Éibhear wrote:
Hi,
I'm in foreign lands right now and low on time, so I don't have much time to research this fully. However, there seems to be an interesting new proposal from the US and some major bodies there for patents. Read the summary at Groklaw and follow the links to read the report and its executive summary.
http://www.groklaw.net/article.php?story=20040428200118940
Éibhear
The proposals read pretty much like "motherhood and apple pie", at least to European eyes. But I understand they may still be in for quite a rough ride from some of the American interests, and the rather conservative "don't change a thing" US small inventors lobby.
The problem, I think, is that faced with these proposals the Europeans would probably respond "but we do all of this anyway, already" -- and yet Europe still has granted vast numbers of really objectionable software patents.
The proposal that seems to generate the most hope among American supporters of the F.T.C. and now the N.A.S. proposals, is the idea of giving all-comers a six-month European-style window to challenge any patent application before final grant.
But actually, even in Europe such proceedings haven't made an awful lot of difference -- huge numbers of bad software patents have still been issued, and even large companies of the scale of say Canon don't feel there's any great benefit in trying to do the Patent Office's job for it, paying good money to do all the legal work of filing an opposition just for the general benefit, unless there is a particular product line or development interest they need to protect.
So: no silver bullet.
The key problem isn't an issue of patent quality -- even if some way could be devised to objectively assess genuine "inventive height" in the discovery of an algorithm, all the evidence is that in almost all cases there simply aren't the societal benefits in granting such a patent to outweigh the costs of sanctioning a 20-year monopoly on it.
The only line which really works is to say "no patents on the pure processing of data" (nor for user interfaces, nor for general "ideas for programs").
The N.A.S. sees enough problems with the USPTO that it didn't see this as yet another sensitive battle it needed to fight.
But in Europe we should be quite clear: Software patents? No thanks.
All best,
James.