Dutch MPs Demand Obviousness Criteria in European Patent Law

lvd writes "In a fine display of common sense and knowledgability, dutch MP Rik Hindriks demands that first sharp and effectual obviousness and novelty criteria be put in the european patent law, and only then the current ban on software patents could be lifted. A summary is available of the hearing on patents between the Dutch parliament, the dutch IT branche organisation, and the Dutch open source society where this was concluded. A statement of Hindriks (in dutch) can be found there as well."

What's not obvious?

[Anoriymous Coward]

I mean, I'm sitting in front of a powerful general-purpose computing machine which is connected in well-known ways to millions of other general-purpose computing machines. You come along with your patent application saying "I can do X" [not the windowing system]. Am I supposed to be surprised? If it's not obvious, it's probably stupid.

Now if youy're trying to patent an algorithm or codec, that's a different matter. But in those cases it's the algorithm that you should be patenting, not the software implementation of it, which, given the public nature of your patented algorithm, is probably obvious.

Obviousness

[Animats]

There's no indication of what the Dutch want to use as obviousness criteria. Nobody seems to have brought that up. They need to work on definitions.

It's often possible to objectively prove non-obviousness in patent cases. If you can show that other people skilled in the art tried to solve the problem and failed where you succeeded, you've shown non-obviousness.

Re:Obviousness

[Bobo the Space Chimp]

> If you can show that other people skilled in the
> art tried to solve the problem and failed where
> you succeeded, you've shown non-obviousness.

Hence the validity of Amazon's 1-click purchasing patent. In a very "The Inmates Are Running the Asylum" sort of way, it was not obvious -- except in retrospect.

Re:What's not obvious?

[scorbett]

I mean, I'm sitting in front of a powerful general-purpose computing machine which is connected in well-known ways to millions of other general-purpose computing machines. You come along with your patent application saying "I can do X" [not the windowing system]. Am I supposed to be surprised? If it's not obvious, it's probably stupid.

And that, in a nutshell, is the thrust of the argument against software patents. No one should be able to patent something that is obvious. Period. If this means that we see a dramatic drop in the number of patent applications that get approved, great.

--

Re:Obviousness

[Account Number Three]

The famous example is batteries. For years, competing battery companies tried to improve the cardboard-sided battery. And one day, the wife of an engineer suggested that he use a can.

The competitors filed suit to have the patent invalidated as "obvious". The judge merely asked how long the competitor had been trying to develop an improved battery and how much they had spent; after being told years and millions, he dismissed the case.

Re:What's not obvious?

[Anonymous Coward]

For those who've had a long day: the above post by "Anoriymous Coward" is just an old-fashioned troll. Moderate down accordingly. Algorithms should be protected by trade secret only, if at all. Implementations should be protected by copyright. Patents were designed to cover physical objects and are just not appropriate for software.

Re:Obviousness

[SlippyToad]

Sorry, I beg to differ. Only someone who doesn't program would say that this is not obvious, and only 20-20 hindsight could do it. The 1-click patent is just an application of already existing technology -- namely cookies. It's more a business process and any dipshit who's done internet programming could have solved that, given the business incentive to do so. Their entire development effort on the 1-click project was probably all of 20 minutes, once they had the customer feedback and incentive to do so.