help-hurd
[Top][All Lists]
Advanced

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

Re: Hurd FS hierarchy in FHS.


From: Thomas Bushnell, BSG
Subject: Re: Hurd FS hierarchy in FHS.
Date: 22 May 2002 17:07:13 -0700
User-agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2

Jeroen Dekkers <jeroen@dekkers.cx> writes:

> Both the LSB and FSH are "Linux standards" we actually don't have to
> do much with. The filesystem is already partly described under the
> Makefile subject in the GNU Coding Standards.

No, that's not correct.

The FSH was explicitly designed as being more than just a "Linux
standard", and the GNU Project has always cooperated with it.  (I have
no particular opinions about LSB.)  The FSH has always known that
different systems would need annexes.  The point of a Hurd-specific
annex is to outline filesystem properties that all Hurd-based systems
should conform to.

The GNU Coding Standards are *coding* standards, and do not attempt to
describe the entire filesystem.  For example, it was the original
intention that the GNU system might well have a libdata directory, for
architecture-dependent static data.  But we thought that it was always
poor practice to write programs that dependent on
architecture-dependent static data.  So we decided that the GNU Coding
Standards should prohibit that, and accordingly they don't include a
Makefile spec for that directory.

However, the GNU system certainly does contain lots of software that
does not comply with the GNU Coding Standard, and accordingly, the GNU
system does need a place to put architecture-dependent static data.
If I were writing the FHS, I'd say "libdata"; the name comes from the
same source as libexec.  But I'm not writing the FHS.

It is a great mistake to confuse the GNU Coding Standards with a file
hierarchy standard.  They are related, but not anything like the same
thing. 

Thomas



reply via email to

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