>>>> "dv" == Didier Verna
<verna(a)inf.enst.fr> writes:
dv> Pete Ware <ware(a)cis.ohio-state.edu> writes:
> Can one share package directories across machines and
architectures
> (cf. etc/)
dv> That's a good point. Perhaps the most important argument against
dv> having all files related to a package under a single directory. There's
dv> already a problem with the lib-src subdir, and will be another one when we
dv> start having packages with C modules in them.
Yes. I'll consider this for the revision.
dv> It would really be a pity not to fix this problem in the Great Package
dv> Overhaul. The fact is that most of the packages subdir already exist in the
dv> GNU Coding Standards.
dv> - etc, info, man exist.
dv> - lisp should go under .../share
dv> - lib-src actually belongs to libexec.
dv> - lib could then be used for C modules.
dv> So I think instead of having a single directory for a particular
dv> package, we could replicate this hierarchy under all needed standard
dv> directories, with the same naming policy.
This would destroy a large part of the rationale for the proposal. I
think this high-level organizational question is better not touched
upon by the package system at all. GNU Coding Standards are one
thing, but few software packages adhere to them, and there are a
gazillion other such standards. Few packaging problems are
satisfactorily solved by directory organization alone, and the main
thrust of the draft is to make things as independent as possible from
it. You can mainly try to not let it get in the way too much.
The approach of the GNU Standards seems to be to move
architecture-specific stuff pretty deep into the hierarchy, across
disjoint portions of it. Probably a workable approach is to
distinguish architectures under <whatever the name of the directory>,
and let the high-level directory structure just do what it needs to
do.
--
Cheers =8-} Chipsy
Friede, Völkerverständigung und überhaupt blabla