Devil's Advocate: IP disputes wholly unsatisfactory

COMMENT Penalising software developers for creating identical code is like punishing writers for occasionally using the same turn of phrase. It's just one more reason why software patents kill innovation, says Martin Brampton.

Do big companies really have any grievances over intellectual property? Or is it all a big scam? The latest revelations from SCO tend towards the latter. Is what they are doing much different from the stream of emails that offer me a share in million dollar funds that just happen to be lying around?

Amazingly we are still subjected to claims that patents will help the European software industry. Hugo Lueders of CompTIA says software patents are needed to ensure the EU can keep to the goals set by the Lisbon Agenda - that the EU will become the world's most competitive and dynamic knowledge-driven economy by 2010.

If by a dynamic knowledge-driven economy he means a bonanza for lawyers, then he might be right. For the most prominent cases reaching the courts have all the hallmarks of people trying it on. It is impossible to know whether SCO has a case for its claim that Linux is a rip-off of proprietary Unix code. But the judge responsible for SCO's case against IBM characterises the evidence produced so far as "a complete lack thereof".

Now, if further doubt were needed, we hear that emails within SCO strongly suggest that searches for incriminating code were a complete failure. The response is apparently to suggest the Linux code has been rewritten to hide the copying. Earlier it was claimed that fragments of Linux code were line-for-line identical to Unix code.

The possibility of basing a legal claim on fragments of code is alarming. Possibly judges can be persuaded that a few lines of code being the same counts as a copyright infringement. But anyone who has written much software would not be at all surprised if fragments of identical code came from independent skilled programmers tackling the same problem. After all, the repertoire of possible statements is far fewer than ordinary language. And the solution is to some extent determined by the problem.

Copyright tends to focus on the actual words used. Patents are even trickier, with the emphasis being on the ideas behind the words. When I sit down to write an article, how can I be sure that I am not saying something really rather similar to somebody else? For we are all subject the influence of the same events. The mere fact that more than one person reacts in the same way is not proof of plagiarism - but it would still be counted as breach of a patent.

Were restrictive legislation something called for by the smaller software companies that are the source of most real innovation, we might be persuaded that patents would promote a dynamic economy of novel products. But the small companies are exactly those most opposed to extensions of patent law into software.

Their fear of litigation by large companies holding portfolios of patents is hardly groundless, given the kind of actions that currently occur in copyright. IBM is well able to take care of itself in a spat with SCO. Small companies have little option but to seek a settlement, since they frequently cannot afford to defend themselves against a patent infringement action.

Of course the big companies know this and understand that their claims of infringement stand a good chance of never being tested in court. Combining that with an apparently casual approach by the authorities granting patents to schemes of highly dubious originality, we have a situation that is anything but conducive to innovation. At the moment, Europe is on track for consolidating this wholly unsatisfactory situation.