Re: Package System specification
I got this from you:
>From rahul Fri Apr 14 12:53:38 2000
Received: (from rahul@localhost)
by reno.cis.upenn.edu (8.8.7/8.8.7) id MAA06118;
Fri, 14 Apr 2000 12:53:34 -0400
From: Rahul Dave <firstname.lastname@example.org>
Subject: Re: Package System specification
To: email@example.com (Jochem Huhmann)
Date: Fri, 14 Apr 2000 12:53:34 -0400 (EDT)
Cc: firstname.lastname@example.org (Rahul Dave)
In-Reply-To: <email@example.com> from "Jochem Huhmann" at Apr 14, 2000 03:44:12 PM
X-Mailer: ELM [version 2.5 PL1]
Content-Type: text/plain; charset=us-ascii
A long long time ago I made a proposal to use RDF for package managing,
incomplete, and non-optimal, to srur discussion. At that time it spurred only
about 3 messages of discussion, and since I am just a user and repackager
for folks at my university(rather than a distribution maker or ISV),
I didnt push, assuming that there wasnt much interest in standardizing on
anything besides rpm, and that other formats would have to be able to
use alien or something to query.
This was late 1998. Its few months into 2000, and we ofcourse still dont have
a LSB(besides FHS), so sometimes I wonder if the LSB is really needed, or
whether the (commercial) market has already taken care
of things by de-facto standrdizing
on redhat. It could be argued that for us folks who have been around Linux
since 50 slackware disks and earlier(what fun that was :-)) that we could
easily do(or someone will do for us) multiple packaging, so it is a non-issue.
Anyway, I offer it up again for discussion. Its at
Note specifically that i also wanted to attack the problem of a proprietary
packaging system being able to use an existing packages database and
contributing to it, as well as for example have a make recognize dependencies
from a package management system without necessarily having to use autoconf,
or even with autoconf, as a standard way of writing dependency macros.
I got this from you:
> * "Philip Rackus" <firstname.lastname@example.org> wrote:
> > Debian and RedHat people are working on the problem. They don't wish
> > to publicize it until there is something tangible in place. Something
> > about to many cooks in the kitchen I believe.
> Then LSB should treat this thing as non-existant.
> Hi! I'm a .signature virus! Copy me into your ~/.signature to help me spread!
> To UNSUBSCRIBE, email to email@example.com
> with subject of "unsubscribe". Trouble? Email firstname.lastname@example.org