[MPlayer-dev-eng] [PATCH] Paths for x86_64

[Christof Buergi]

</lurk>

D Richard Felker III wrote:
> FHS is not a legitimate standard. It's been crap from day one.

Maybe it is, maybe it isn't. But it still is the best accepted
standard available. In fact, several distributions switched to it,
because their userbase demanded it. Thus, FHS has some kind of
legitimacy.

Oh, and it made SuSE stop putting their init scripts into
'/sbin/init.d', so it can't be that bad. ;-)

> This is still nonsense. Package managers should relocate files when
> installing NON-NATIVE packages. Or better yet, refuse to install
> non-native packages at all. WE WILL NOT SUPPORT THIS NONSENSE!

We had a heated discussion some weeks ago in the lists of the Linux
Usergroup Switzerland about that very problem. In the end, the general
consensus was, that

a) a pure 64bit operating system with encapsulated 32bit applications
   is cleaner then a hybrid system, but impractical.
b) usability is more important then ideology.
c) a pure 64bit system for x86_64 is not necessarily the best solution,
   as the Opteron/Athlon64 is not a pure 64bit CPU, and the Xeon with
   x86_64 support is no 64bit CPU at all.

In other words: The lib64 advocators (which I was not part of) won the
argument. And two former debianists are now running Gentoo and (gasp!)
Fedora on their Opteron boxes. What I'm trying to tell you: The
backwards compatible hybrid approach with lib64 is likely going to be
the de facto standard, used and accepted by the majority of the
community. It's not very nice to tell them, that their system is
broken, with no reason but that you don't like the lib64 idea.

-- 
                     _ ___      http://www.p2501.ch/             / \
|/  /\   /\   /\  | \   |  |    christof at buergi.lugs.ch          \ /
|\ /--\ /  \ /  \ | /_  |  |                                      x
                                Say NO to HTML in mail and news  / \