# [texhax] problem with a defined command

bnb at ams.org bnb at ams.org
Fri Jan 17 14:24:24 CET 2014

On Fri, 17 Jan 2014, Philip Taylor wrote:

> I do not see it as santioning its use, but rather means of getting
> aolder docs to run withpout replacing in the body

Well, each can interpret it as he/she chooses, but MJD wrote :

> Work is under way on an auxiliary package called breqn that addresses
> not only this problem but a number of others; at the time of this
> writing, however [September 1999] it has only progressed as far as a
> beta release. Some workarounds:

> \def\bal#1\eal{\begin{align}#1\end{align}}

He does not add (e.g.,) "but I do not recommend this for use
in new documents".

he would have if there hadn't been pressure
not to.  he was quite unhappy about having
to allow \beq / \eeq, but recognized that
since ams would have to deal with documents
prepared using the original article.cls,
that was the only way to avoid excessive
amounts of manual cleanup and checking.
(yes, i know that \beq and \eeq *should be*
able to be reliably replaced by an automatic
filter, but authors don't always do the
expected.

it's also true that authors submit articles
using author-defined macros in the title,
abstract, and references.  these must all
be tracked down and expanded, since these
elements are converted to html for posting
to the web and distributing to bibliographic
aggregators, and must therefore stand alone.
and it is *not* feasible to construct filters
that reliably deal with such random macro
expansion, since authors also can be guilty
of redefining commands within the body of
the text, even though they are asked not to.
(i have encountered authors who are quite
proud that they never read user documentation.
some days i wonder why i even bother to take
so much care writing it.)
-- bb