Factors affecting the use of open source software in tertiary education institutions by David G. Glance, Jeremy Kerr and Alex Reid
Open Source Software (OSS) is software that has been released under a license which requires the distribution of the software’s source code with any binaries. It is often available at no cost and is mostly supported by developers providing their services for free. Considerable interest has been shown in OSS by tertiary education institutions (TEIs) because of the promise of a reduced total cost of ownership of the software, potentially better support, freedom from vendor lockin, ability to tailor the software and pedagogic benefits of being able to view the source code. To find out the extent of use of OSS by TEIs in Australia, New Zealand and the United Kingdom, a survey was sent out to technical personnel at all TEIs in these countries. The results of the survey show that OSS is already being used by all TEIs who responded to the survey and that the major reasons for this was lower Total Cost of Ownership and freedom from software vendor dependence. It is clear however that the majority of the OSS software being used is in server infrastructure with a lesser amount being used on normal desktop machines.
Open Source Software (OSS) describes software released under a particular style of license which requires the distribution of the software’s source code when any binaries are provided. Typically, open source software is available for no cost and has large numbers of developers adding new features and fixing bugs. Many examples have gained the reputation of being featurerich, reliable, robust and secure. These attributes make OSS of particular interest to tertiary education institutions (TEIs), increasingly concerned with escalating software licensing costs. Security is becoming another driver towards OSS with the neverending deluge of attacks against Microsoftbased systems and the constant requirement to update these systems against new vulnerabilities.
The rate of adoption of OSS in TEIs is unknown. Surveys have been conducted on its use in corporations and public institutions in Europe (Wichmann, 2002) and the military in the U.S. (Kenwood, 2001) which showed that the majority of OSS deployed was for serverbased infrastructure, and that the reasons for choosing OSS were reduced Total Cost of Ownership (TCO), greater stability and security and customisability.
It was decided to conduct a survey targeted at TEIs to explore the extent of the current usage of OSS and the perceived benefits and weaknesses of its use.
The extent of the uptake of OSS in TEIs, the reasons for this uptake, and the successes and failures of this experience have been undocumented. A survey was constructed and sent to a list of technical contacts at TEIs in Australia, New Zealand and the U.K. This list covered all of the major universities in these countries. The survey was emailed to the technical contacts, completed electronically and then emailed back to the authors. Of those institutions contacted, technical contacts at 34 different institutions replied.
The number of systems each respondent was responsible for ranged from 10 to 18,000, with an average of 2,700 per respondent. Thirtytwo responses to the survey were received. Twentytwo were Australian, six from the U.K., three New Zealand and one Fijian institution.
The questions on the survey were split into three categories:
- Experience and Skill in Open Source. This section contained questions which were aimed a gauging the level of exposure of the TEI to OSS and the skill set of the staff in dealing with the OSS.
- Benefits of OSS. This section contained questions examining the perceived benefits of OSS from the TEIs. The benefits were categorised as relating to TCO, supportrelated, open standards, and interoperability and vendor independence.
- Support. This section contained questions regarding the perception of support issues around the use of OSS.
Experience and skill in Open Source
Fourtyseven percent of respondents had evaluated open source software and decided on deployment. Eightyone percent of respondents were aware of open source and had experience with it.
Seventyeight percent of respondents reported having staff with skills in one or more open source packages. Seventytwo percent were familiar with the open source development process.
Fifty percent had deployed open source in a significant way. A further 44 percent had deployed open source in a limited way.
In terms of deployment of OSS, 100 percent had it deployed as infrastructure (Web servers, proxies, firewalls, file servers) with 50 percent deployed in administration, 53 percent using it in teaching, 56 percent using it in student labs and 50 percent using it in research.
Only three percent were contributing to development. Fourtyfour percent had contributed in a limited way in terms of filing bugs.
Benefits of OSS
Eightyfour percent stated that the OSS had reduced TCO. Fiftythree percent cited being able to customise the software; Fiftythree percent claimed better response for support and bug fixes. Seventyeight percent claimed less reliance on a specific vendor, and 59 percent adoption of open standards and interoperability.
Eightyseven percent stated equivalent or better support (47 percent stated better) with 13 percent stating worse support.
Sixtyeight percent stated support requirements were not higher than those of proprietary software. Thirtytwo percent stated they were higher.
Sixteen percent of the respondents had put on extra services due to the use of open source software, including extra help desk support and extra staff.
There is a scarcity of information regarding the issues surrounding the use of OSS in TEIs. Several studies have been conducted in other industries and report positive findings in terms of the functionality, reliability, maintainability, robustness, security, and scalability of OSS software compared to their proprietary equivalents. Interest in exploring the use of OSS within organisations is gaining momentum, with many optimistic about the possibility of reductions in TCO and avoidance of vendor control and potential lockin.
From the results of the survey, it can be seen that OSS has already made a significant impact within TEIs in Australia, N.Z. and the U.K. Ninetyfour percent of the respondents reported already using OSS in their institutions. It is also clear that the perception is that the TCO for OSS is less or at least the same as proprietary software, that it is no harder and sometimes easier to support, and that it offers significant advantages over proprietary software.
Perhaps the most interesting observation from the survey is that the primary reasons for adoption of OSS were the perception that there was a saving in TCO and that it freed the institution from reliance on a given vendor. This observation is in line with one of the original motivations of this study, which was that OSS potentially provides TEIs with the opportunity for significant reduction in overall software purchase costs, software support costs and general maintenance costs in addition to the "nonmaterial" benefits. The issue with this, however, is that it is likely to be a perception and not something that has been assessed in a scientific way. It is unlikely, to our knowledge, that any of the participants in the survey have conducted thorough and ongoing assessments of their costs with regards to OSS or compared these to any the alternative proprietary software approaches.
False perceptions on both sides are thus likely to have a large influence on decisions to deploy or not to deploy OSS. It is for this reason that a study should be conducted to actually examine the issues of using OSS amongst the different user roles and assess the issues, solutions and costs associated with the replacement of proprietary software with OSS.
About the Authors
David G. Glance, Jeremy Kerr, and Alex Reid are from the School of Computer Science and Software Engineering at the University of Western Australia.
Email: email@example.com; firstname.lastname@example.org; email@example.com.
This work was supported by a project grant made by CAUDIT (Council of Australian University Directors of IT), for which the authors are very grateful.
C.A. Kenwood, 2001. "A Business Case Study of Open Source Software," at http://www.mitre.org/work/tech_papers/tech_papers_01/kenwood_software/kenwood_software.pdf, accessed 3 December 2003.
T. Wichmann, 2002. "Use of Open Source Software in Firms and Public Institutions Evidence from Germany, Sweden and UK," at http://www.berlecon.de/studien/downloads/200207FLOSS_Use.pdf, accessed 7 January 2004.
Paper received 15 January 2004; accepted 25 January 2004.
Copyright ©2004, First Monday
Copyright ©2004, David G. Glance, Jeremy Kerr, and Alex Reid
Factors affecting the use of open source software in tertiary education institutions by David G. Glance, Jeremy Kerr, and Alex Reid
First Monday, volume 9, number 2 (February 2004),