Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!samsung!usc!rutgers!mcnc!duke!romeo!crm
From: crm@romeo.cs.duke.edu (Charlie Martin)
Newsgroups: comp.text.tex
Subject: Re: frequently asked questions
Message-ID: <17972@duke.cs.duke.edu>
Date: 5 Mar 90 12:39:36 GMT
References: <1990Feb25.182757.23500@cs.rochester.edu> <22450@uflorida.cis.ufl.EDU> <814@lectroid.sw.stratus.com> <1513@loria.crin.fr>
Sender: news@duke.cs.duke.edu
Reply-To: crm@romeo.UUCP (Charlie Martin)
Organization: Duke University CS Dept.; Durham, NC
Lines: 37

In article <1513@loria.crin.fr> hermann@loria.crin.fr (Miki HERMANN) writes:
>The solution of the first question is  easy. It requires  only to read
>Lamport's  LaTeX manual properly (Rule  of thumb: If  it does not work
>for  the  n-th  time, read the  reference manual!). The explanation is
>given   on  the page   155.  The \baselinestretch   parameter does the
>required work.  You can use it in the following way:
>
>AFTER the \documentstyle..., but BEFORE \begin{document} you write
>
>	\renewcommand{\baselinestretch}{2}
>
>for doublespaced documents. As it is explained in  the manual, you can
>use a decimal number (1.5, 2, 3.14, etc) for \baselinestretch.

And, just like the LaTeX book says, doing so results in a really ugly
hacked version of double spacing.  (Okay, I admit Lamport didn't say it
quite that way....)

If you have to produce a finished document that is doublespaced, in
LaTeX, you should really modify the style in much more significant ways:

(1) Pick an appropriate baselineskip value.  I've been using 12pt type
on 20pt centers, so \baselineskip 20pt plus 2pt minus 1pt works pretty
well.

(2) Adjust the before and after spacing on sections, subsections, etc.

(3) Adjust the space before and after for floats.

(4) Make sure the table of contents etc uses a smaller center:
doublespaced TOC etc looks funny.  (The Dragon At The Graduate School
Gate here will buy a single-spaced TOC even though the text must be less
grey.....)

But simply setting baselinestretch once and for all is not an adequate
solution.
Charlie Martin (crm@cs.duke.edu,mcnc!duke!crm)