Re: Software patents ...

[<halvar () gmx de>]

Hey all, Gadi,

> They simply re-wrote those files and were over with it.
>
> I suppose the same could theoretically happen with the open source 
> community, and then this list of patents wouldn't be worth much.

Which basically void's Ballmer's assertion that Governments would
have to brace for patent lawsuits (if they were stupid enough to recognize
US patent law) -- as soon as someone sues them over the nonlicensed
use of the patent, the patent number would have to be disclosed.

So I guess it all boils down to Ballmer being a FUD-monkey I guess...

Does anyone have an idea for sensible software patent reform ? Realistically
speaking, algorithmic patents will not be abolished (or not introduced) as 
the
lobby behind it is too large and well-financed. So what would be useful
measures to reform it ? A few ideas:

1) Shortening the expiry time of software patents to 3 to 5 years
Reason: 20 years in software is about 30% of the total existence of the
industry, and at the rate of innovation simply unrealistically long.
2) Forced licensing of patents at relatively low rates _or_ proportional to
the proportion of the patent in a product. This would reduce the licensing
fees for one-click shopping to a few cents hopefully.
3) 12 months of disclosure of patent applications with a challenge to the
industry to show prior art or explain why the solution is obvious before
a patent is granted
4) Putting the burden of paying the challenger's expenses in challenging a
patent (if the patent is overturned) on the patent holder. This would 
basically
mean that if I get myself a trivial patent and somebody decides to challenge 
it,
and it is successfully overturned, I have to pay the challenger's costs. 
This would
create incentive to do a better prior art search and discourage trivial 
patents.
5) In every field and every patent application, the applicant has to pay the
patent office to hire an engineer from a different company in the same field
to solve the particular problem solved by the patent application within n 
(2, 5, 10?)
days of work. If the engineer successfully solves the issue in the same way 
as
in the patent, the patent is not non-obvious, and thus cannot be granted.

Any other proposals or comments on the above ?

Cheers,
Halvar 

_______________________________________________
Dailydave mailing list
Dailydave () lists immunitysec com
https://lists.immunitysec.com/mailman/listinfo/dailydave