Message-Id: <9305031347.AA13824@sc.zib-berlin.dbp.de>
Reply-To: Mailing list for the LaTeX3 project <LATEX-L@vm.urz.Uni-Heidelberg.de>
In-Reply-To:  <01GXPWF9I1IQI8SCZI@MATH.AMS.ORG>
Date:         Mon, 3 May 93 15:46:13 CET
From: Michael Downes <MJD@MATH.AMS.ORG>
Sender: Mailing list for the LaTeX3 project <LATEX-L@vm.urz.Uni-Heidelberg.de>
To: Multiple Recipients of <LATEX-L@vm.urz.Uni-Heidelberg.de>
Subject:      Re: documentstyle option versions
Status: R


Cameron Smith wrote:

> Michael Downes asks how one may guard against "subtly wrong results"
> when A sends B a LaTeX document that uses a non-standard but widely
> popular foobar.sty and B uses an older version of foobar than A used
> in creating the document.
...
> I ask a counter-question: what kind of "subtly wrong results" are we
> worried about?  If they're that "subtle", what difference do they make?
>
> LaTeX is supposed to be a structural rather than a visual markup
> system. ...  If a change in a style file merely causes a paragraph
> to re-flow or a figure to be re-positioned or some other "subtle"
> *visual* incompatibility, then I claim that that is not a problem.

No, in my lexicon "wrong" and "different-looking" are not synonymous.
But I acknowledge it's reasonable for you to question that particular
point, since you don't have a copy of my lexicon (and if you did your
version probably wouldn't be in synch with my version :-).  By "subtly
wrong results" I really meant wrong results, not just
different-looking; but not so glaringly wrong that they would
necessarily be noticed in a casual skimming of the output. For
example: Omission of all \cite's in the body of the text.  Printing
all cites as number 1 instead of different numbers.  Minus signs in
math formulas printing as cap Gammas. Disappearance of all cap Greek
letters in math formulas. Omission of equation numbers.  Quotations
losing their formatting so that they are indistinguishable from the
surrounding text. Changing of boldface variables in math formulas to
math italic.  Elements in a picture environment that were supposed to
be equidistant coming out non-equidistant (not by a pixel or two, from
output device differences, but by a significant amount that renders
the picture nonsensical). Omission of all footnotes.

LaTeX is a complex system, so there are more ways for things to go
wrong than anyone can begin to imagine. I don't think LaTeX document
interchange has major problems in practice---the methods for managing
interchange work remarkably well---but I have seen real cases where
things went wrong.  What if you submit your pride-and-joy article to a
publisher and see it come out in print with a deficiency along the
lines of the ones listed above, because the publisher had on hand a
different version (maybe older, maybe newer) than you of one of the
documentstyle options that you used? And you didn't spot the problem
beforehand because you believed that since you submitted the document
electronically it wasn't necessary to re-proofread every word and math
formula? It's rather difficult to unpublish a corrupted document when
the medium is the printed page.

Here's a hypothesis: If you submit a document to a publisher in .tex
form rather than .dvi form, either the publisher must be able to
replicate your LaTeX system exactly (TeX version, LaTeX version+date,
versions of all documentstyle options), or the document must be
re-proofread word for word after it has been run through the
publisher's in-house processing.

The intent of my postings is not to claim that the above hypothesis is
correct (maybe it is, maybe it isn't) but to provoke discussion of
ways to make the interchange process more failsafe.

>  As long
> as the text of the document isn't mangled (and what kind of style file
> change would actually cause text to be lost?) ...

Style file changes are usually done by human beings.

Michael Downes                              mjd@math.ams.org (Internet)