Friday, October 8, 2010

Open Source Content and Document Management in the Public Sector

Yesterday I attended another of the Manchester BCS's open lectures, this time an excellent talk by Graham Oakes entitled "Open Source Content and Document Management in the Public Sector". As the title implies, the talk focused on issues associated with using open source solutions for content management systems within public sector organisations and essentially reported the conclusions from a meeting held by the BCS Open Source Specialist Interest Group at the start of the year.

That meeting in turn had been inspired by a Guardian article reporting on a website project by Birmingham City Council ("Why can't local government and open source be friends?", 7th August 2009). The article suggested that the spiralling cost of the project could have been significantly reduced if the council had opted for open source software (with the defining characteristic of being free to install and use for any purpose) for the content management system (CMS) behind the website, rather than a costly proprietary solution.

Certainly there is now plenty of mature and robust open source software that is recognised by industry analysts as being "enterprise-ready" (possibly the most famous example being the Apache webserver, "the most popular web server on the Internet since April 1996"), and this includes open source CMS's being used successfully to run the websites of prominent public bodies (Graham cited several notable examples including various UK police forces, the C.I.A. and even the White House). So open source is certainly a viable option for these applications.

Also there are many apparent benefits over proprietary software:
  • Lower front-end costs: the software is free to install and use; also it's cheap to experiment with before committing to it wholesale.
  • Easier to work with: the source code is visible and can be modified as required; also there are no licensing issues for virtualization or cloud computing applications (which can be a real problem when using proprietary software).
  • Favours incremental delivery: you can start small and build up over time, rather than having to deliver one huge project all at once.
Another potential benefit for public sector organisations is that use of open source can demonstrate a commitment to openness by the organisation. However Graham was keen to point out that there are also a number of risk factors which counterbalance these benefits:
  • Misunderstood costs: probably the greatest misconception about open source: even though it might be free to download doesn't mean it costs nothing to run - someone still needs to support the system day to day. But perhaps more significantly, the biggest costs associated with the implementation of any CMS - whether open source or proprietary - are for things like content migration and training. (It's probably in these areas that the aforementioned city council website project actually went wrong, as the CMS software is likely to have been around only 10% of the total budget.)
There are other "misperceptions" (for example, having access to the source code is not an intrinsic benefit - it also requires that your organisation has the expertise to exploit it). Graham also cited some cultural factors that might work against the successful use of open source by public organisations:
  • Mismatch of scale: generally open source is on a much more smaller scale than government organisations, which are better at interacting with large corporations.
  • Broken procurement models: government procurement is geared towards buying licences rather than buying services, and this bias tends to favour proprietary solutions over open source. (An interesting later observation was that this model tends also to favour large programmes over smaller ones - but that in his opinion the larger a project is the greater the chances of failure are.)
(A further point was that people can have a "philosophical bias" towards or against open source, leading to unreasoned decisions about which technology to use, regardless of how well it matches the requirements.)

In conclusion, while there are potential benefits to public sector organisations using open source, it's certainly not a given. In many ways using open source isn't that different from proprietary software. He closed with a few conclusions:
  • All open source is not the same: there are variations in quality, capabilities and levels of community support, so these should be assessed before committing to a particular solution.
  • Focus on your problem and not on the technology
  • Consider the total life-cycle costs: look beyond just the start-up costs when comparing open source and proprietary software.
  • It's the team and not the technology that creates success: good people will still succeed using mediocre tools.
Hopefully I've accurately communicated the key points of a Graham's fascinating talk. I'd add that many of his conclusions chimed with those from "Open Source for the Enterprise" (Dan Woods & Gautam Guliani, O'Reilly Media) which makes the case for open source in the private sector - and reiterates the point that informed consideration of benefits and risks of any technology is vital in making good choices.

No comments:

Post a Comment