>>>> "David" == David Bush
<dbush(a)adn.alcatel.com> writes:
David> On 9 Jul 1998, Didier Verna wrote:
> Now, the argument falls back to the eternal matter-of-taste
> discussion: when you have a package "pack" that provides libs
> and includes, the policy I like is installing in
> /usr/include/pack and /usr/lib/pack. The policy I dislike is
> installing in /usr/pack/include and /usr/pack/lib.
David> I'm with you but who are we to fight against the combined
David> "wisdom" of at least four major Unix workstation vendors
David> and the OpenGroup? :-)
"We is _us_, and dooooon't you ferGIT it!"
The vendors and the Open Group presumably have what they consider
sufficient reason to do this. However, remember that "sufficient
reason" often includes things like intra-organizational turf wars and
repository permissions. If the clients complain loudly, the internal
voices of sanity may be heard, too.
We may be able to prevent insanity on the part of minor vendors. (As
a _single_ beta tester I got the /usr/ml---multilingual---hierarchy
removed from the new TurboLinux-JP distribution. The /usr/jp
hierarchy stayed, unfortunately ... due to a turf issue.)
And we should prevent pollution of systems without CDE. I recently
got the "no bold italic version of fixed" error, and it recommends
setting a resource to something in /usr/dt/.../fonts, which I don't
have. Bletch.
--
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences Tel/fax: +1 (298) 53-5091