On Sat, Nov 18, 2006 23:06:33 PM -0600, M Harris
([EMAIL PROTECTED]) wrote:
> On Saturday 18 November 2006 02:38, M. Fioretti wrote:
> >
> > This is correct, because forking is useless or undoable when the
> > problem is software patents. See my other message in this thread.
> No...
>
> The fork (done properly) removes the patent issue.
Removes my foot. Again: sw patents are not copyright. Please learn the
difference. Patents forbid the use of _algorithms_, not that of
already typed files of source code.
> no GPL package can (by legal definition) contain patentable code...
see your confusion? Code is not patentable, is copyrightable. Besides
that, a developer should know in advance (that is, have legal
expertise and spend a _lot_ of time digging patents databases) that an
ALGORITHM is already patented and cannot be used NO MATTER HOW YOU
TYPE IT.
> if it is GPL and if it ships at all the *entire* thing becomes
> GPL... period.
Absolutely not. What you say _may_ happen only if it is the patent
holder who also releases the code under GPL. But if the generic
developer ships as GPL an implementation of an algorithm already
patented by somebody else (something not so easy to find out unless
you're a specialized lawyer) the code can be used.
> The only thing that needs to happen is to remove the prop packages
> from the distro (sanitize the logos,.. not strictly required) and
> repackage.
and here you prove that I am right, that is that no fork can solve a
patent problem. What if the "prop package" is the kernel or gcc? What
if the problem is, say, the ALGORITHM in the kernel that schedules
processes? how easy it is to rewrite the kernel so that a whole
distro works with another algorithm, and how useable would the result
be if all the unpatented algorithms left have very poor performance?
> Remember... there isn't a patent controversy here... only FUD. We
> always need to keep reality and fantasy separate.
Please start yourself then. All the Uncertainty and Doubt you and
others are spreading by confusing all the time patents and copyright
doesn't help. It doesn't even really matter if and how the Linux
kernel or any other GPL package violates some sw patents.
What matters in reality is how users (especially corporate and
governments ones) and developers are scared away from GPL systems and
how feasible it is for almost everybody to survive (financially) a
patent lawsuit from Microsoft. Even if they are surely wrong, you'd go
bankrupt long before proving it. This happens all the time, not just
in software.
If you were right (both in theory and in practice), Stallman and the
FSF would not spend half their time worldwide to scream against sw
patents. The reasons they do is exactly because they have effects like
the ones I have described.
Ciao,
Marco
--
The right way to make everybody love Free Standards and Free Software:
http://digifreedom.net/node/73
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
