UK Judge: Who needs software patents?

Glyn Moody writes "C|Net has a surprising story about a seminar given by a top judge at the U.K.'s Court of Appeal who specializes in intellectual-property law. According to the article, he has "questioned whether software patents should be granted, and has criticized the U.S. for allowing "anything under the sun" to be patented." Is the tide turning?"

Comments

The tide isn't turning

[Bogtha]

The EU Commission are trying to push through software patents again. There's a write-up on Groklaw. I think their idea is to keep trying again and again until we get sick and tired of protesting it.

Re:There are a few good patents as well

[CastrTroy]

RSA encryption is a mathematical algorithm. You can't patent that. You can't patent a program that implements a mathematical algorithm either (or you shouldn't be able to). I don't think i've seen a piece of software that should be patentable.

Re:Pay for the Progress Bar You Use!

[Doctor Faustus]

No, because a baking thermometer does not track progress. During baking the temp usually remains constant.
The temperature of the oven, yes. The temperature of the food, no. Food doneness is usuallly checked by the temperature in the middle of the food: 140-180F for various meats, and a little over 200F for bread.

Re:Fat, Ignorant American Assholes

[Bezben]

Thats funny, I was always under the impression that the US got into world war 2 after they were attacked, a few years after it started... My history has always been shakey though.

MS on patent reform

[badriram]

HEre is what Microsoft is pushing for

Sure it does account for no patents at all on software. It does call for better patent quality, a faster and more open system.

Re:Pay for the Progress Bar You Use!

[tambo]

Sir, here at the USPTO we take pride in granting patents without consideration of trifling concepts such as; gross obviousness, unoriginality and indeed patentability itself.

To be fair, when it comes to software, the USPTO has struggled under two logistical problems:

• It's really tough to find prior art on a lot of software inventions. Sure, patents like Amazon's OneClick method and Compton's "embedding multimedia on a CD-ROM" method are evidence of obviously deficient examination - prior art should have been easy to find. But consider, say, an algorithm embodying a specific video codec - the only instances of prior art might be embodied in closed-source, commercial video players. Short of decompiling and reverse-engineering a bunch of complex software, the examiner is incapable of demonstrating prior art - especially in light of the USPTO's examiner productivity metrics, which severely limit the examiner's time frame for searching and finding prior art.

• The USPTO has been really abused by Congress. They appoint patent administration officers not on the basis of patent experience, but as political favors. Worse, Congress views the USPTO as a cash cow and siphons off its excess funding - thereby depriving the USPTO of funding to improve its examination process (e.g., hiring more examiners.)

It's true that the USPTO could have done better with the resources at its disposal. But it's possible that it just hasn't been given enough resources to meet its responsibilities.

- David Stein