Community Patent Review Project Announced

PatPending writes "American companies General Electric, IBM, Microsoft and Hewlett-Packard have joined with the New York Law School and the U.S. Patent and Trademark Office to inaugarate a new system of peer review for software patents. The four companies, plus Red Hat, the world's biggest listed open source software business, are the lead sponsors behind the Community Patent Review project. The one-year pilot program will begin in early 2007 and focus on published but not-yet-granted patent applications relating to computer software. Scientists and engineers will be able to submit prior art to patent examiners at the USPTO using an online system. All Community Patent review project documents will be available on the internet for public comment. 'High-quality patents increase certainty around intellectual property rights, reducing contention and freeing resources to focus on innovation,' said David Kappos, vice president of IP law at IBM."

Comments

In Soviet Russia...

[2.7182]

they had a system like this.

Re:Wasting resources to stop wasting resources.

[Red Alastor]

While this may appear to be a good thing, what it amounts to is these companies further wasting resources in an effort to stop the resources that are already being wasted due to a flawed patent system.

True. But if we had a decent system, peer-review would have a role to play in it. The way I see a good patent, system :

• There is time penalties for every time you troll the patent office. If you submit a stupid patent, your next one will be queued longer. That way, you cannot use the "throw noodles at the wall, see what sticks" approach. If we make it a monetary punishment, big companies could still abuse the system but a time penalty is fair for everybody.
• Peers (read competitors) have the first opportunity to review the patent. They'll jump on the occasion to point prior art. After that, the patent office can do its own research if that wasn't enough but if good prior art was found, it should be very quick.
• The patent office pays court fees every time a patent is overturned in court. It should make them nervous of approving anything that isn't bullet-proof.

This Makes Sense to Me

[magixman]

To fairly grant a patent, the patent office must be able to understand a) what has already been done b) what is just plain obvious.

To achieve either you have to be skilled in the art of the subject at hand and that is just not something one could reasonably expect of a patent examiner who must be a generalist. A community skilled in the art must get involved and I really think this a good thing and could turn the patent system around. No matter how evil you think patents are, they are not going away anytime soon. The best we can do is to better the current situation by supporting efforts such as this.

Re:This Makes Sense to Me

[bit01]

To fairly grant a patent, the patent office must be able to understand a) what has already been done b) what is just plain obvious.

True. The legal fiction that a minor government bureacrat would be able to assess all of human knowledge and arbitrarily decide whether something is original is just mind blowing. Scientists spend their entire working lives in very narrow fields and even they sometimes make mistakes. Not to mention the idiocy of allowing a minor, empire building government department to act as gatekeeper on all of technology.

To achieve either you have to be skilled in the art of the subject at hand and that is just not something one could reasonably expect of a patent examiner who must be a generalist.

True.

A community skilled in the art must get involved and I really think this a good thing and could turn the patent system around.

Not a chance. Have you seen how many software patents are being issued? There is absolutely no chance of a comprehensive review of a significant fraction of them. Particularly since the ones who should do the review, the people who do the actual work, quite rightly regard it as legal fiction make work.

No matter how evil you think patents are, they are not going away anytime soon.

So the patent mafia like to claim. The US is only 5% of the world's population. The software patent situation in 95% of the world's population is in flux. With only a bit of luck the USPTO might find itself out in the cold, to the benefit of everybody. Including the US at large.

The best we can do is to better the current situation by supporting efforts such as this.

No, the best we can do is to get our congressional representatives to fix the law so that it actually implements the objectives laid out in the constitution, to also make sure this massive interference in the citizen's business has a scientific basis and maybe send some of the bribed examiners at the USPTO to jail (with the amount of anonymity, ambiguity and money involved it's a certainty there's a lot of corruption going on). These are all unlikely but not impossible.

---

Scientific, evidence based IP law. Now there's a thought.

A sensible patent system would not cover software.

[Bruce Perens]

How about a patent system that was based on sound economic policy? We don't really have much reason to believe that software patenting promotes innovation in software. Indeed, if you look at the tremendous success of Open Source, the thing they do differently from the rest is that they eschew most intellectual property protection. And they are certainly not short on innovation.

A lot of this is because software is one of the few products that can be effectively produced with reasonably low capital input and has essentially no cost of manufacture once it's been created. Letting everyone get their hands on an idea will be more effective for that sort of product.

So, I submit that a sensibly defined patent system would not have software and business methods within its scope.