Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!rutgers!njin!princeton!phoenix!tbrakitz
From: tbrakitz@phoenix.Princeton.EDU (Byron Rakitzis)
Newsgroups: comp.text
Subject: Re: Spacing in math mode in LaTeX
Message-ID: <8501@phoenix.Princeton.EDU>
Date: 17 May 89 23:14:14 GMT
References: <433@pbseps.UUCP> <3847@utastro.UUCP> <1404@uw-entropy.ms.washington.edu> <2286@csc.anu.oz> <1431@uw-entropy.ms.washington.edu> <1292@jarthur.Claremont.EDU> <1436@uw-entropy.ms.washington.edu>
Reply-To: tbrakitz@phoenix.Princeton.EDU (Byron Rakitzis)
Organization: Princeton University, NJ
Lines: 52

In article <1436@uw-entropy.ms.washington.edu> charlie@mica.stat.washington.edu (Charlie Geyer) writes:
>
>In article <1292@jarthur.Claremont.EDU> dhosek@jarthur.UUCP
>(Donald Hosek) writes:
>
>> of the final copy rather than the author. I read (I think in TeXline) that
>> some journals find it more economical to re-key articles than try and deal
>> with author-supplied manuscripts which are rife with all sorts of spurious
>> fine-tuning!
>
>but to lists, headings, figures and tables, and so forth.  Many people
>eventually learn to let the program do it it's own way.  Some people
>don't have that much self restraint.  The just have to fiddle.
>


Well, I don't know about y'all, but I certainly learned my TeX by
"fiddling". (That should be `fiddling' ! :-) ) In any case, the answer
isn't the avoidance of all "spurious" control sequences, but rather
the rigorous implementation of user-defined control sequences. 
That is an ugly turn of phrase. What I meant by it was: (like Humpty Dumpty!)
"If you're going to fiddle, put all your `fiddly' control sequences in
\def statements at the top, and stay away from TeX primitives (even
plain TeX `primitives') inside the body of a document."

This way, in the worst case the editor can \let all your macros
\relax, so to speak. 

Knuth himself warned against the practice of repeatedly using
primitives (in the TeXbook, somewhere close to the beginning).

If you need to make a macro for sets or the infinitessimal in the
integral, do something like:

\def\setbraces#1{\{\,#1\,\}}

or 

\def\dtheta{\,d \theta}

or whatever catches your fancy. You'll probably find your manuscript
to be much more `readable' (i.e., easy to take in at a glance. Is
there a word for that? I don't think `legible' quite fits the bill,
does it?) as a result.

Cheers.

-- 
"C Code."
"C Code run."
"Run, Code, run!"
Byron Rakitzis. (tbrakitz@phoenix.princeton.edu ---- tbrakitz@pucc.bitnet)