Message-ID:  <E0wk7s1-0000bQ-00@heaton.cl.cam.ac.uk>
Reply-To: Mailing list for the LaTeX3 project
              <LATEX-L@RELAY.URZ.UNI-HEIDELBERG.DE>
In-Reply-To:  Your message of "Fri, 04 Jul 1997 13:57:22 -0000." 
              <199707041205.OAA07322@centre.univ-orleans.fr>
Date:         Fri, 4 Jul 1997 13:53:46 +0100
From: Robin Fairbairns <Robin.Fairbairns@CL.CAM.AC.UK>
Sender: Mailing list for the LaTeX3 project
              <LATEX-L@RELAY.URZ.UNI-HEIDELBERG.DE>
To: Multiple recipients of list LATEX-L
              <LATEX-L@RELAY.URZ.UNI-HEIDELBERG.DE>
Subject:      Re: Availability of Class files (was: LaTeX3 goals)
Status: R

Michel Lavaud writes:

> > Are you all familar with Graham Williams catalogue? in many ways, he
> > has already laid the foundation for what we are talking about, with a
> > BiBTeX database looking like this:
>
> I think it is extremely useful to have the TeXIndex of Jones updated.
> But I am not sure that the texlive key, with the contents described
> above,  is a  good idea?

David Jones hasn't got the time/inclination/something or other.  I
asked him some while back, and he said he was quite happy for the file
to be zapped from the archives.  I had forgotten to do that, and have
just moved it to obsolete/help

The reason the texlive key is there is that the file is also used (as
a BibTeX file, no less!) in preparing the TeX Live documentation.  The
key doesn't appear in the other derived format, CTAN
help/Catalogue/ctfull.html

> "Not a 'must have'" seems to me usselessly aggressive to the package
> writer, at least if Graham's document is publicly released.

It depends how you interpret the phrase.  I would claim that it's
pretty readily interpreted as `take this if you want it, but it's not
part of a standard distribution'.

> Moreover,
> if  somebody wants to do something and the "xyz" package
> exactly does what he needs, what is the use for him to know that the
> package has been classified by somebody as a "not a must have"?
> And for those who are not interested in what "xyz" package does,
> getting an opinion on it is useless too. So, why giving any opinion
> at all ?

Give an opinion if you want to classify, don't if you don't.  Don't
try to classify without giving any opinion: that way lies madness...

> And cataloging a package as "generally agreed important application"
> is not very serious, is it ? As it will reflect the opinion of, from
> one individual (for many packages) to ten (say hundred for the
> "bestof" package) individuals, out of hundreds of thousands of users
> (somebody spoke about 20 billions).

No-one's talked about billions of users yet, have they?  There are
only 100 million PCs (or something) in the world!

> I am not sure that users really need opinions. I think they need
> much more an exhaustive catalog, as the original TeXIndex is,
> updated and with improved abstracts and demo files. Once they have
> it, they can decide by themselves if the package is useful for them
> or not.

This is precisely what Graham's stuff gives us.  We're very lucky that
he was willing to take up the baton when David Jones had to give it
up.

> If people have free time to discuss about a package, and have an
> opinion at the end, I suppose they will have written a set of test
> files? Therefore,  I would propose that they release their test
> files, without emitting any opinion,  instead of emitting an opinion
> and no test files. And, about the texlive key, replace it by a
> pointer to these test files ?

As I said, the texlive key performs an entirely different function.  I
didn't read this discussion as claiming that these cataloguers were
going to construct test suites, or anything of the sort.  I'm happy
that the cataloguers `merely' collect opinions and distill them into a
digestible form, as Graham Williams has already started to do.

In my view, a project to construct a test suite for every package on
CTAN is doomed to failure.  (For some reason, my fingers kept wanting
to type `dodo' where I meant `doomed' ;-)

Robin F