Print

Print


On Tue, 24 Nov 1998 16:33:59 +0000
Robin Fairbairns <[log in to unmask]> wrote (inter alia)

> sebastian writes (of chris rowley)
> > anyway, you are (and you know it!) deliberately misrepresenting my
> > basic view point....

[deletions]

> the reason i have got away with as much latex-ery as i have, over the
> years, is that it's been useful to have a repository of last resort of
> answers to people's questions around here.  i regularly get asked for
> help by people all over the department, occasionally by people in
> other parts of the university.  sometimes, latex is unavoidable; and
> latex isn't so well documented that everyone can trivially do every
> sensible thing they might want.

I find myself in much the same position as Robin.
I also find myself asked about installing TeX and friends on assorted
platforms - Unix, Linux, DOS, OS/2, Win95 and WinNT so far.
I find that even people who have been using TeX and LaTeX for years
can come up with some obscure bit of code which interacts in an
unusual way with some of the packages.
I have managed to get (nearly) everybody to use \newcommand instead of
\def, but even then some publishers packages can do funny things.

The latest was a staff member putting a book together.
He had used \newtheorem{example]{Example}[section]
and got a most obscure error message.
I tracked it down to part of the publisher's class file, which also
defined an "example" environment.

But even if we eventually get away from TeX/LaTeX into something else,
we will still have the problem of interactions between different macro
(or whatever they may be called) definitions.
And doing any sort of extensive writing in mathematics *without* some
sort of macro facility is just too awful to contemplate.
Has anyone ever tried to write something in *raw* TeX

> i don't believe there's a real solution to this problem, in the short
> term.  whether a meta-language (`designers' interface') will help
> isn't clear to me.  whether the xml bandwaggon, or whatever is
> tomorrow's buzz-word takes out latex altogether, i can't possibly
> guess (though the evidence of the sgml takeover doesn't inspire me to
> assume that xml will do all that much better).
>
> but even if xml _does_ take over, there are going to be people wanting
> advice on how to use xml tools to produce an acceptable visual
> result.  these things _won't_ be well-enough documented (at least, the
> versions the researchers can afford), any more than latex is now.

I wholeheartedly agree with the last two paragraphs.
One problem with visual display is the very limited resolution,
compared with the printed page.  This can be overcome, for small bits
of a page, by zooming, but this gets horribly tedious even in xdvi
And I am quite sure that a certain person in charge of a large company
in USA would be rather unwilling to commit the required amount of
money into adding a zoom facility to the internet viewer just to
satisfy the wishes of a (relatively) small mathematics community.

> robin

Ken Smith
[log in to unmask]