mailing-list for TeXmacs Users

Text archives Help


Re: Feedback


Chronological Thread 
  • From: Henri Lesourd <address@hidden>
  • To: "Bear F. Braumoeller" <address@hidden>
  • Cc: address@hidden, address@hidden
  • Subject: Re: Feedback
  • Date: Wed, 24 Jan 2007 15:08:06 +0100

Bear F. Braumoeller wrote:

Hello all,

I'm responding to the requests for feedback on your website. I've just spent the afternoon installing and working with TeXmacs, and I have to say, I'm very impressed! For a long time, I have wished for a simple LaTeX editor that would let me work directly with the text, rather than forcing me to plod through lines and lines of code. I hadn't realized that TeXmacs existed, and I was very excited to get a chance to try it out. Thank you, very much, for putting so much time into such an impressive product.

I doubt that I will end up being able to use it, unfortunately, for the simple reason that doing so would involve a massive citation re-formatting effort that would require far too much time. Let me explain, in the hopes that this will be of some use to you.

When I first succeeded in getting citations to work, I ran into the same problem that Javier Arantegui did in this exchange on your list:

http://lists.texmacs.org/wws/arc/texmacs-users/2004-10/msg00036.html

I recognized the problem, however, because I've run into it in LaTeX before. I use chicago.bst, but in order to make it work properly I must load chicago.sty at the beginning of the document. The chicago.sty file redefines many citation and bibliography entries, so that they look correct when they are printed. Without it, chicago.bst creates terrible-looking entries in both the citations and the bibliography. (I have attached a copy, in case you are curious about exactly what it does.)

With chicago.sty, I can use commands like \citeN[123]{foo99} to get Foo (1999, 123), or \cite[123]{foo99} to get (Foo 1999, 123), or \citeNP[123]{foo99} to get Foo 1999, 123 -- you get the idea. And the bibliography is formatted just as political science editors want it to be, in particular, without bibliography labels. There are some other packages that do similar things, but all of them require some sort of .sty file at the beginning: at a minimum, they need \def\@biblabel#1{} in order to get rid of the bibliography labels. My guess is that the need for some sort of .sty file at the beginning will be very common for social scientists, actually.

It's true that in your case, one probably needs to redefine some
predefined TeXmacs macros (e.g. \cite{}), although I'm not even
sure that since 2004, things didn't changed in a way which would
obviate this (recurrent) problem of using different bibTeX styles.

This being said, it should be noted that bibTeX is wholly reused
as a full component of the TeXmacs software : it means that quite
often, as a matter of fact, problems with formatting the bib entries
are to be solved at the level of bibTeX, they often have nothing
to do with redefining/adapting TeXmacs itself. This comes in
stark contrast to the connection with LaTeX, whose functionnality
is reimplemented from scratch in the TeXmacs rendering
engine : thus currently, we use 0% of LaTeX inside
TeXmacs (except for the fonts, although one can use
only Type 1 fonts, and thus become completely independent
of this part of LaTeX, too).


I don't even think that I understand the LaTeX style file all that well, let alone how I could re-render it as TeXmacs! The only thing that comes to mind is a LaTeX-to-TeXmacs style file conversion program, which I would think (?) would be straightforward, given that TeXmacs imports LaTeX files, but I have no idea whether the style file syntaxes are even remotely comparable.

Automatically translating LaTeX style files seems easy at first, but it appears
that it is a quite difficult problem, because of the need of also importing
the pure TeX commands that can be found in lots of LaTeX style files.

The TeX language being a quite low-level language, it is hard to translate
it in a high-level language like the TeXmacs macro language.

What we would need in order to increase the speed of spreading TeXmacs
would be a (handmade) TeXmacs implementation of each one of the
common LaTeX style files (especially the ones from the publishers).

And for this we need volunteers !

This being said, to my knowledge (correct me if I'm wrong) purely syntactic
LaTeX macros (i.e., the ones which define a purely syntactic macro expansion)
that can be found in the preamble of a LaTeX document definitely **can** be
imported in TeXmacs.


I always tell my students that, when they raise problems, they should also be willing to offer solutions. In this case, I fear that I don't know enough about your program to offer any intelligent solutions. I suppose it might be possible to convert chicago.sty over to a TeXmacs style file, but I took a look at the "writing your own style file" page, and frankly, it's quite daunting.

It should not be so complex, because TeXmacs macro language is exactly similar to
purely functional s-expressions, thus some basic knowledge about Lisp or functional
languages should kick-start you very easily.

Perhaps the following tutorial:
<<
http://www.ags.uni-sb.de/~henri/texmacs/aTeXmacsTutorial.pdf
>>

, which exclusively proceeds by means of examples you can do by
yourself while reading the tutorial, would fit your needs better.

I would be pleased to hear more news / opinions from you about that, because
as far as non-"hard science" people (i.e., maths / physics / computer science
people), are concerned, we suspect since quite a long time now that our
tutorials are perhaps not completely adapted, they are perhaps still a little
bit too much in the boring style of the UNIX man (i.e., "read everything and
forget nothing, otherwise you can't understand") . Thus, any hint / info in this
respect would be very useful (although the next problem, then, would become
finding the time to do the job of improving the tutorials : but the 1st step is to
know what to do).


As usual, any other comments / criticisms / dialectic points of view are
hotly welcomed.



  • Re: Feedback, Henri Lesourd, 01/24/2007
    • Message not available

Archive powered by MHonArc 2.6.19.

Top of page