Educational institutions in the UK consume a lot of software. Office suites, student record databases and virtual learning environments (VLEs) are just a few pieces of the software that are procured by UK universities and colleges.
While some of the software available is proprietary, meaning it carries licensing restrictions on its use and distribution, there are many open source options which are free to use, modify and distribute without additional permission from the copyright holder or supplier.
At OSS Watch, the unbiased open source advisory service based at the University of Oxford, we conducted our fifth National Software Survey in 2013 on behalf of Jisc. The survey was set up to establish how proprietary software and open source software is used in UK higher education (HE) and further education (FE) and identify trends over the past decade.
While the results of the survey show some commonalities between how the two sectors approach the different types of software, we can also see clear differences.
Open source vs proprietary software
Of the software actually deployed, all HE institutions surveyed reported at least some usage of open source, particularly on servers. By contrast, only about 50% of FE institutions reported using any open source at all. I don’t advocate that all software used should be open source, but I certainly advocate that it should be considered equally to proprietary solutions.
If this was happening in practice I’d expect to see closer to equal proportions of open source and proprietary being adopted in the HE and FE sectors, so what’s going wrong, and why is FE’s adoption of open source so much lower? Looking at the responses given in other sections of the survey sheds some light on the issue.
Looking at the criteria used to select software for procurement, both sectors indicated ‘meeting user expectations’ as the top criterion, with ‘total cost of ownership’, ‘performance of the software’ and ‘interoperability’ also scoring highly. HE also considered staff preferences and staff expertise to be important factors, while FE considered support quality and support costs more important.
It’s notable that while interoperability was cited as important, risk of vendor lock-in preventing easy transition to an alternative solution was one of the lowest ranked criteria.
Why not open source?
When asked to give reasons for not choosing an open source solution, interoperability and migration problems were again cited by both sectors. Looking at actual software products in use, it’s clear that Microsoft solutions are highly dominant in both sectors. It is possible that the perceived lack of interoperability is really a symptom of institutions being locked in to one company’s ecosystem.
Another factor cited by both sectors for not choosing an open source solution was a perceived lack of such solutions for a particular need. Over the past year I’ve been compiling and maintaining a record of open source solutions for education which lists solutions for educational software needs, from back office administrative systems to subject-specific teaching tools, covering a broad range of subjects. This may simply be a hindsight rationalisation for a lack of knowledge and support for procuring open source effectively.
While looking at the current state of software deployed, it’s important to look at the software policies in place at UK universities and colleges.
The software policies that exist may not bear relation to the software used on-the-ground in universities and colleges today. However they should give us an indication of how the sectors are evolving and the levels of adoption we can expect to see in the future. It’s here we see clear differences in approach between HE and FE:
- Over 70% of ICT policies in HE state that open source should be considered as an option, while only 40% of FE policies state this
- 35% of FE institutions state that proprietary software is preferred over open source, while less than 20% of HE institutions state this. Neither sector has such a preferential policy towards open source.
Half of HE institutions reported that they considered open source and proprietary software equally, up from 30% at the last survey in 2010. Only 25% of FE institutions reported equal consideration for open source, a slight decrease since 2010.
Contribution and engagement
Engaging with an open source project allows you to get maximum value from it by benefitting from community support and helping drive development in a direction that matches your needs.
In HE, contribution is far more likely than in FE, showing an upward trend since 2008 to today where all HE institutions have some level of contribution to open source. By contrast, only 60% of FE institutions currently contribute to open source projects, the lowest level since 2008.
The VLE market
The only area where FE has a larger proportion of open source deployments than HE is in the VLE space. Both sectors reported a majority of institutions running Moodle, but in FE it was almost ubiquitous while a reasonable proportion of HE institutions are running the proprietary Blackboard or WebCT VLEs.
As the VLE market is a relatively recent development and the big proprietary players carry a price tag prohibitive to most FE institutions, Moodle will be the first VLE many of these institutions will have had. As users become used to a system over time, they can be resistant to the change that moving to an open source alternative would bring. However as there was often no VLE in place when Moodle was considered, this made the open source option an easy choice.
The future of open source
I believe that the differences we see here indicate a need to support effective consideration of open source during procurement, especially in the FE sector. The responses from our survey show reservations about the suitability of open source options. However the success of Moodle shows that it can be an effective solution for the sector, if vendor lock-in can be mitigated.
The situation looks more positive in HE, but if we want to see a more balanced software market in the future, change in policies and practice need to happen much faster. If current trends are maintained it’ll be another 20 years before all institutional policies consider open source and proprietary solutions equally.
Read the survey in full on the OSS Watch website.