Wolfgang Bornath a écrit :
2011/7/8 andre999<[email protected]>:
Wolfgang Bornath a écrit :
2011/7/7 nicolas vigier<[email protected]>:
On Thu, 07 Jul 2011, Wolfgang Bornath wrote:
I must admit I do not understand the cause of this discussion, maybe I
am thinking in too simple ways. Free goes in core, non-free goes in
non-free. If a non-free software has a restrictive license it goes in
tainted. A free software can not have a restrictive license, if it has
it is not free and goes in tainted.
Tainted is not about restrictive license but patents. A free software
can have a free license, but do something which is maybe patented.
Yes, right. I made a mistake there - just replace "restrictive
license" with "patents" in my sentence.
"free" means that it can be redistributed with source code, with a free/open
source license.
"non-free" (in terms of the repos) means that it can be redistributed, but
either not with source code, according to the license + or we simply don't
have/can't get the source code.
"tainted" was mostly for packages affected to some extent by tainted
patents.
Such packages could be free or non-free, that has nothing to do with being
in "tainted".
Some discussions in the past considered that the likelihood of a patent
impacting a particular software (in the few countries that do accept
software packages to some extent, like the USA), should affect whether it
goes into tainted or not. I don't know what consensus there was on this
point, if any.
That exactly was the reason why "tainted" was created. The gathering
of such software in one repo to make it easy for users and mirror
maintainers in those countries to avoid them if they chose (or are
forced) to do so.
Since as far as I know, nobody introduced any evidence during the debate of any
mirror being pursued for carrying potentially patent-affected packages, despite
the fact that many mirrors in the USA, for example, carry such packages, any
problem for mirrors is a moot point.
However users (or mirrors) that wish to respect patent claims would evidently
appreciate avoiding the contents of a "tainted" repository.
Remember that very few patent claims against software are ever validated by the
courts. Most pursuits in the USA are against companies with very deep pockets,
which are often settled out of court for convenience. Generally in these
cases, the defending party is earning revenues from the subject of the patent
claim. Which of course doesn't apply to software distributed for free by mirrors.
(e.g., if I remember correctly, Novell paid license fees to Microsoft for
patent claims, while other companies successfully contested the same claims in
court.)
--
André
