Re: [Patents] US patent reform strategy inaugurated by myths & lies

[Gregory Aharonian <srctran@world.std.com>]

Sylvain cites:
>- Bessen J. et Maskin E (2000), "Sequential Innovation, Patents, and
>  Imitation", MIT

Please stop citing this study.  It has no real world industry data in it.
It is rather a simplistic artificial model of a small subset of the
software industry.  It cannot be used to prove much of anything about
software patent policy.

I haven't looked at the other studies, but I suspect they are as deficient
of real world data as the Bessen study.

A more rational way to affect the effective length of patent data is
to increase somewhat the maintenance fees of issued patents.  There
was a nice study of German patents done a few years ago showing that
most patents aren't renewed after about ten years or so (except for a
few patents for valuable pharmaceuticals).  Increase the maintenance
fees (for all patents, not just software patents), and I suspect 
patents will be abandoned even earlier, without having to relying on
artificial and ill-supported arguments that the life of a software
invention (or a hardware invention or many other inventions) is or
should be only 5 years.

Greg Aharonian