Re: support for 'macro' formatting languages

[David Carlisle <davidc@xxxxxxxxx>]

> (though this is the American _Mathematical_
> Society, who might be expected to be fussier about layout than their
> more applied cousins, and thus even less enthusiastic about MathML).

Aha. Thus proving there is a fallacy in your argument somewhere;-)
The AMS are one of the main backers of mathml (including providing
the services of Patrick Ion of Math Reviews who is co chair of the group
and co editor of the spec:-)


> Neither I nor any of the folk I've asked about this can think of an
> advantage in being able to search the maths content of a paper,

Not so long ago (when I was a full time mathematician, as opposed
to a full time MathML and OpenMath person) I would probably have
said the same. But the fact is that publishers are trying to do this
_now_ with your papers (or at least the abstracts of them).

> Notation is not consistent enough to make a search on maths sensible,
> or make it worth while loading a big equation into Maple from a paper.

Depends on the area. In Algebraic Topolgy (where I grew up) this is
probably true. In other areas, the flow goes the other way, you get the
computer algebra system to write chunks of your paper for you.
(Have a look in the library if you don't believe me:-)

> That's the level of validation that would be feasible
> for maths, producing that level of irritation.

XML validators don't try to check the maths for correctness
they just try to check you haven't done something equivalent to
$ a + \section{b} $ to get a bold `b'. You might say that no one
would ever do that. You'd be wrong. If (like Sebastian) you work for a
while for a publisher accepting author written TeX you find that
authors will put _anything_ in their TeX files, and scroll past any
number of errors, if the output looks OK.

I am not one to knock TeX (as some people on this list may know, I
am one of the authors/maintainers of latex2e which is probably
the most widely used flavour of TeX) But in a world where all
information is being passed as XML (a world which might or might not
happen) having Mathematics encoded in TeX would mean that  mathematics
does not have the status of a `natural language' but rather than as a
`graphic': some strange blob of data that has to be handed to a special
system. Having spent the best part of the last two decades speaking
mathematics, I definitely view it as a language rather than as binary
data. 

I accept that the problem of how to input the stuff is not yet solved
(not even close) so I don't suggest that you should switch now.
But sometime in the next millennium I think you (or your students, or
your students grandchildren) will switch.

LaTeX users will be the last people to gain from or use MathML. they
already have a structured markup language with convenient mathematical
input and excellent output. But MathML can be written from CA systems,
from MSWord, and (one day) from tex-mathml translators, and it (one day)
will be rendered in your web browser and be the format in which you get
results from queries into math reviews and the like. TeX is a fixed
point, but the interfaces surrounding MathML will improve until one day
even the most diehard tex hacker will convert.

5 years ago the most frequently asked questions on comp.text.tex
were about TeX. Today they are about converting to or from MSWord, or to
or from a web format such as HTML. It is nice being a TeX user, but it's
also useful sometimes to communicate with the rest of the world.

David
[in prophet mode today, must be the season]


 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist