Open Source Software is now a viable and credible alternative for government, says a report published yesterday by the UK's Office of Government Commerce. The report, detailing the verdict on a series of proof of concept trials of OSS, which were carried out in conjunction with Sun and IBM over the past year, notes that the three main areas of implementation are at different stages of maturity, but notes that cost savings can be achieved, and signposts OSS' attractiveness as a possible escape route from vendor lock-in.
The OGC is not by a long chalk evangelising open source. On the contrary, throughout the document it maintains a measured and balanced tone, addressing the issue of whether it's feasible to consider OSS against proprietary systems in national and local government procurement, and if so, what kinds of roles represent the 'low-hanging fruit' where OSS can be deployed most cost-effectively. So the OGC is not saying that UK government should switch to open source as a matter of policy, it is saying that in many areas OSS can be better, and more cost-effective, than proprietary solutions. Ironically, one pilot study where proprietary lock-in proved such an insuperable problem that the pilot had to abandoned took place at, er, the OGC's executive agency, OGC Buying Solutions. Proprietary lock-in also seems to have been rather more of an issue when it came to communicating with central government systems than it was elsewhere, so Whitehall clearly has some distance to go before it can walk the talk.
The report finds that OSS is "viable and credible for infrastructure and for meeting the requirements of the majority of desktop users." Desktop deployments are still limited by "lack of complex functionality", while "for business applications, the lack of Open Source products to compete with large-scale proprietary enterprise-level products" is also an obstacle. It recommends that public sector bodies should "examine carefully the technical and business case" for OSS implementation, consider it for server consolidation, and consider the potential costs and benefits of "migration to an OSS desktop for transaction users, (potentially in conjunction with use of 'thin client' architecture solutions)".
The longer range recommendations are more interesting. It suggests identifying the role of open standards in future strategy, taking account of the eGov Interoperability Framework (eGIF), developing skill in OSS development, deployment and operation, reviewing infrastructure and apps well ahead of any planned procurement and renewal, and considering the benefits of incremental change that involves OSS deployment. The forward planning should "determine whether current technologies and IT policies inhibit future choice" and if so, "what steps may be necessary to prevent future 'lock-in'", while suggested incremental changes are in the areas of "Email, LDAP, Web and internet Browser."
The lock-in issue comes up repeatedly in the document and the pilot reports and case studies included with it; unsurprisingly, given that the OGC's prime directive is to achieve value for money in government purchasing, and single vendor scenarios are therefore from its point of view A Bad Thing. It isn't however the OGC's role to define overall government IT strategy and policy, so the OGC cannot itself provide a specific answer to the question, 'fine, but where do we go now?' That role falls to the e-Government Unit (formerly the Office of the e-Envoy) which is due to publish an update to its OSS policy shortly.
The individual pilot and case study reports are interesting reading in their own right. The MoD Defence Academy, for example, has a Linux-based system using Apache, Zope and Plone, and takes the view that an OSS server infrastructure "is inherently more secure than one based on proprietary software." Central Scotland Police is another notable user, while StarOffice and OpenOffice are (depending on the level of requirements) viable as deployments on both OSS and Windows platforms (the case study with 5,000 StarOffice licences which doesn't want to be named is probably, by the way, Bristol Council, which by a miraculous coincidence in currently known to be deploying 5,000 StarOffice licences). A government department which also wishes to remain anonymous meanwhile volunteers the information that it thinks "the Microsoft monopoly position [is] unhealthy from a procurement viewpoint". Even in the cases where existing systems blocked cost-effective OSS deployment, the view of the subjects tends to be not that OSS is therefore a route that they should not go down, but that they should examine how they got locked in, and investigate how they might be able to break out in the future. Ofwat's experience with its Aquarius software (which uses Excel and relies on Visual Basic macros), for example, "suggests that any new software project should be planned using open standards to allow maximum choice of desktop in future."
This sort of talk sounds ominous for Microsoft, but as recent trends in UK government IT procurement have indicated, it isn't necessarily wholly positive from the point of view of the average open source developer. The focus on value for money (a charge that's been led by the OGC) has contributed to the concentration of contracts in the hands of a very few large companies, the NHS IT project (where the most widely-used current GP system is actually being squeezed out) being the most obvious example right now. There is a danger that giving OSS the government seal of approval will simply put Sun and IBM forward in bids as Linux-toting Tweedledums to Microsoft's Tweedledee, and that these two will be just as capable of locking customers in as Microsoft is.
An OGC spokesman stressed to The Register that this was not the intention, and that the organisation was aware of the importance of smaller developers. "We look at innovation, we don't look at headcount, and you can quote me on that," he said hopefully. So we did.
Microsoft issued a statement on the report, saying: "We understand that it is the role of Government to promote a level playing field and to foster increased competition in any market. However, having read the report in detail the findings do not align fully with feedback we regularly receive from our customers in the market place who have evaluated Microsoft software against Open Source software. We would encourage interested parties to read the report, its recommendations and conclusions, in detail in order to enable them to reach their own informed conclusions."
NHS OSS white paper is 'disappeared'
'Independent' report used MS-sourced data to trash OSS
Gov.UK and MS upgrade licensing deal
OGC streamlines purchasing portals
Gershon retires from the Office of Government Commerce
Microsoft, Sun, IBM and the war for government desktops
Open source not ready for desktop, IBM told UK government