[texhax] MS Word & Mathtype to TeX

Barbara Beeton bnb at ams.org
Mon Dec 19 15:25:01 CET 2011

let me try to explain why peer-reviewed
journals request (at least some of them
don't *require*) tex.

you want to try to live in a low-impact,
self-sufficient world, with, ideally, low
cost to accompany your reduced footprint.

most math journals these days cannot afford
the cost of skilled professional composition.
the route they have chosen to avoid this is
to ask authors to use a tool created by a
competent mathematician/computer scientist to
produce his own works to the standard of quality
he expected from traditional composition.
(after all, an author knows more about his/her
subject than the best compositor.)

in order to reduce costs further, to keep
subscription prices affordable (and here, i
can speak only for journals published by
professional societies, academic institutions,
and other not-for-profit organizations), and
to make possible publication in electronic form
in parallel with the printed copy, publishers
require a uniform, predictable, and reliable
input source.  latex has proven itself to be
such a stable format.

at the american mathematical society, submissions
are peer-reviewed before they come to the office
for production.  what the reviewers and subject
editors accept for publication, we at the office
have to deal with.  if an author is unable to
produce a file in the "normalized" form required
for production, then it will be modified by a
skilled keyboarder (someone on our own staff, if
available, but sometimes "contracted out"), or,
in the worst case, rekeyboarded.  this is very
much more expensive than handling a file created
by an author knowledgeable in latex who has read
and adhered to the guidelines.  with some extra
effort by the author, everyone benefits in the
end because of lowered costs and, perhaps more
importantly, the decreased likelihood of
introduced errors.

so this is not meant as a conspiracy against
microsoft or any other commercial software
purveyor.  the fact is, a math book carefully
prepared in latex is "stable".  the original
files can be pulled out of an archive ten or
twenty years later, updated with new content,
and reprocessed with minimal attention to
whether the software will produce the same
results.  similarly for articles repackaged
into collections -- reprocessing through this
reliable software produces vastly better (and
more compact) results than digital scans of the
originals, which is the other popular method.

no one will try to persuade you that latex is
something that can be learned in a couple of
days.  it *does* require attention and effort.
but there are many who argue that it *is*
worth the effort.
						-- bb

More information about the texhax mailing list