Re: [Axiom-developer] [#210 Pamphlet support on MathAction] [#210 Pamphlet support on MathAction] [#210 Pamphlet support on Mat

[michel . lavaud]

On 7 Oct 2005 at 21:49, root wrote:

> Changes
> http://page.axiom-developer.org/zope/mathaction/210PamphletSupportOnMa
> thAction/diff -- Actually, I continue to maintain the position that
> documents should be available in SOURCE form, not pdf, not dvi, etc.
> Only source allows you the ability to use the information in any way.

Yes, of course, I was comparing pdf and dvi.

However, I don't agree with you when you say source form, not dvi. In my 
opinion, it is
necessary that documents be available in both forms, source _and_ dvi. Indeed, 
the
source form is potentially ambiguous because it can be rendered differently on
different installations (cf example below). This is a consequence of the fact 
that a
source document in TeX is in fact a computer program.

On the contrary, a dvi file is not a computer program, it is a sequence of 
instructions
that describe how glyphs are placed on pages, It is  the electronic equivalent 
of a set of
sheets of paper. It is completely freed from any ambiguity that could arise 
from the
programmatic nature of the source document, and it is rendered the same way on 
all
machines. Once it has been checked by the author (as one does with galley 
proofs for
paper articles) it is not necessary to check it anymore. It is unfalsifiable, 
as a printed
article in a review.

A simple example : some years ago, a buggy version of a TeX style file for 
French
language, broke the rendering of fractions. As it is possible to include any 
set of TeX
commands in a TeX format, this means that any French user who would have 
included
this file into his formats (to accelerate compilations) would have an 
installation that
could display incorrectly some results.Of course, this is a very extreme and 
improbable
example, but nothing can rule it out, nor rule out more subtle differences 
between
formats that could make a document rendered differently on different 
installations
Another problem with source files : referencing equations is done not by 
numbers but
by labels, so an author who would want to cite a result of an article could not 
use
"Equation number 12 page 32".

For users (i.e. not the author), source file + dvi file allows double checking. 
For
authors, source files are the equivalent of the drafts that one writes when 
doing
research work, and dvi files are  the equivalent of published articles, for the 
outer
world. Both are necessary, in my opinion.

Best wishes,

address@hidden
http://www.univ-orleans.fr/EXT/ASTEX
ftp://ftp.univ-orleans.fr/pub/tex/PC/AsTeX
liste de discussion: address@hidden
Abonnement à la liste: envoyer un message de contenu
"sub astex Nom Prenom Etablissement" à address@hidden