help-hurd
[Top][All Lists]
Advanced

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

Re: Hurd FS hierarchy (was Re: LD_LIBRARY_PATH troubles)


From: Carl Wilhelm Soderstrom
Subject: Re: Hurd FS hierarchy (was Re: LD_LIBRARY_PATH troubles)
Date: Sun, 24 Mar 2002 15:44:08 -0600
User-agent: Mutt/1.2.5.1i

On Sun, Mar 24, 2002 at 12:24:07PM -0800, Thomas Bushnell, BSG wrote:
> Carl Wilhelm Soderstrom <chrome@real-time.com> writes:
> >     what's the logic behind the /com name?
> We needed a place for mutable data that could be shared between
> machines.  
<snip>
> The AI lab put it in a directory called "com", which stood for
> "common".

the main problems I see with the name are:
1. from my DOS and VMS experience, COM means 'commands' (namespace
collision).
2. isn't this the same thing as what '/usr/share' is all about? so why not
call it /var/share or just /share (except that /share has already been
populated with the stuff from /usr/share, which you may or may not agree is
similar enough to what you're thinking of, to mix together).

there's enough stuff in /var that *can* be shared between machines, I don't
think there's reason to leave it sacrosanct and say it's for unshareable
stuff; let alone make a new tree on / for 'shared variable stuff' as opposed
to 'unshared variable stuff'. the main problem with doing that, is that what
may not be considered shareable at the moment by one set of people, could be
shared by another group, for their own reasons. (for instance, I can easily
concieve of sharing /var/log, such that a group of machines log to a common
syslog file, and can all view it locally. since this isn't commonly done,
should the standards body put syslog in /var/log, /com/log/, /var/share/log,
or /log?)

Carl Soderstrom.
-- 
Network Engineer
Real-Time Enterprises
www.real-time.com



reply via email to

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