy toward open source: An overview. In
R. Hahn (Ed.), Government policy toward open source (pp. 1–11). Washington,
D.C.: Brookings Institution Press.
Hignite, K. (2004, August). An open mind on open source. Retrieved
December 1, 2004, from National Association of College and University
Business Officers (NACUBO) Web site: http://www.nacubo.org.

450 Summer 2007: Volume 39 Number 4

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
 Hirst, T. (2001). The open source teaching project (OSTP): Research note.
United Kingdom: Open University. (ERIC Document Reproduction Service
No. 474 093).
Hoepfl, M. (1997). Choosing qualitative research: A primer for technology
education researchers. Journal of Technology Education, 9(1), Retrieved June 30,
2006, from the Virginia Tech Web site: http://scholar.lib.vt.edu/ejournals/JTE/
v9n1/hoepfl.htm.
Honebein, P. C. (1996). Seven goals for the design of constructivist learning
environments. In B. G. Wilson (Ed.), Constructivist learning environments: Case
studies in instructional design (pp. 11–24). Englewood Cliffs, NJ: Educational
Technology Publications.
Iyer, H. (2003). Web-based instructional technology in an information
science classroom. Journal of Education for Library and Information Science,
44(3), 296–315.
James, C. A. (n.d.). The care and feeding of FOSS (or, the lifecycle of software
technology). Retrieved February 24, 2005, from http://www.moonviewscientific.
com/essays/software_lifecycle.htm
Jasinski, M. (n.d.). EDUCHAOS-disruptive technologies. Message posted
to electronic mailing list, archived at http://community.flexiblelearning.net.au/
GlobalPerspectives/content/article_5333.htm.
Kelty, C. M. (2004). Culture’s open source. Anthropological Quarterly, 77(3),
499–506.
Kuali project: A financial system of, by, and for higher education (n.d.). Retrieved
June 4, 2004, from Kuali Project Web site: http://www.kuali.org/faq/
Leinonen, T., Hakkarainen, K., Appelt, W., Dean, P., Gomez-Skarmetav,
A., Ligorio, B., et al. (2001, June 25). ITCOLE project: Designing innovative
technology for collaborative learning and knowledge building. Paper presented
at the ED-MEDIA 2001 World Conference on Educational Multimedia,
Hypermedia & Telecommunications, Tampere, Finland. Retrieved
August 4, 2004, from Association for the Advancement of Computing in
Education Web site: http://www.editlib.org/index.cfm?fuseaction=Reader.
SearchResults&search_query=AND%28Leinonen%29& SEARCH_STRING=
Leinonen&CFID=37088808&CFTOKEN=60784793
Linden, A., & Fenn, J. (2003, May 30). Understanding Gartner’s hype cycles.
Retrieved February 3, 2005, from Gartner Group Web site: http://www.gartner.
com.
Maxwell, J. (1996). Qualitative research design: An interactive approach.
Thousand Oaks, CA: Sage Publications.
The MERLOT community. (n.d.). Retrieved December 1, 2005, from the
MERLOT Web site: http://www.merlot.org.
Miles, M. & Huberman, A. (1994). Qualitative data analysis (2nd ed.).
Thousand Oaks, CA: Sage Publications.
Moore, G. A. (1991). Crossing the chasm. New York, NY: HarperBusiness.
Moore, G. A. (2005, April 5). Dealing with Darwin: The role of open source in
computing. Paper presented at the open source Business Conference (OSBC),
San Francisco, CA. Retrieved April 29, 2005, from http://www.osbc2004.com/
live/13/events

Journal of Research on Technology in Education 451

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
 O’Mahoney, S. C. (2002). The emergence of a new commercial actor:
Community managed software projects. Unpublished manuscript, Stanford
University. http://opensource.mit.edu/papers/omahony.pdf
Overby, E. M., Bharadwaj, A. S., & Bharadwaj, S. G. (n.d.). An investigation
of firm-level open source software adoption: Theoretical and practical implications.
Retrieved May 9, 2005, from Emory University Web site: http://userwww.
service.emory.edu/~eoverby/files/overby_open_source_adoption_study.pdf
Papastergiou, M. (2006). Course management systems as tools for
the creation of online learning environments: Evaluation from a social
constructivist perspective and implications for their design. International Journal
on E-Learning, 5(4), 593–622.
Pavlicek, R. C. (2000). Embracing insanity: Open source software development.
Indianapolis, IN: SAMS.
Perens, B. (1999). The open source definition. In C. DiBona, S. Ockman, &
M. Stone (Eds.), Open sources: Voices from the open source revolution (pp. 171–
188). Sebastopol, CA: O’Reilly & Associates, Inc.
Perkins, D. N. (1992). Technology meets constructivism: Do they make
a marriage? In T. M. Duffy & D. H. Jonassen (Eds.), Constructivism and the
technology of instruction: A conversation (pp. 45–56). Hillsdale, NJ: Lawrence
Erlbaum Associates.
President’s Information Technology Advisory Committee (PITAC). (2000,
October). Developing open source software to advance high end computing: Report
to the President. Arlington, VA: National Coordination Office for Information
Technology Research and Development. (ERIC Document Reproduction
Service No. ED 462 967).
Raymond, E. S. (2001). The cathedral and the bazaar: Musings on Linux and
open source by an accidental revolutionary. Cambridge, MA: O’Reilly.
Rogers, E. (1995). Diffusion of innovations (4th ed.) New York, NY: The Free
Press. (Original work published 1962)
Sakai (n.d.). Retrieved December 6, 2004, from The Sakai Project Web site:
http://www.sakaiproject.org
Scacchi, W. (2001, December). Understanding the requirements for developing
open source software systems. Retrieved November 7, 2004, from Free/Open
source Research Community Web site: http://opensource.mit.edu/papers/
Scacchi.pdf
Stallman, R. (1999). The GNU project. Retrieved July 16, 2004, from The
GNU Web site: http://www.gnu.org/gnu/thegnuproject.html
Stewart, K. J., & Gosain, S. (2004, August 23). The impact of ideology on
effectiveness in open source software development teams. Retrieved September
20, 2004, from Free/Open Source Research Community Web site: http://
opensource.mit.edu/papers/stewartgosain.pdf
The TLT Group history and data (n.d.). Retrieved December 1, 2005, from
The TLT Group Web site: http://www.tltgroup.org/about/history.html
Villano, M. (2006). Open source vision. Campus Technology, 19(11), 26–36.
Von Krogh, G. (2003). Open source software development. MIT Sloan
Management Review, 44(3), 14–18.

452 Summer 2007: Volume 39 Number 4

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.
 Weber, S. (2004). The success of open source. Cambridge, MA: Harvard
University Press.
Williams, S. (2002). Free as in freedom: Richard Stallman’s crusade for free
software. Sebastopol, CA: O’Reilly & Associates, Inc.
Wilson, B. G. (1996). What is a constructivist learning environment?
In B. G. Wilson (Ed.), Constructivist learning environments: Case studies in
instructional design (pp. 3–10). Englewood Cliffs, NJ: Educational Technology
Publications.

Journal of Research on Technology in Education 453

Copyright © 2007, ISTE (International Society for Technology in Education), 800.336.5191
(U.S. & Canada) or 541.302.3777 (Int’l), [email protected], www.iste.org. All rights reserved.