Message-Id: <199402022127.AA28420@mail.cs.tu-berlin.de>
Reply-To: Mailing list for the LaTeX3 project <LATEX-L@DHDURZ1.BITNET>
Date:         Wed, 2 Feb 1994 22:25:14 +0100
From: Frank Mittelbach <MITTELBACH@mzdmza.zdv.uni-mainz.de>
Sender: Mailing list for the LaTeX3 project <LATEX-L@DHDURZ1.BITNET>
To: Multiple recipients of list LATEX-L <LATEX-L@DHDURZ1.BITNET>
Subject:      Re: On the death of \footheight
Status: R

Paul,

i think you are wrong and below is why.

david said in another message it has been a long day---so it has for me
and i can't say that i'm too pleased either about some of the messages
i've seen this evening after 10 hours hard work. (not talking about this
message). since this is online excuse me if something is unclear.

generally:

many people complain about many things and it is certainly easy to
speak about time bombs etc. but if you think a bit about what you ask for
you might find that you yourself have conflicting goals.

on one hand people complain about stuff getting to large or
that package X is not running if used together with Y and Z and on the
other hand you really ask for big blue compatibility. don't forget
that IBM has at the same time always sold larger machines.

coming back footheight: what does 2e currently provide?

a) the dimen is removed in native mode to make it more likely
that package X does work with package Y without overflowing the
fixed number of dimens

result a style that just defines it (without using it as did latex209)
need to remove this definition. how many classes do you think will have
this problem?

a style that really uses it needs to declare it to be able to run it
in 2e, again how many styles will that be?

b) it is let to a scratch dimen in compatibilty mode

this will make it possible to run any old document that contains
a setting for it in the premable (something unnecessary since it is
not used, but somebody might have thought it was necessary)

> David Carlisle:
> > That does not mean that all the other thousands of LaTeX users should
> > have one of the very limited dimen registers wasted by being allocated
> > but never used.
>
> \dimendef\footheight=0 would solve most of these problems.
>
> You can assign to it without trouble (unless you're stupid enough to think
> that \dimen0 retains its value after your code surrenders control),
> and since it is nowhere guaranteed to have a meaningful value, can read it.

but  in that case it is of no use to a style that really uses it
like seminar since such a style might not want to get a random value
whenever it reads it out.

> It you want an error message,
> \def\footheight{\begingroup\afterassignment\no@footheight\dimen0=}
> \def\no@footheight{\@latexerr{footheight is obsolete}\endgroup}
> would even preserve \dimen0, though you couldn't read \footheight.

i think that is even worse since then you waste space (something
people complain about) and you gain nothing since Tim still has to
change seminar to make it run.

so i guess you can't win them all.

> Whilst LaTeX2e is compatability mode looks more or less compatible,
> this sort of thing is going to mean that it will be a LONG TIME
> before people (can) switch to native mode.

speaking about LONG TIME: when i joint the project of writing the companion
there was nothing like latex2e around or thought of. we had dropped
the idea of producing 2.10 long time ago.
the companion was perhaps finished to 70% when the discussion was restarted
and to 90% in april last year when it was finally decided that such a version
is important for the reasons given in the announcements.

as a result the companion became a sort of test case since the latex at my
home changed daily. in the middle of it i  froze the whole environment
and although i had added by then all the text describing 2e I used very an
early implementation of latex2e (in some cases only partly implemented stuff)
to produce the book. in other words the book was basially produced with
latex209 + NFSS2 + a few macro halffinished macros to produce the 2e
examples.

this book uses nearly all style files it describes in the single latex run
that is producing it (some example are done offline and pasted via eps fgures)

now in the last few days i have started to rewrite the companion class file
in latex lingua, eg removing handmade code for newcommand the new
minipage options etc. after doing this and after finding a bug in
the changebar.sty that did not show up with the 209 output routine
the companion compiled under 2e without any change to any package files
used (and it produced the same page and line breaks as before)

so much for the number of changes necessary to styles to run under 2e
(of course, there will be styles that need a bit of helping but not many)

as for the compatibility mode: it processed leslie manual without problem
and it did processed all the test files produced last year by the volunteer
group without showing differences (well, after fixing a few stupid bugs,
ie the version you should find on the servers tomorrow or so).

so i really don't think that there will real problems if people are
only willing.

however one can frighten any everybody one only have to be raising
the voice enough. whether or not it is to the point does not really matter.
--- unfortunately if i may say so.

good night
frank