🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: comp.text.tex
Article #664

Re: TeX methodology

#664
From: Ajay Shah
Date: Sun, 02 Aug 1992 00:49
55 lines
3249 bytes
Richard.Kooijman@dnpap.et.tudelft.nl (Richard Kooijman DNPAP) writes:

>rcoadh@melomys.co.rmit.oz.AU (Alexander Doktorovich) writes:

>>  Thirdly, the comment: I believe the ASCII interface is outdated. It may have
I think you have consumed too much hype from the industry :-)

>By using that ASCII interface, I can work on LaTeX via modem from home,
>at work, everywhere.
A very valuable feature, IMHO.

>Frankly, I don't care much for a graphical interface. Nowadays, with
>modern wordprocessors you still are tempted to fiddle to get things right on
>paper and with LaTeX you just don't (can't), and you stick to writing a
>report logically.
Absolutely.  TeX encourages you to concentrate on the logical
structure of your document.  Most of the chores of typesetting are
done invisibly, by extremely smart code.  As a rule of thumb, the
typical document produced using LaTeX by a novice is far more
beautiful than the same document produced by him using a GUI editor,
since he's likely to have tried his hand at twiddling some buttons.

There are deeper issues in quality too.  I keep getting shocked at the
weaknesses of (say) MS Word's inter-character spacing and hyphenation.
Considering the market they are targeting, they probably don't need to
fix these weaknesses in a hurry too.  These vendors have a way to go
before they know typesetting as well as Don Knuth.

>The ASCII interface has more advantages than those I mentioned above.
>You can use all kind of generic utilities that can read and handle
>ASCII files: ispell, awk, grep and of course your favorite editor!

Yes!  You can use the editor of your choice.  You can integrate the
process with `make'.  You can write programs which extract things from
inside your documents, and you can write programs which generate TeX.

These are not abstract possibilities.  E.g. I routinely turn results
of a simulation into LaTeX using a awk filter.  Similarly, my `ols'
program (a multiple regression Unix tool, on comp.sources.reviewed)
has a option to produce LaTeX.  I have used this option with much joy,
and have got enthusiastic email about it from other users.

Perhaps the funniest story concerns the index.  I won't explain the
gory details, but I needed to do something repetitive to the index.
The cheapest solution turned out to be a awk filter which worked on
the .ind file after makeindex has done his job.  It sits in the
makefile, cleaning up the index every time I say "make main.dvi".

Thus I think of programs which read and write textfiles as being Open,
and that is a extremely important criterion for me.  Now if only dvi
files were stored in plaintext (like postscript is) ...

        -ans.
-- 
Ajay Shah, (213)749-8133, ajayshah@usc.edu
Technical Details
Message-ID: <l7mc9fINN5ef@almaak.usc.edu>
Path: arc1-ca.pugleaf.net!sparky!uunet!caen!sol.ctr.columbia.edu!usc!news
References: <rcoadh.712658976@chudich.co.rmit.oz.AU> <richard.712705651@dutepp13>

Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!caen!sol.ctr.columbia.edu!usc!news
From: ajayshah@almaak.usc.edu (Ajay Shah)
Newsgroups: comp.text.tex
Subject: Re: TeX methodology
Date: 1 Aug 1992 17:49:51 -0700
Organization: University of Southern California, Los Angeles, CA
Lines: 55
Sender: ajayshah@almaak.usc.edu (Ajay Shah)
Message-ID: <l7mc9fINN5ef@almaak.usc.edu>
References: <rcoadh.712658976@chudich.co.rmit.oz.AU> <richard.712705651@dutepp13>
NNTP-Posting-Host: almaak.usc.edu