Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!att!ucbvax!mtxinu!jaap
From: jaap@mtxinu.COM (Jaap Akkerhuis)
Newsgroups: comp.text
Subject: Re: Good WPs for books?
Message-ID: <1991Mar29.194547.18335@mtxinu.COM>
Date: 29 Mar 91 19:45:47 GMT
References: <1991Mar23.145145.23613@canon.co.uk> <1571@aupair.cs.athabascau.ca> <1048@hrshcx.csd.harris.com>
Reply-To: jaap@mtxinu.COM (Jaap Akkerhuis)
Organization: mt Xinu, Berkeley
Lines: 27

In article <1048@hrshcx.csd.harris.com> steved@hrshcx.csd.harris.com (Steve Daukas) writes:
 > [among other things]
 > 
 > Another reference to the useability of roff was in the preface of a Unix
 > text.  The authors comments basically said that it was far too difficult
 > to get troff to do what they needed and what the publisher wanted.  They
 > concluded with the following statement:  "we have concluded that there
 > must be a better way."  This is paraphrased (I'm not in my office at home, so
 > I can't find the text), but I will provide the actual reference if you desire.
 > 

This remark is the 4.3BSD book by Leffler at all. and is citated
completely out of context. It refers to the fact that bottoming
out the pages wasn't easy and had to be done by hand. Implementing
the design wasn't too bad.

Since I was instrumental in getting the book on the shelves, I can
assure you that producing the book (with the same design) using
TeX, Scribe, Frame, Word, Interleaf etc.  would have been just as
complicated or likely even more complicated (and in some systems
impossible), due to some of the details in the specs.

The bottom line here is that when there is a design which has
features different then the design as build in the typesetting
programs, one has to do a lot of work to adapt the programs/macros.

	jaap