Archive for May 31st, 2007

How sustainable is open source software?

May 31, 2007

There has been a significant upsurge in interest in the use of open source software (OSS) solutions in recent years in both the private and public sector. Education, due to its specialist nature, is particularly interested in OSS for delivering IT solutions in areas where it feels that traditional, closed source solutions have not always catered for its needs. Prominent examples include the Moodle e-learning environment, the Sakai collaborative learning environment and DSpace, a digital content repository system.

One of the big debates to date has been the question of how sustainable open source solutions are likely to be in the long run. The popular image of open source is that a disparate group of software hackers come together in a fairly ad hoc manner, and, usually led by a charismatic figure (or ‘benevolent dictator’) like Linus Torvalds, produce some software. In time, the people involved will go off and work on other, newer, perhaps more interesting projects. This is a worry for education, which needs the reassurance of long-term stability. Questions arise such as who will be maintaining this code, can I read a manual, is there an O’Reilly book, who amongst my staff will understand the programming language that’s been used? These are timely questions and a project that I’ve been indirectly involved in, which attempts to deal with some of these issues, has just come to fruition. Oxford University’s OSS Watch service have announced the publication of their report, “Sustainability Study: a case study review of open source sustainability models”.

If you’re interested in OSS but have previously felt it was too ‘techie’ for you, then I’d recommend this report. My colleague, Gaynor Backhouse, did the editorial on this and I know she was keen to really tell the ‘stories’ involved in order to provide context for the issues and make them more accessible. It was an interesting project to work on, not only because of the subject matter, but also because it involved, in effect, interviewing some of the key figures in OSS development. We joke about it being ‘extreme journalism’, in the style of extreme programming, as there was an iterative process of development and checking with authors. It is, however, quite a long report (around 60 pages) but it is divided into chapters, so you can dip in and out of it as it suits you.