[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: TeXlive 2008 in Fedora 11?



2009/1/29 Nicolas Mailhot <nicolas mailhot laposte net>:
> Le jeudi 29 janvier 2009 à 00:51 +0100, Kevin Kofler a écrit :
>> Tom "spot" Callaway wrote:
>> > I'd have to say the chances of this are extremely unlikely. I'm much
>> > more inclined to propose that we drop texlive altogether.
>>
>> No TeX? That would make Fedora pretty much useless for Mathematics and
>> anything related!
>>
>> If concrete licensing issues are found by somebody, of course those need to
>> be addressed, but dropping all of TeX on the ground that it's too hard to
>> audit is a horribly bad idea.
>
> People who want TEX kept should create a task force and feed TEXlive
> 2008 in rawhide bit by bit, checking for each bit the licensing is ok.
> That's about the only way the burden of checking TEX can be distributed
> enough it won't depend on a couple packagers, when the amount of work is
> too much to be humanly doable by a few people alone in a realistic
> timeframe.
>
> Our curent TEX packaging would be similar to try to package 80% of CPAN
> in a single source package. Of course people crap out when they try to
> check licensing correctly (and TEX stuff has more licensing variability
> than Perl)

I have actually been spending a chunk of time looking carefully at the
way in which TeXlive is put together from the various upstreams to see
if we can't do better than taking the monolthic tarball dumps of
texlive as we have in the past. To say that the way the bits and
pieces of TeX are put together is a horrific mess is an
understatement. However, there is some hope. Forefront in my mind is
that we do want to capitalise on the integration work that is done by
texlive, as we don't have the manpower to reproduce that effort. On
the other hand we want something much better integrated and modular
than texlive presently provides for distribution packaging. At the
moment though, I only have time to work on this at weekends, so
progress so far has been slow, and has been entirely learning.

J.


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]