standard dependancy system

[Nicolas Mailhot]

Le dimanche 25 novembre 2007 à 12:44 +0100, Kai-Uwe Behrmann a écrit :

> > > I do it as well (sometimes). But I doubt that anybody from distro
> > > maintainers read each line of config.log for each update of each
> > > package to verify, that all features were included.
> > 
> > Which is funny in a really sad kind of way, because you would think that
> > any company whose main business it is to compile other people's software
> > would try a little bit harder to get consistent results. 

A lot of the packaging work is not done company-side (and usually is
more rigorous than company-side or even upstream packaging BTW)

Also don't forget that many distribution build for multiple arches, and
upstream authors have frequently been knwon to ignore their own warnings
in the first place.

> Agreed. Let me extend this aspect even further.
> 
> Is a package featuring a plug-in system still the same when not delivered 
> with its plug-ins? Should at least the name say that it is not complete?
> Who decides about this? Should'nt the original maintainer have a last 
> say?

The maintainer has a say. If he makes the effort to document properly
his system people will take his wishes into account. If knowing his
wishes requires web site or mailing list archeology, he gets what he
sown.

> Distributors should be forceable to meet quite other standards.

Please add this at the start of your web sites so I know never to touch
anything you produced. Either you make it easy to do the "right
thing" (whatever the right thing is in your book) or you can get lost.
Know that every deviation from your wishes in downstream packaging is a
direct result of lack of documentation or a problem in your code the
downstream packager had to work around.

Life is too short to second-guess upstream authors, fix their problems
in their stead, or change the rest of the distribution so it fits the
ideal of some upstream maintainer.

Most distribution packagers will consider that a package that works 90%
is acceptable if doing the last 10% is so much work than packaging the
code in the first place is no longer a good bargain. If some authors
have a problem with this they can either force packagers through
licensing or stupid interactive setups (short way to blacklisting) or
make sure their code never has this 10% of not efficiently solvable
problems in the first place.

Also if a distribution decided the right answer to some question is
different from what an author thinks, the distribution choice will take
precedence to the choices of the authors of each individual app. That's
how the game is played, distributions provide some sort of consistency
to users, people who don't like the consistency choices of one distro
have always been free to setup their own around other choices.

-- 
Nicolas Mailhot
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: Ceci est une partie de message
	=?ISO-8859-1?Q?num=E9riquement?= =?ISO-8859-1?Q?_sign=E9e?=
Url : http://lists.freedesktop.org/archives/xdg/attachments/20071125/ddfe9a05/attachment.pgp