Re: [boost] [xint] Boost.XInt formal review

Subject: Re: [boost] [xint] Boost.XInt formal review
From: Peter Dimov (pdimov_at_[hidden])
Date: 2011-03-09 21:24:09

Chad Nelson wrote:
> On Wed, 9 Mar 2011 10:10:25 -0800
> Scott McMurray <me22.ca+boost_at_[hidden]> wrote:
>
> >> I'm leery of looking at any GMP code... it's probably pure paranoia,
> >> the GPL can't apply to just *looking* at code, but I'd rather stick
> >> to descriptions to ensure that my code doesn't resemble anyone
> >> else's and nobody can claim that I've copied from them.
> >
> > Your paranoia may be justified:
> >
> > "The courts have made it clear that, under copyright law, proof of
> > substantial similarity between your work and another work, along with
> > proof of access to the other work, may be enough to prove
> > infringement, even when you don’t realize that you’re making a copy."
> > ~ <http://rosenlaw.com/lj8.htm>
>
> Sheesh... that's even worse than I thought. :-( In that case, I think it
> best if I don't even download GMP.

I don't think it's that bad.

Algorithms aren't subject to copyright, and if two pieces of code implement 
the same algorithm, they will necessarily display a similarity. This 
similarity doesn't count as infringement. They either have to share 
substantial copyrightable elements that are not intrinsic to the algorithm, 
or there has to be some other proof that one is a direct copy of the other.

The "mental copying is infringement" precedents have likely been established 
by companies suing ex-workers, and I suspect that these ex-workers were 
probably unable to afford a proper defense, because it's much more likely 
for someone to remember an algorithm than it is to memorize lengthy code.

Either way, I've never heard of a GPL copyright holder pursuing "mental 
copiers".

Not that I'm a lawyer, or something. :-)