Last week I went along to the grandly-named Westminster eForum Keynote Seminar on Open source software: in business, in government. The good news was that it offered one of the best line-ups of open source know-how in the UK I have come across. The bad news was that the seminar's venue was quite small and not even full: these people really deserved a much bigger audience. The poor turnout was a sad reflection of how far open source still has to go in this country in terms of mainstream recognition and interest.
Things began with Karsten Gerloff, President of the Free Software Foundaion Europe, giving a nice, gentle intro to all the basic concepts. There then followed a number of sessions, all of which had names that will be familiar to readers of this column.
For example, one on "Implementation and the costs of open source and free software" had fellow blogger Andrew Katz, as well as Alan Lord, Director of the Open Learning Centre (his speech is now online). Taking part in a discussion of "The challenges of deploying open source software in the public sector" was Mark Taylor, a tireless advocate of open source in business and government, and also a blogger here on Computerworld UK.
Another member of the latter panel was Mark Wright, "Cabinet Member for the Transformation, Value for Money and IT, Bristol City Council." His was a long, sad tale of trying to save local government money by switching to OpenOffice.org. As he explained, he was brought to the brink of defeat not through any problems with or deficiencies of the software itself, but by the fact that everyone else in local government preferred to pay Microsoft large sums of money for its Office suite instead (more details in this report.)
It was a classic case of lock-in, and led very nicely to the last session of the day, on "Open standards and software: consumer rights and legal implications," and the principal reason I was there. I was ready to say all the usual stuff about how open source runs the Internet, powers leading companies like Google, Facebook et al., and yet still can't get a look-in when it comes to central and local government procurement in the UK. I put this down to lock-in of the kind described by Wright in Bristol: people understandably can't be bothered dealing with the problems raised by trying to shift from the old to the new, even if the latter is better and/or cheaper. This gives existing suppliers power over customers, when it should be the other way around.
I was going to point out that the way out of this situation was to insist on open standards. These create a level playing field, allowing all companies to compete on a fair basis. They allow products to be swapped in and out with ease, creating true competition among suppliers, and a Darwinian selection that drives down prices and pushes up quality.
But for that to happen, the standards must be truly open: that means anything nominally claimed as "patented" would need to be released under Royalty-Free licences, not the more restrictive Fair, Reasonable And Non-Discriminatory (FRAND) (for the full half-hour argument see my previous post on the subject.) Otherwise, the majority of open source projects would be excluded, and with them much of the healthy competition and the benefits it brings.
That's what I had planned to say, and I did, in fact, say most of it. But things were rather changed by an interesting presentation that preceded the session on open standards. This was given by Francisco Mingorance, "Senior Director of Government Affairs, Business Software Alliance."
It is a sure sign of my approaching senility that I had failed to notice those last three words when I skimmed through the conference agenda beforehand. For, as readers of this blog may recall, it was only a couple of weeks ago that I expended some effort pointing out the fallacies of the truly awful BSA report on the alleged deleterious effects of software piracy on the world's economies.
So once I had focussed my age-addled brain on the presence of the BSA at this event, I then re-read the title of the talk - "The European debate on software interoperability, openness and freedom of choice", and noticed the words "freedom of choice". I suddenly knew exactly where this one was going. A short press release (not online that I can find) sent out by the BSA shortly afterwards reveals exactly what kind of "freedom" we're talking about here:
In keynote remarks at today's event, Francisco Mingorance, BSA senior director of government affairs, said, "Public entities should procure the software that best meets their needs—based on functionality, performance, security, and cost of ownership—and avoid dogmatic preferences in favour of open source or commercial software. Governments that attempt to favour one software development model as superior in terms of a particular variable risk making incorrect choices among the full range of solutions available in the marketplace."
Amusingly, this release is headed: "BSA calls on UK Government to avoid archaic open source vs. proprietary software policies". Gotta love that "archaic", as if the UK government has been favouring open source since the 18th century. The thing, of course, is that there has been no actual presence of open source in government procurement, ever. The current moves to introduce some are merely rectifying a situation – nay, an archaic situation – whereby certain manufacturers have had an almost total monopoly on government IT, and newcomers are locked out.
But as I've indicated above, I actually agree with Mingorance: I think we should have a totally level playing field for government procurement. But for that to be the case, we must have Royalty-Free, not FRAND, licensing for any nominal patents that are contained within relevant standards.
My impression was that Mingorance's presentation depicted such RF licences and the software that required them in a rather negative light. The GNU GPL, for example, which is one of the main licences that cannot accommodate FRAND licensed-standards, was described by him as "IP-restrictive." Since "IP" is actually a government-granted monopoly – and most people agree that monopolies are bad things – that's like describing the police as "criminal-restrictive": true, but a decidedly odd way of looking at things.
Mingorance also noted that existing major standards such as GSM or Wifi are FRAND not RF. But these standards were drawn up decades ago, before free software represented a viable alternative to proprietary software, so it's hardly surprising that the relevant standards bodies opted for FRAND. By contrast, it's striking that a more modern standards body, the World Wide Web Consortium, does insist on Royalty-Free for its standards:
The W3C Patent Policy governs the handling of patents in the process of producing Web standards. The goal of this policy is to assure that Recommendations produced under this policy can be implemented on a Royalty-Free (RF) basis.
And as Mingorance said himself, we must "focus on the future", not be trapped in that archaic past.
The trouble, then, is that BSA is inconsistent: it asks for "dogmatic preferences" to be avoided, with no "favour" being shown to "one software development model", but fails to recognise that its own preference for FRAND, rather than RF, licensing is "dogmatic" in precisely this way, since it favours certain development models over others, like those producing GPL-licensed software.
So when it comes to creating level playing-fields through open standards, let's avoid double standards. RF licensing discriminates against no one, and favours no one; it maximises the number of rival offerings and hence increases the overall competition. This allows customers to procure their software "based on functionality, performance, security, and cost of ownership": who could possibly be against that?