Message-Id: <9311142001.AA27129@sc.ZIB-Berlin.DE>
Reply-To: Mailing list for the LaTeX3 project <LATEX-L@vm.urz.Uni-Heidelberg.de>
In-Reply-To:  Message of Sun,
              14 Nov 93 11:57:56 -0700 from <michal@GORTEL.PHYS.UALBERTA.CA>
Date:         Sun, 14 Nov 93 20:52:20 SET
From: Goossens Michel <GOOSSENS%CERNVM.bitnet@vm.urz.Uni-Heidelberg.de>
Sender: Mailing list for the LaTeX3 project <LATEX-L@vm.urz.Uni-Heidelberg.de>
To: Multiple Recipients of <LATEX-L@vm.urz.Uni-Heidelberg.de>
Subject:      Re: psfonts.dtx
Status: R


>> `chancer3.sty' is a relic of the *previous* discussion about names for
>> styles. Before the `nf' prefix was invented, i suffixed everything
>> with `3' to point the way to LaTeX3...
>>
>> sebastian
>>
>
>Here is a perfect example for one of reasons why I protested against
>all this "prefix" (make that "suffix" as well) business.  Everything is
>fine and dandy if you work on some document, here and now, and it will
>be neither widely distributed, to a location(s) with an unknown version
>of a system and rather unsophisticated users, nor it will be archived
>for future references/retrieval.  This pictures instantly changes if
>you have 6000 documents on line, like somebody wrote that he has.
>Then you take your 'nfchance' if this will work, or not, because some
>style name has changed.
>
>  Michal
>  michal@gortel.phys.ualberta.ca
>  ntomczak@vm.ucs.ualberta.ca

And this is precisely the reason why we wanted to make a clean break
and call our styles something different with the new LaTeX (or with
NFSS2 in the case of Sebastian). I thank that Michal makes the trivial
assumption that times.sty is a standard style (like article.sty or
bezier.sty), which is distributed together with LaTeX. Well, to be
sure, it is NOT. Before Sebastian and Karl Berry tried to put some
order in the PostScript-fonts-with-(La)TeX-mess several hackers had
come up with their version of a modified lfonts.tex, and proudly called
it times.sty (because in the good old days Apple LaserWriters only
had those fonts available). I really do not see the big mental step
in the brain of some people that keeping the name of a file that defines
(amongst other things) Times-Roman as the text font is a guarantee that
their old files will continue to work, as by magic, or at least that
it will be much clearer to point to that file as the culprit.

I can accept the argument of giving the full eight characters to the
developers to call their files lucimath, lucid, lucidbrb, lucidbry
or any other straightforward name, but I refuse to admit that this
will make a difference to the ease with which documents are transported
between sites. I am working in a Physics laboratory, where we have
over 2000 resident scientists and engineers, plus another 2000 visitors,
from all over the world. And I can assure you that they want
their documents to run on all the sites of a collaborating set of
laboratories. Therefore, to speak the same language, they set up
a comparable (functionally identical) LaTeX system at all those sites,
and the people coming from outside will have to try and make their
documents work with that "standard" system. That is a one-way change.
As for the thousands of documents, which we have in our archive,
we try to help the user to run his document if he ever needs it again,
with no guarantee that everything will work if he used non-standard
styles (like the times.sty at his site in Ottawa, Paris, Frankfurt
or Moscow). This is the whole idea behind LaTeX2e, to have a clear
break, and make a series of styles work with the "standard" LaTeX,
providing at the same time commands for checking which version of
LaTeX you are using and for verifying whether the given style was/is
compatible with that version. That is a long way from file names and
being sure that a file in your archive, which ran with times.sty, as
it was available at Alberta University three years ago, will run
immediately with LaTeX2e, merely because there is now a file times.sty,
coming with new LaTeX, which has the same name (and hopefully does
something similar as the original times.sty).

I hope the sterile debate on file names is now closed. We have given
the user visible files a "intuitive" name, keeping the ones which were
already available before (although I do not know, e.g., even for
Helvetica, which file name is best chosen, since at our site
I have for that already helv.sty, helvetic.sty and helvet.sty).

>Might it be better to aspire to getting references to fonts out of
>\document...  commands altogether (in line with the "separate the role of
>the document-designer from that of the author" philosophy)?  I.e.,
>* as far as possible, avoid \document...[12pt]...  and avoid
>  \document..[..times..]..  and avoid \document..[..helv..]...
>* regard font things as document-designer things rather than as
>  "ordinary user" things.
>If this were done, the question of what things were called would matter
>to document-designers, but not to ordinary users.
>   ..... lines deleted .....
>
>To some extent, the amateur publisher (e.g., postgraduate doing thesis) is
>aspiring to be "professional" anyway, so the above remarks may still apply.
>In practice, the "maintainer of the University of X thesis-styles" might
>give a choice of designs, from which the "ordinary user" can choose (like
>choosing wallpaper).  The ordinary user might need to know whether the
>design can be previewed by software that just "knows about" Computer
>Modern, or whether software that "knows about" PostScript is needed.
>But all other details can be hidden.  I.e., I think "the maintainer" would
>be better hiding references to [nf]times.sty (if the design needs them)
>from "the ordinary user", rather than telling "the ordinary user" to put
>[nf]times.sty in a \document...  command.
>
>Rather than to spend a lot of time worrying about how font-specifications
>should be named in a \document...  command, mightn't it be better to aspire
>to getting font-specifications out of \document...  commands as far as
>possible (so that their names don't matter to "the ordinary user" anyway)?
>
>                                                                David Rhead
Let us hope that David is right, and that we can convince our many scientific
authors that they _are not_ master typographers, that they better spend their
time on improving the content of a paper, rather than its presentation.
But if the latter is important, then there should, of course, be tools for
doing the job. Yet, that can be left to the package designer, who can choose
the fonts that are needed.

Michel

-----------------------------------------------------------------------
 ,=====,======  Dr. Michel Goossens      Phone: (+41 22) 767-3363
/  ,----\, |    Documentation Section    Fax  : (+41 22) 767-7155
|  C E R N |\|  AS Group - CN Division   Email: Michel.Goossens@cern.ch
|\|      / \ |  CERN European Laboratory
| "====="  /\        for Particle Physics
    '--/--'  \  CH-1211 Geneva 23
      /       \ Switzerland
-----------------------------------------------------------------------