TeXhax@cs.washington.edu (TeXhax Digest) (07/04/89)
TeXhax Digest Friday, June 9, 1989 Volume 89 : Issue 58
Moderators: Tiina Modisett and Pierre MacKay
%%% The TeXhax digest is brought to you as a service of the TeX Users Group %%%
%%% in cooperation with the UnixTeX distribution service at the %%%
%%% University of Washington %%%
Today's Topics:
Printing LaTeX postscript files to an LPS40
Request for modified TeXtures LFONTS file
Down-sizing fonts for footnotes in Textures
TeX T-Shirts
Re: TeXhax Digest V89 #47
RE: Oldstyle digits in small caps font
Importing TeX output into WordPerfect
Underlining in LaTeX
A LaTeX Bug?
Re: Comments on TeX
Comments on TeX
EOF problems in TeX 2.95
Running TeX in batchmode
----------------------------------------------------------------------------
Date: Tue, 6 Jun 1989 15:09:32.62 CDT
From: <horn%SHSU.BITNET@UWAVM.ACS.WASHINGTON.EDU> (James T. Horn)
Subject: Printing LaTeX postscript files to an LPS40
Keywords: LaTeX, PostScript, LPS40
We have just installed an LPS40 onto a VaxCluster and was
wondering
if anyone is having problems, and if not how do you create the
postscript file to submit to the LPS40 postscript queue?
Any help would be appreciated.
James T. Horn
Sam Houston State University
HORN@SHSU.BITNET
-------------------------------------------------------------------------
Date: Tue, 6 JUN 89 07:45:15 PDT
From: "Micro Mauler" <MICRO2.SCHWER@CRVAX.SRI.COM>
Subject: Request for modified TeXtures LFONTS file
Keywords: TeXtures, LFONTS
I am looking for a suitably modified version of the
TeXtures (MacIntosh TeX/LaTeX application) LFONTS file that will
use (download) the PostScript Apple LaserWriter+ fonts, in
particular Times-Roman, rather than the default CMR fonts.
If you would be willing to share said same, please
respond to:
Len Schwer Micro2.Schwer@crvax.sri.com
P.S. BinHex/StuffIt format would be fine.
---------------------------------------------------------------------------
Date: Tue 6 Jun 89 11:50:55-PDT
From: David Kreps <F.KREPS@GSB-WHY.Stanford.EDU>
Subject: Down-sizing fonts for footnotes in Textures
Keywords: TeXtures, fonts, footnotes
Help! I am using TeXtures to compose a book, in which I am using
Palatino for all the text and Computer Modern for the math.
Everything is fine except that when I downsize all the fonts for
footnotes, the numbers used in the math expressions (which are being
taken from cmr) are coming in at cmr10 (instead of cmr8). Since I'm
using Palatino at 8pt for the text, I'm completelyat a loss what to
do. Any suggestions? Thanks.
--------------------------------------------------------------------------
Date: 2-JUN-1989 10:16:40 GMT
From: FPS%VAXA.CC.IMPERIAL.AC.UK@UWAVM.ACS.WASHINGTON.EDU
Subject: TeX T-Shirts
Keywords: general
cf Knuth, The TeXbook, p483 `You can buy TeX T-shirts at these meetings'
(i.e. TUG meetings). as usual, all life's answers are contained in The Book.
i bought mine at the '84 TUG, and although faded, it is an object
of much coveting.
i believe that doug henderson usually does them, but this is
only a rumour. i'm going to suggest to TUG that they start
selling TeX and MetaFont t-shirts. i hope you're a TUG member.
(in fact, i hope that all TeXhax recipients are TUG members --
if not WHY NOT? we do want to know.)
malcolm clark
---------------------------------------------------------------------------
Date: 4 Jun 1989 0736-PDT (Sunday)
From: lamport@src.dec.com (Leslie Lamport)
Subject: Re: TeXhax Digest V89 #47
Keywords: laTeX, environments
Jon Warbrick writes:
Within an enumerate enviroment if the text for an `item' is in a center
enviroment then the lable for the item is centered along with the text.
The following demonstrates the problem:
In a list, the label is part of the first line of the first paragraph
of the item.
Leslie Lamport
---------------------------------------------------------------------------
Date: Mon, 5 Jun 89 09:08:40 PST
From: WMO@muon.JPL.NASA.GOV
Subject: RE: Oldstyle digits in small caps font
Keywords: Oldsyle digits, fonts
Peter Flynn (TeXhax V89 #12) asks "if it would be possible to fudge things
so that text in small caps (cmcsc10) would have \oldstyle numerals."
My recommendation is that he scavenge the METAFONT CM source code and
build his own font. All he has to do is to change "romand" to "olddig"
in the tenth line of csc.mf and then regenerate the fonts in as many
sizes and magnifications as he likes. This will alleviate the problem
of trying to get TeX to paste fonts together behind the scenes.
----------------------------------------------------------------------------
Date: Fri, 2 Jun 89 11:39 CDT
From: CHRDCWW%engvms.unl.edu@UWAVM.ACS.WASHINGTON.EDU
Subject: Importing TeX output into WordPerfect
Keywords: TeX, WordPerfect
Does anyone know of a way to take either *.dvi files or HP LaserJet specific
print files (*.hp) and convert them to a graphics file that is compatible
with WordPerfect 5.0 (a *.wpg file)? The primary intent is to incorporate
mathematical equations created by TeX (or LaTeX) into WordPerfect documents.
Clifford W. Walton
CHRDCWW@ENGVMS.UNL.EDU
University of Nebraska-Lincoln
Department of Chemical Engineering
236 Avery Laboratory
Lincoln, NE 68588-0126
(402) 472-2751
--------------------------------------------------------------------------
Date: Fri, 02 Jun 89 19:01:50 GMT
From: Vivian Harrington <HARRINGT%IRLEARN.BITNET@UWAVM.ACS.WASHINGTON.EDU>
Subject: Underlining in LaTeX
Keywords: LaTeX, underlining
A collegue here wishes to underline several continuous lines in a
document. The underline command is fragile and does not work when
more than one line is to be underlined. Is there any simple way
around this difficulty ? thanks in advance
Vivian Harrington
Advisory Service, University College Dublin
----------------------------------------------------------------------------
Date: Fri, 2 Jun 89 13:39:30 EDT
From: toms@ncifcrf.gov
Subject: A LaTeX Bug?
Keywords: LaTeX, bug
Hi TeXhaxers!
If I run the newest LaTeX:
% /users/beav/bin/virtex '&lplain' notthere.tex
and the file does not exist, it will say
This is TeX, C Version 2.95 (no format preloaded)
! I can't find file `notthere.tex'.
<*> &lplain notthere.tex
Please type another input file name:
If I then type a control-c, it will ignore me. If I type a control-d, it goes
into an infinite loop that cannot be broken by either control-d or control-c!!
I have to kill the process. We just set up this version of TeX, on a Sun 3/260
running Unix. Is this a bug or is something odd about the way we set it up?
Tom Schneider
National Cancer Institute
Laboratory of Mathematical Biology
Frederick, Maryland 21701-1013
toms@ncifcrf.gov
-------------------------------------------------------------------------------
Date: Fri, 2 JUN 89 13:49:01 BST
From: CHAA006%vaxb.rhbnc.ac.uk@NSFnet-Relay.AC.UK
Reply-To: Philip Taylor (RHBNC) <P.Taylor%vaxb.rhbnc.ac.uk@NSFnet-Relay.AC.UK>
Subject: Re: Comments on TeX
Keywords: TeX
Michael Barr <INHB%MCGILLC.BITNET@UWAVM.ACS.WASHINGTON.EDU> recently sent some
opinions of TeX. Whilst I would be the last person to suggest that TeX is
perfect, and could never be improved, I really feel that Michael is going over
the top with some of his criticisms, and that others are rather missing the
point:
>For example, I think that TeX's arithemtic is terrible. The
>inability to give an expression as numerical parameter is very
>troublesome.
Surely what is wrong in this area is the lack of orthogonality: in some
contexts, 2\x will multiply the current value of "x" by 2; in others
it will concatentate "2" and the current value of "x". Very confusing
for beginners. If I set \baselineskip = 12 pt plus 1 pt, and set
\parskip = 2\baselineskip, as a beginner I might reasonably expect \parskip
to hold the value 24 pt plus 2 pt; in practice the <skip> is coerced to a
<dimen> and \parskip ends up as 24pt, with no stretch or shrink.
> Macro writing would be much simplified if you had the
>ability to look ahead at the next token (\futurelet gives an awkward way
>of doing this, but is not really satisfactory).
In what way is \futurelet unsatisfactory ? The only restriction I've
found is that it (necessarily) staticises the \catcode of the following
character(s) [just sufficient characters to form the next <token>].
[Afterthought: is that true; does it perhaps staticise the \catcode
of at least one more character ? If not, how does it ``know'' that
that character can't form part of the next <token>. ?]
> Numerical (not just
>integral) and string variables would be extremely useful. (I don't
>understand how token variables work, but they don't do what I would
>want.)
Real variables: yes, I can certainly see a use for those, but string
variables: well, I find that token variables do what I want, but then
Michael admits that he doesn't understand how they work.
>
>My publisher wanted the book I am writing set in such a way that if
>facing pages are the same length, then the left hand page is the longer.
Then your publishers want miracles, or are working in a non-Euclidean
geometry.
>I'm not sure why they consider this important, but I am not a book
>designer. At any rate, I had to tell him that if they insist on that,
>then they are going to have to typeset it in the traditional way, for I
>see no way of coercing TeX into doing that. I don't say it can't be
>done; I suppose it might be possible to make TeX think it is doing
>two-column style and cut the page myself, but this is far beyond my
>ability of TeX programming. I also think I would run out of memory.
If, on the other hand, we assume that what your publisher really wants is
that if facing pages are NOT the same length, then the left-hand page should
be the longer, then I would ask him/her how you are to deal with chapters
which end on a left-hand page. They will frequently be considerably shorter
than chapter opening pages on the corresponding right-hand page. However,
once you have eliminated the special cases, a modified output routine should
be capable of achieving all that you require; why should you think you
"would run out of memory" ?
>
>Finally, I don't disagree with his remarks on hand-formatting. In fact,
>I think that TeX still leaves too much of it to the user.
TeX doesn't "leave too much to the user"; it allows the user to "[tell] a
computer exactly how the manuscript is to be transformed into pages whose
typographic quality is comparable to that of the world's finest printers;"
[Knuth: ``The TeXbook''; Addison-Wesley 1984]. That is exactly what TeX
sets out to do (and what, in my opinion, LaTeX seeks to take away again).
>I have
>published a book that includes some 600 diagrams, all done with LaTeX
>picture mode. But the Latex book suggests that the way to prepare
>diagrams is to lay them out on graph paper and read off the coordinates!
>This is as far from logical typesetting as possible.
On the contrary, it is extremely LOGICAL typesetting, but I suspect that
isn't what you want. Instead of saying "I must have a line of length
7.4pc, with relative origin (0,0) and orientation pi/4 radians", I think
you want to be able to say "I want a line that looks like this:"; that's
fine for some diagrams, and useless for others. It's a bit like the
"pull it around" approach of mouse-based document preparation: when the
chap came here to demonstrate Aldus PageMaker, he pulled in a half-tone
image, which he then shrank BY EYE to fit the column width; when I asked
him "but why can't you just TELL PageMaker to shrink it to EXACTLY fit the
column", he said: "Oh, you can't do that; you just keep magnifying the
display until you get it as accurate as you need". Do you want to work in
a world like that ? I don't.
So, I 'm sorry to be critical of Michael's criticisms, but TeX can't defend
itself (and Knuth is probably far too busy to defend it). Despite that,
I really would like to see an ongoing discussion of real deficiencies in
TeX, and suggestions as to how in a future system (UfY ?) these deficiencies
might be resolved. I'm specifically not suggesting that we discuss how one
might implement a totally new document preparation system, nor how WYSIWYG
features might be integrated into a TeX environment: I'm sure we all have
an intuitive {\it feel} for what makes TeX TeX, and for what makes it quite
different from Ventura Publisher or Aldus Pagemaker; within the constraints
of the original design philosophy of TeX, what improvements would YOU like
to see ?
Philip Taylor
Royal Holloway and Bedford New College.
---------------------------------------------------------------------------
Date: Mon, 05 Jun 89 18:13:22 EDT
From: INHB000 <INHB%MCGILLC.BITNET@UWAVM.ACS.WASHINGTON.EDU>
Subject: Comments on TeX
Keywords: general, TeX
I would like to respond to Philip Taylor's response to some of my
remarks. It is clear, first of all, that we are on totally different
wavelengths as far as what we want from a typesetting system. What I
mean by ``logical'' typesetting is a system in which I give a verbal,
rather then pictorial, description of what I want and the system turns
it into a document. This is the way Lamport has used this term
repeatedly, in TeXhax and elsewhere. If Mr. Taylor wants to use it
differently, he may go ahead, but he should make it clear that he is.
Now LaTeX for the most part is an extremely logical system. This means,
among other things, that you don't have to think about your layout and
you don't have to spend much time previewing. The picture mode is
something else again. You have to give a precise description of the
actual physical layout and you may well have to preview a moderately
complicated one 10-20 times, occasionally more.
If one took Mr. Taylor's definition of logical seriously, one would have
to call logical a typesetting system in which you had to specify
individually the postion of every character on the page or perhaps every
dot.
I suggested that in order to implement my publisher's requirement that
if facing pages were of different lengths (sorry about the typo), then
the left-hand one be longer, I would run out of memory. Mr. Taylor
disputes this. Well, the last time I ran the book through TeX, it
reported having used:
65449 words of memory out of 65535
Since there is no way of achieving the desired result without compling
two pages at a time, I think I may be pardoned for suspecting that the
compilation of two pages at a a time would have certainly exhausted
memory. If I am wrong, so be it.
The one time I tried using token variables, the attempt foundered on the
apparent inability to compile an empty string. I was attempting to
coerce TeX into accepting more than nine parameters by some such
business as
\def\cs<#1><#2>{\setparms<#1>...}
\def\setparms<#1,#2,...>{\tokena=#1 \tokenb=#2 ...}
and everything worked fine until I tested it with something like
\cs<,...><...>
that is the first parameter empty, and then everything blew up. Since
tokens are barely mentioned in the TeXbook, I certainly do not feel I
understand how to use them. It is said that they you can store a token
list and if they can store only a non-empty token list, the book should
have said so. In the example above, I finally tried the equivalent of
\def\setparms<#1,#2,...>{\def\tokena{#1} \def\tokenb{#2} ...}
and that worked fine in that case. However, on a different trip through
TeX's rather complicated alimentary canal it might not have.
I will stick to my guns on the statement that TeX's arithmetic is
terrrible. Suppose you want to kern 3 times \indent plus 5 points.
The code you need is
\newdimen \dimenvar \dimenvar=\indent \multiply\dimenvar by 3
\advance\dimenvar by 5pt \kern\dimenvar
Surely it would have been possible to have designed TeX so that it could
have accepted \kern{3*\indent+5pt}. And this is a simple example! In
the macros I have for doing commutative diagrams, there are much more
elaborate examples. The lack of orthogonality is annoying, but only
part of the problem.
The reason I find \futurelet unsatisfactory is that it ties the
lookahead to one of TeX's defining word. I would like a procedure that
would allow me to take a token from the input stream, assign it to a
token variable, or use it in a definition and then put it back. Before
I began using TeX, I had designed and implemented a math formatter. It
worked quite well, but wasn't anything like TeX in its ability to put
out print quality material. One of the most useful procedures was
ungetchar that put the most recently read character back into the input
stream. It was much easier both conceptually and in practice than
\futurelet (one of whose worst features is its name).
Virtually every user of plain TeX has told me that LaTeX takes away all
the flexibilty of the former. When challenged, however, none of them
has come up with a single example of something they actually do in plain
that can't be done, usually with less pain, but sometimes with more, in
LaTeX. I'm not saying examples don't exist (although with enough work,
it is evident that anything from TeX could be done in LaTeX), only that
is the very rare TeX user who will do such a thing. To take one trivial
example, several methematicians have told me they won't use LaTeX
because it doesn't allow them to put in their own equation numbers. Of
course it does! Try running the following sample text through LaTeX:
\documentstyle{article}
\begin{document}
\begin{equation}
e=mc^2
\end{equation}
$$e=mc^2\eqno \rm off-the-wall$$
$$e=mc^2\leqno\hbox{\rm(\theequation a$'$)}$$
$$e=mc^2\eqno\hbox{\refstepcounter{equation}\rm(\theequation $''$)}$$
\end{document}
In conclusion, may I remind Mr. Taylor and others that each of us has
his own needs and one man's features are another's deficiencies. If
someone says he misses a feature, he probably does. One feature that I
would dearly love to see is a procedure that places one box over another
opaquely, that is wiping out a portion of the other. Take my word for
it, I have sufficient reason to want it. The main reason I hesitated to
mention it before is that such a feature would render every device
driver obsolete.
Michael Barr
----------------------------------------------------------------------------
Date: Sat, 3 Jun 89 23:02:25 PDT
From: lgy@newton.phys.washington.edu (Laurence G. Yaffe)
Subject: EOF problems in TeX 2.95
Keywords: TeX 2.95, EOF problems
How can one fix the response of TeX to end-of-file conditions so that,
for example, "tex somefile < /dev/null" will not cause an infinite loop
if somefile.tex tries to include a missing file? Will this be fixed in 2.98?
------------------------------------------------------------------------------
Date: Sat, 3 Jun 89 22:59:06 PDT
From: lgy@newton.phys.washington.edu (Laurence G. Yaffe)
Subject: Running TeX in batchmode
Keywords: TeX, batchmode
Given a dumped format file (called "phyzzx.fmt") which includes the command
"\everyjob {\input myphyx.tex}", why is it that "virtex &phyzzx a.tex" creates
a log file called "a.log", while "virtex &phyzzx \batchmode \input a.tex"
creates a log file called "myphyx.log"? Is there any way to run tex in
batchmode on a given input file, using an arbitrary format, and ensure that the
log file receives the same name it would if tex were running interactively
(without, of course, changing the input file itself)?
-------------------------------------------------------------------------------
%%% Concerning subscriptions, address changes, unsubscribing:
%%% BITNET: send a one-line mail message to LISTSERV@xxx
%%% where xxx is the nearest geographical site in the
%%% tree shown below
%%% SUBSCRIBE TEX-L <your name> % to subscribe
%%% or UNSUBSCRIBE TEX-L
%%%
%%% Here is the BITNET re-distribution tree as shown in a recent
%%% REVIEW (The geography is guessed at from the subscription list)
%%%
%%% CLVM TAMVM1 FINHUTC
%%% | | (Finland, UK, Scand, CERN)
%%% | | |
%%% TeXhax ----> UWAVM ----- MARIST ----- EB0UB011 ----- BNANDP11
%%% | (France,Italy,Spain) (Belgium)
%%% | |
%%% UBVM HEARN --- DEARN
%%% (Netherlands) (Germany)
%%%
%%% Internet: send a similar one line mail message to
%%% TeXhax-request@cs.washington.edu
%%%
%%% All submissions to: TeXhax@cs.washington.edu
%%%
%%% Back issues available for FTPing as:
%%% machine: directory: filename:
%%% JUNE.CS.WASHINGTON.EDU TeXhax/TeXhaxyy.nn
%%% yy = last two digits of current year
%%% nn = issue number
%%%
%%%\bye
%%%
End of TeXhax Digest
**************************
-------