[Discuss] TeX Resources?

Daniel M German dmgerman at uvic.ca
Thu Jul 19 22:46:46 PDT 2007


 Adam Parkin twisted the bytes to say:

 Adam> Alan W. Irwin wrote:
 >> TeX is probably too low level for your needs so go with LaTeX (which
 >> is 

 Adam> I actually meant both, I've been playing with LaTeX so far so that's
 Adam> really what I had in mind.

You really need to know what you are doing to resort to TeX. So skip
it at first.

First I'll state my bias: I used to do Framemaker, and I loved
it. Framemaker had the potential to supplant latex, but it was sold to
Adobe and it has been downhill from there. So I had to learn
--reluctantly--Latex in my MS. Today I can't write papers with any
other tool (Ok, I do, and I hate it). Word is nice with grammar, but
that is all. It is a _pain_ in every single other aspect.

There are 3 beauties of Latex:

1. Separation of content from presentation
2. Easy control of references and bibliography
3. _Beatiful_ typesetting

The Companion is the book to have. I wish it was electronic. You need
to own one.

1: First get an editor/tool/mode that supports LaTeX. I use AucTex (a
   mode for emacs) that provides all the niceties expected for an
   editor to support a: color highlighting, compilation, code folding,
   project management, code completion, etc.

2: Get some papers from your supervisor (I can get you a couple
   if you want). Use them as the basis for your first attempts. 

3: Lamport's book is the best intro (IMO). 

4. Once you have some experience writing with Latex start browsing the
   Companion. The Companion does not teach how to use Latex, it
   teaches how to better use latex.

5. Don't fight latex (at least until you know what you are doing). Let
   latex make the decisions for you (even if you don't like the
   result). Perhaps the biggest change in paradigm is how it handles
   floats. Once you start facing this issue, read the floats section
   in the companion (understand its algorithms and parameters). Also,
   don't fight styles. If the style does not do what you want, find
   another than that does (at least until you build
   experience). Remember, latex is about writing structurally, not
   typographically.

6. Sometimes it is easier to edit the text that fight orphans and
   widows. I find that the best way to get papers to fit the
   restrictions of number of pages, and the way floats are placed is
   by rewriting a paragraph, or merging two paragraphs.

7. When faced with challenging issues, check the companion, and the
   documentation for the module you are using. 

8. UofWaterloo has a TA who is paid to be a latex consultant. I wish
   UVic had a similar position.

9. Collaborate with people using a version control system. Don't use
   the "token ring" approach to collaborative writing. Latex is well
   suited for version control. I have successfully co-written articles
   at the last minute (Abram can attest to that) using svn and cvs.

10. Finally, once you get some skill, try to separate typesetting from
    structure yourself. Learn to write macros and environments, and
    create you own personal modules/styles for the things you
    like. One example: when we have nomenclature in a paper, we use
    macros; if the nomenclature changes (say alpha instead of beta) we
    only have to change the definition of the macro.

dmg


 >> DocBook front-end to LaTeX itself.  For example, if you write your
 >> document in
 >> DocBook, you can generate all kinds of document formats (including
 >> LaTeX .....

 Adam> Thanks for the tip about DocBook, as I had only roughly heard of
 Adam> it. Having done a bit of GooFu and reading up on it though I have
 Adam> concerns that DocBook is a bit too high level.  In particular, it
 Adam> seems to me that DocBook isn't as suited for mathematical formulae,
 Adam> which is something I do need.  Please correct me if I'm wrong though.

No, you don't want to do Docbook. Docbook is best for technical
manuals (not thesis, and particularly not for articles), and it is
really SGML/XML that is typeset via Latex (Latex is the backend).

 Adam> As well, there is a much greater volume of documentation on
 Adam> LaTeX than on DocBook (or so it would seem in my rough search
 Adam> for tutorials & the like).  There's the "Definitive Guide to
 Adam> DocBook" (which seemed like more of a reference than a
 Adam> tutorial), but aside from that I didn't find much.

One of the reasons you are "invited" to learn Latex is that there is a
thesis style at Uvic for latex, and every respectable conference will
provide you a style for Latex. 

 Adam> Most of the praise I've been reading about DocBook seems to be of the
 Adam> "well it's XML so if you have a background in HTML it'll be really
 Adam> easy to read & write compared to LaTeX", which is fair to a degree
 Adam> (LaTeX *is* ugly), but I don't see DocBook being easier to learn *for

No, I don't think so. DocBook is praised because it can be typeset in
a variety of formats: texinfo (nothing to do with tex), LaTeX, HTML,
etc. XML is ugly to write.

 Adam> Having said all that, I am a bit intrigued by the ideas behind
 Adam> DocBook, and will be throwing it into my list of technologies I
 Adam> "should" check out at some point. =8->

 >> astrophysical context.  I am virtually positive there will be some
 >> equivalent of AASTeX that is suitable in a computer science context.

 Adam> I'm pretty sure you're right, I haven't checked yet.  IIRC, the CSC
 Adam> department has a LaTeX thesis template as well, but again I haven't
 Adam> checked yet (given that I just installed LaTeX today). =8->


dmg


 Adam> -- 
 Adam> --
 Adam> Adam Parkin
 Adam> E-mail: pzelnip at gmail.com
 Adam> Blog: http://pzelnip.blogspot.com/
 Adam> ----------------------
 Adam> Hey, this is America. Knowing nothing and choosing one of two options
 Adam> isn't a poll. It's an election.
 Adam> 	-- Bill Maher



-- 
--
Daniel M. German                  
http://turingmachine.org/
http://silvernegative.com/
dmg (at) uvic (dot) ca
replace (at) with @ and (dot) with .


More information about the Discuss mailing list