Received: by nummer-3.proteosys id <01C19443.9B94A84C@nummer-3.proteosys>; Thu, 3 Jan 2002 11:44:23 +0100 Return-Path: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C19443.9B94A84C" x-vm-v5-data: ([nil nil nil nil nil nil nil nil nil][nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil]) X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message Subject: {1} Re: {?} uppercase Date: Sun, 23 Feb 1992 17:25:54 +0100 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: From: "Frank Mittelbach" Sender: "LaTeX-L Mailing list" To: "Multiple recipients of" Reply-To: "LaTeX-L Mailing list" Status: R X-Status: X-Keywords: X-UID: 583 This is a multi-part message in MIME format. ------_=_NextPart_001_01C19443.9B94A84C Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable > > Actually when I said URGENT I meant ,,to be solved before LaTeX 3.0`` = :-) And this is exactly why I said it isn't urgent. I'm sorry when this is disappointing some of you but please try to understand my position here. There are only two possibilities for working on LaTeX: - adding kludge to kludge to the current syntax, adding a new feature whenever somebody is requesting it. - redesigning it from scratch with the proper internal tools to handle most if not all problems for typesetting in various diciplines. I can't do both, I already spend much more time on the LaTeX project then I should and trying to improve the current version while working on the next major release would double this time. Also the first alternative isn't very good because any syntax addition is something which I can't easily change for ltx3 and this means it should be considered very carefully before something is officially added to latex209. We did so with NFSS and also the amstex.sty stuff is something which probably will find its way more or less painlessly into ltx3. But many of the currently important problems not solved with the current latex cannot be added to it in a consistent manner without touching all internals all over the place. For this reason I like you to remind everybody that this is not a discussion list on how to solve problems with with latex209 but rather how to design ltx3. I can only repeat: bringing up the problems like the \uppercase one by Yannis is very important. Also suggestion what to do with them, e.g. how to solve them. But please accept this should not be done under the viewpoint of adding stuff to latex209 on short notice and also that there are unfortunately many many other problems that are equally important (perhaps to others). > Here is a more concrete proposal on which I would be glad to have = Frank's > and other's opinion: > make auxiliary commands \addtochapter, \addtosection, \addtosubsection > etc, which > 1. may appear several times > 2. will add their argument to the current section/chapter/subsection = etc > name, after it has been \uppercase'd. In the inbetween one may = change > \uccodes ad libitum. > This solution seems to me upwards compatible (no syntax changes, just > new commands added). This proposal is exactly one of the above kind. Of course it is a syntax change since we have a lot of new commands. I don't mind if somebody is implementing something like this in the current latex (as a style option) but I'm strictly against adding this unreflected officially to the current system. Michael has outlined a solution that would work internally, instead of scanning for $ signs; one can surely develop a different scanning system for more general goals. I'm not sure but perhaps we will finally do character scanning by hand in commands like section this would allow to produce such effects without any additional costs although it wouldn't be very efficient. > I still consider the whole thing to be URGENT since this small problem > destroys a big effort on Greek (and Cyrillic and other) TeX. Avoiding > styles which use \uppercase is not a solution. Come on, please. Reconsider this. There are so many of these small problems around. What is better: ltx3 within 2 years or 20? Frank ------_=_NextPart_001_01C19443.9B94A84C Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable {1} Re: {?} uppercase

>
> Actually when I said URGENT I meant ,,to be = solved before LaTeX 3.0`` :-)

And this is exactly why I said it isn't urgent. I'm = sorry when this is
disappointing some of you but please try to = understand my position
here. There are only two possibilities for working on = LaTeX:

  - adding kludge to kludge to the current = syntax, adding a new
feature whenever somebody is requesting it.

  - redesigning it from scratch with the proper = internal tools to
handle most if not all problems for typesetting in = various diciplines.

I can't do both, I already spend much more time on the = LaTeX project
then I should and trying to improve the current = version while working
on the next major release would double this time. = Also the first
alternative isn't very good because any syntax = addition is something
which I can't easily change for ltx3 and this means = it should be
considered very carefully before something is = officially added to
latex209.

We did so with NFSS and also the amstex.sty stuff is = something which
probably will find its way more or less painlessly = into ltx3. But many
of the currently important problems not solved with = the current latex
cannot be added to it in a consistent manner without = touching all
internals all over the place.

For this reason I like you to remind everybody that = this is not a
discussion list on how to solve problems with with = latex209 but rather
how to design ltx3. I can only repeat: bringing up = the problems like
the \uppercase one by Yannis is very important. Also = suggestion what
to do with them, e.g. how to solve them. But please = accept this should
not be done under the viewpoint of adding stuff to = latex209 on short
notice and also that there are unfortunately many = many other problems
that are equally important (perhaps to = others).


> Here is a more concrete proposal on which I would = be glad to have Frank's
> and other's opinion:
> make auxiliary commands \addtochapter, = \addtosection, \addtosubsection
> etc, which
> 1. may appear several times
> 2. will add their argument to the current = section/chapter/subsection etc
>    name, after it has been = \uppercase'd. In the inbetween one may change
>    \uccodes ad libitum.
> This solution seems to me upwards compatible (no = syntax changes, just
> new commands added).

This proposal is exactly one of the above kind. Of = course it is a
syntax change since we have a lot of new commands. I = don't mind if
somebody is implementing something like this in the = current latex (as
a style option) but I'm strictly against adding this = unreflected
officially to the current system.  Michael has = outlined a solution
that would work internally, instead of scanning for $ = signs; one can
surely develop a different scanning system for more = general goals. I'm
not sure but perhaps we will finally do character = scanning by hand in
commands like section this would allow to produce = such effects without
any additional costs although it wouldn't be very = efficient.

> I still consider the whole thing to be URGENT = since this small problem
> destroys a big effort on Greek (and Cyrillic and = other) TeX. Avoiding
> styles which use \uppercase is not a = solution.

Come on, please. Reconsider this. There are so many of = these small
problems around.  What is better: ltx3 within 2 = years or 20?

Frank

------_=_NextPart_001_01C19443.9B94A84C--