Subject: Re: [boost] Conflict with patent
From: Mathias Gaunard (mathias.gaunard_at_[hidden])
Date: 2010-07-20 18:51:07
Le 20/07/2010 21:19, schnarf_at_[hidden] wrote:
> Mathias Gaunard wrote:
>> Patents become invalid if you can demonstrate prior art.
>> I wouldn't worry too much about it in this case.
>
> Patent invalidation is a very difficult thing to do, and not a good way to build a defense. Attempting to invalidate a patent can bankrupt you with legal fees, and finding prior art that legally works can be tricky.
It only matters if the patent holder tries to enforce the patent, which
is realistically not going to happen in this case, because they would
ultimately lose. The relation to his work is fairly remote, there is a
massive amount of prior art (it's in TAoCP for god's sake), and it's a
quite trivial addition to a widely known technique.
For these reasons, my opinion is that the Boost community shouldn't care
about that particular patent if the code was ever considered for review.
Of course that opinion is just that of a simple developer with no
corporate experience, and I'm far from being a lawyer.
I do not know if there ever have been pieces of code rejected due to
such far-fetched patents. I think there were discussions with a few
atomic concurrency things and people managed to work around the patents,
so surely some people know about subject very well.