[Top][All Lists]

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

Re: [Gnustep-cvs] r31321 - in /tools/make/trunk: ChangeLog GNUstep.conf.

From: Yavor Doganov
Subject: Re: [Gnustep-cvs] r31321 - in /tools/make/trunk: ChangeLog configure
Date: Fri, 17 Sep 2010 21:49:57 +0300
User-agent: Wanderlust/2.15.5 (Almost Unreal) SEMI/1.14.6 (Maruoka) FLIM/1.14.9 (Goj┼Ź) APEL/10.7 Emacs/23.2 (i486-pc-linux-gnu) MULE/6.0 (HANACHIRUSATO)

SPUeNTRUP - Kai Henningsen wrote:
> Am Fri, 17 Sep 2010 10:00:37 +0000 (UTC)
> schrieb Yavor Doganov <address@hidden>:
> > The FHS people *tried* to make it widely adopted by soliciting
> > input from many parties, and failed.  They probably failed because
> > they didn't listen at all to that input.
> Well, I remember those debates, and that isn't my impression. My
> impression was that few other-system people were at all interested in
> the first place, and most who were were interested only in making the
> FHS *identical* to what they were accustomed to. A standard isn't worth
> much if it just describes the mess that made people want a standard;
> necessarily, everyone must compromise. The Linux guys had a strong
> motivation; everyone else pretty much did not.

Yes, I agree with you that this is a better description of the "FHS
standartization process".  Basically, GNU (GNU/Linux only at that
time) distros began to realize that the minor discrepancies in the
layout between them were causing only trouble (inter-distro patch
exchange, user confusion when switching from one distro to another,
etc), so they decided to eliminate them, or perhaps, document them.
The GNU (GNU project) and BSD folks were mostly spectators, and the
proprietary system developers even more so.  The latter had only a
remote interest because of the eventual PR benefits from the
"FHS-compliant" rubber stamp.

There was a lot of hype and energy in those days, but the efforts to
maintain the standard up-to-date with recent developments have been
... non-existent.  Up to the ridiculuous point that Debian Policy had
to add a special exception for /selinux (as per that policy, any FHS
violation is a Debian Policy violation, but FHS doesn't mention
/selinux at all, and amusingly, SELinux is the invention of roughly
the same folks who coined up the FHS :-)).

In general, it is more or less irrelevant whether we like the FHS or
not.  As long as GNUstep works with this layout, and Nicola was kind
enough to implement the revolutionary changes in gnustep-make 2.x
(which made our lives significantly easier), what the default layout
is doesn't matter (at least for distro maintainers, that is).

@David: IMHO you don't have to use or test the FHS layout in order to
consider it "supported".  Just invite users encountering problems to
report bugs to the distro, or (in case they build from source and
install with the FHS layout), forward them to Savannah so that others
can examine and look at the issue at hand.  That way, bugs will get
fixed one way or another, instead of declaring FHS as "second class
citizen" and GNUstep installed in this manner "broken".

reply via email to

[Prev in Thread] Current Thread [Next in Thread]