Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 22 Feb 1997 14:12:35 -0500 (EST)
From:      John Fieber <jfieber@indiana.edu>
To:        Paul Richards <paul@demon.co.uk>
Cc:        ports@freebsd.org
Subject:   Re: Perl5 modules
Message-ID:  <Pine.BSF.3.95q.970222135626.12505B-100000@fallout.campusview.indiana.edu>
In-Reply-To: <87914htuw1.fsf@originat.demon.co.uk>

next in thread | previous in thread | raw e-mail | index | archive | help
On 22 Feb 1997, Paul Richards wrote:

> The only argument I heard in favour of the current scheme was that if
> some hacker was looking for a package to do something, say web
> related, then they'd probably go and look in www.

...Which can be (carefully) implemented in via secondary entries
in the CATEGORIES field of a port...

> It would be so much easier if "real" perl programmers could go to
> /usr/ports/lang/perl_cpan/ and see immediately if the package they
> want is part of the ports collection or not.

A "what it is" classification dictates this approach.  A "what it
is used for" calssification dictates what we have now.  As I've
described in the past, the latter can never be done completely,
and the more complete it is, the more unworkable it becomes until
it is eventually crush under its own weight.

The only thing worse is a classification system with an ambiguous
policy and/or practice.  Predictability is a very important
component of usability (with the standard exception of video
games).

-john




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.95q.970222135626.12505B-100000>