The European Parliament discusses tomorrow whether to amend or reject the proposed Directive on the Patentability of Computer-Implemented Inventions, or 'software patents directive', as it's commonly known. (Lots more information, and a history of this directive, is
here.)
For an issue which is so important to my constituents - I've had more lobbying messages on this in the past few months than on any other issue I can remember! - it got no mention at all on the BBC news this morning. To be fair, it does coincide with the G8 build-up and the Olympic bid, but you'd think they'd find space for a mention of such a crucial vote which will effectively determine the future of software development in Europe.
But the debate is certainly hotting up in Parliament, and I've noticed that two myths are circulating about this directive. They threaten to make it sound innocuous when in reality it's very far from innocuous.
Myth 1: This directive is just about "computer-implemented inventions" like washing machine programs and automatic braking systems. What harm can there be in that? This is not so. Although the directive does cover washing machine programs and automatic braking systems, the draft before Parliament in fact also covers all kinds of computer-implemented inventions, including software designed for PC programs or for the internet.
The Council's current position defines "computer-implemented inventions" very broadly and very vaguely, but one thing that almost all commentators agree on is that it will open the door for
software concepts to be patented. That is, under the current proposal, any software concept - an on-screen progress bar, an internet shopping trolley, a resizeable icon, anything - will be patentable and will therefore be protected by law as soon as a company applies for a patent. The only requirement will be that the software implements a "technical invention", which is a disturbingly vague phrase.
In fact, Parliament's original first-reading position redefined "computer-implemented inventions" much better. It specified that only programs which had a
physical effect on the real world could be patented. That definition really
does refer only to things like washing machines and brakes. It makes it clear that 'pure' software concepts - that is, programs and bits of programs which have no physical effect on the real world because they don't control a spin cycle or drive a car - are not patentable.
That first-reading position was welcomed by the anti-patent lobby because it outlined a very reasonable (albeit arbitrary) limit to patentability. But that position has been reversed by the Council's second-reading position.
Myth 2: The people who oppose software patents are just "open-source supporters", a bunch of pie-in-the-sky lobbyists with no connection to modern commercial reality. Nothing could be further from the truth. Software patents are opposed by small software companies, businessmen and individual developers, as well as academics and the open-source community.
Small software companies and individual developers fear trampling by large corporations, who will be able to register vast numbers of patents and then defend them in court against their rivals, or use the threat of legal action as leverage against them. Small businesses and individuals will not have the legal or financial resources to fight back effectively, or to patent their own ideas.
Even if a given concept is not actually patented, small businesses will be very wary to use that concept in their own work because they won't be able to afford to check the patent situation. Software patents will therefore stifle creativity - as well as being painfully anti-competitive.
Only a small proportion of our constituency lobbying has come from "open-source supporters". These are people who believe that software concepts should be free in principle, so that amateurs and the software community can gradually improve and distribute them.