Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!rutgers!sri-spam!ames!ucbcad!ucbvax!decvax!decwrl!pyramid!prls!philabs!tg!scott
From: scott@tg.UUCP
Newsgroups: comp.text
Subject: Some ditroff deficiencies, could they exist in TeX?
Message-ID: <117@tg.UUCP>
Date: Mon, 2-Feb-87 03:45:04 EST
Article-I.D.: tg.117
Posted: Mon Feb  2 03:45:04 1987
Date-Received: Wed, 4-Feb-87 03:27:34 EST
Organization: Townsend Greenspan & Co., Inc.  New York, NY
Lines: 51
Keywords: phototypesetters, publishing

As I remember writting the ditroff output driver for a VideoComp 500
typesetter, I remember some inadequacies that I would like to know if
they exist under TeX.

1) (This may not be in TeX) But in the more highly distributed WWB (and I
think fixed in the bsd version), the ability to define a character as
slanted does not exist!  This problem only occurs on phototypesetters
(not lasers) that sell Italic fonts not slanted and allows the users to
slant these characters to meet their needs.  It is impractical to put
the \S command for all italicied fonts and there is noway to control
this when jumping in and out of math.  The solution we came up with is a
bit of a kludge, but works.  I wonder if I can specify a 15-degree slant
on all Italic charaters in my tfm.

2) Not all characters on a phototypesetter exist on a single physical
font.  For example, the VideoComp typesetter has subsets of a physical
font (e.g. the Times Roman is font 2 with subsets 0-2) and things like
ligatures and some necessary special characters (question marks, etc)
appear in these subsets.  How does one specify these subsets without
specifying these characters as specials (imagine the search of special
fonts for all the periods) and not breaking ditroff.  This again was a
bit of a kludge (but could have been done better--always known after the
fact :-}).  Can a "forwarding" of fonts be specified in TeX's tfm files
to point to these subsets as being part of the Roman but on a
"different" font?

3) This one has not be figured out (read I am not working on this
anymore :-}) but needs to addressed by people using phototypesetters.
How can different ranges of a font be defined.  On a phototypesetter, a
character is designed to be printed on a certain range of point sizes.
Printed on a smaller or larger ps can cause the character to not be
printed properly.  This is controlled by having different font ranges.
Now if your "normal" Roman font prints in range 4-28, how can you ask
for a different range of the roman font if you are, say, doing book
covers in point size 48 in the roman font?  Defining it as a different
font is a solution (not the optimal), but I would like to be able to use
the full range of point sizes without having to have the keyboarders
have to enter different things during text entry?  Any thoughts?

My only other question has to do with usage of troff or TeX.  Is there
anyone in netland using TeX or troff in publishing--either comercial
book/magazine publishing or the publishing of journals for non-profit
organizations?  I have said in the past that the Americal Physical
Society publishes 45,000+ pages per year in many journals using the old
C/A/T troff (being phased out) and ditroff.  So if you know of any
others, I would like to know!

Thanks!

Scott Barman
{philabs, pyrnj}!tg!scott