Message-Id: <9210021048.AB25965@sc.zib-berlin.dbp.de>
Reply-To: Mailing list for the LaTeX3 project <LATEX-L@vm.urz.Uni-Heidelberg.de>
Date:         Fri, 2 Oct 92 11:47:05 CET
From: Frank Mittelbach <MITTELBACH@MZDMZA.ZDV.UNI-MAINZ.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 How to get rid of most users of TeX
Status: R


> Subj:	How to get rid of most users of TeX

> The TeX community seems bent on shooting itself in the foot.  I have
> seen three proposals this week, each of which would have the effect
> of getting rid of at least 80% of the TeX users: make PS the
> standard output, let us stop supporting MS-DOS and now make the use
> of MF required for all TeX users.  Let us think in terms of
> enlarging, not drastically curtailing the use of TeX.

I guess that would leave about one percent :-)

 I do agree with Michael that this can't be the goal and I hope that
the misunderstanding of ``making PS the standard output'' is cleared
with my other mail.
 I also agree with the statement that MF is not a solution to all our
problems.
 Who is trying to stop supporting DOS? I'm sure that the emTeX
installations are of a number of x00000 with x > 1 and with even more
powerful computers they will probably become even more important.
 [By the way, emTeX with its \specials supported for all drivers has
already shown that it is possible to have a uniform interface for the
important graphic operations (like rotation) for various output
devices]

> A propos of MS-DOS, I do have one suggestion to make.  Since MS-DOS
> does allow only 11 chars in a filename, why do waste three of them
> to convey exactly 2 bits of information.  A font or font related
> file can have, if I have counted correctly, only 4 possible
> extensions, tfm, gf, pxl or pk.  If there are any I haven't heard
> of, they can be accomodated by what I suggest.  Which is that we use
> only one byte for the type and the other two bytes of the extension
> for the design size.  Thus cmr10.tfm would be renamed cmr.10m.  The
> other three would be cmr.10g, cmr.10x and cmr.10k respectively.
> This would require modification to mf and tex, as well as a couple
> of conversion programs, but the modifications would be minor.  I
> would use .07? for 7 point and so on.  The main objection would be
> for fonts over 99 points, but they are not handled in the current
> scheme anyway.  The upshot would be that all 8 bytes of the name
> could be used descriptively.

I find this suggestion very interesting and would suggest that this
should be discussed further in the TeX community and on the
TeX-implementors list since this has only a chance when it would
become a standard in all TeX and MF change files!!

I one would use hex notation for the size one could even go beyond
99pt and still have a relative descritive name (eg cmr.0am, cmr.0cm
etc for cmr10.tfm and cmr12.tfm).

But I think the latex-l list is not the right forum to discuss this
further as I said it has only a chance if *ALL* installation change.

Frank