Re: GPL, yet again. (The kernel is a lot like a shared library)
> We SHOULD be taking an individual work and analyzing it
> for creative content. Not cooking up arbitrary hypotheses
> and pretending they mean something
I am going to try taking some hours this weekend and verify one
of the programs + libcurl + openssl cases. I'll get back to you
next week.
> I don't have a specific answer. That depends on the creative content
> of X and Y.
>
> But, in general terms, I'd be looking for creative ideas (ideas first
> expressed in Y) which are incorporated into X. For example, if
> X will serve as a replacement for Y, that would strongly hint (but
> not guarantee) that X is a derivative of Y.
One last thing: not even this. Following this reason, the GNUTLS SSL
shim would be a derivative work of OpenSSL. And it's not. Remember, the
whole argument I'm giving is:
1. the GPL can only restrict the distributions of an original work or
its derivatives; and
2. the only way to determine if a work is a derivative work of another
is going thru the abstraction/filtration/comparison process.
--
HTH
Massa
Reply to: