[comp.sys.amiga.introduction] TeX

xanthian@zorch.SF-Bay.ORG (Kent Paul Dolan) (01/15/91)

In comp.sys.amiga.applications,

ben@contact.uucp (Ben Eng) writes:
-----
| Unlike a WYSIWYG system, such as WP5 or a page layout program, TeX
| handles all the kerning, paragraphing, spacing, margins, page breaks,
| indenting, centering, and other subtle details automatically. Title
| pages, Abstracts, Tables of Contents, List of Figures, List of Tables,
| Chapters, Sections, Subsections, Bibliographies, References, footnotes,
| figures, tables, references to numbering of tables/figures/etc, ALL
| numbering (page, chapter, section, etc.), and virtually everything else
| is all done for you.
-----

wally@pallas.UUCP (Wally Hartshorn) writes:
-----
| I know very little about TeX.  You say it does all of this stuff
| automatically?  How could it do that?  I'm confused.  Suppose I say
| that I want a table of contents.  I still have to go through and tell
| it what to include in the table of contents, right?  How is that any
| different from the way WordPerfect does things?
-----

Well, a TeX document is actually a computer program that you write,
using the TeX language, with embedded data (your text), a lot like a
BASIC program, which TeX interprets into a page layout with font and
style selections and locations and sizes for each character on the page.

A post processing step then maps the actual character image bits onto
the correct locations on the final document, which gives the opportunity
for various fonts and output devices to be supported by various post
processers, all running from the same TeX output. This allows the TeX
output to be sent to the screen for preview, to a dot matrix printer for
draft or final output, or to a laser printer or even a phototypesetter
for higher publication quality output.

Like a BASIC program that hides the details of doing floating point
calculations on a processor like the 6502 that has no inherent floating
point capability, the TeX interpreter hides the details of getting all
the columns lined up and the hanging indents done right and the index
items bookkept and the lines fudged closer or farther apart to avoid
unattractive "widow" lines alone at the top of a page, and the space
reserved for footnotes and the proper footnoting indices matched up, and
right side and left side page style differences kept straight, and
chapter headers set at the top of each page and book titles on the
facing page and on and on.

However, also like a BASIC program, TeX involves a learning curve, and
learning a new language.  Like most computer languages, it is possible
to do frighteningly complex things in TeX, and also to write buggy
programs, which end up creating unexpectedly ugly documents.

To help with this problem, TeX has several sets of designed-in extra
smarts available.  Low level TeX is an extremely primitive and detailed
language.  Built on top of this, so no-one but the original authors
have to deal with TeX at its worst, are .fmt files that provide macros
or subroutines built up from the lower level routines, that civilize
TeX a lot.  The most common is called plain.fmt, and TeX provides a way
to "compile" this into the TeX routine, sort of like chaining a setup
routine ahead of your BASIC program, so that the TeX you run looks like
plainTeX insead of like raw original TeX.

Above these routines are .sty (style) routines, which provide even more
smarts, but trade off against flexibility a bit, rather like programming
in C++ instead of C, or in a fourth generation language instead of in
COBOL. The most important of these are the set bundled by Leslie Lamport
as LaTeX, which provides styles that become fill in the blank documents
for business letters, resumes, various technical journal standard
formats, and so on, and Oren Patashnik's BibTeX, which provides
facilities for laying out bibliographies to satisfy various different
journal and textbook series publishing standards.

At the top, it is possible for TeX gurus to create "house styles" that
cover the complete layout of a professional journal or a novel or
textbook, so that the same consistent style can be created fairly
painlessly by different authors without a lot of editorial feedback or
repeated galleys.

The big win using TeX in any of its flavors is that you are writing
_about_ the structure of your document, as with nroff or troff, but with
much more sophistication, and letting TeX create that structure, rather
than directly creating the structure yourself, so that when you go back
and change something early in the document, the changes are rippled
through by TeX rather than by you, and you are relieved of a lot of the
layout detail work.

The big downside is that the document can easily double in size if you
like to use word processors as fritterware, changing fonts and sizes and
styles at the drop of a hat, and the original text, as in raw Postscript
programs, can be nearly impossible to read among the embedded commands.
This takes acclimation and self discipline to overcome.

-----
| There are some nice WYSIWYG word processors for the Amiga, but none of
| them seem to have all of the features of Word Perfect.
-----

That's OK; among those who have achieved a higher state of Amiga to Zen
enlightenment, ;-) those word processors are referred to as What You See
Is _All_ You Get. Learn TeX, and you'll never look back. It has become
the standard of scientific publication.

-----
| Do you know what the list price of AmigaTeX is?
-----

No, but in the range $200-$400, probably toward the low end of that
range.

As this has become a .introduction document, I'll cross post it and point
followups there.

Kent, the man from xanth.
<xanthian@Zorch.SF-Bay.ORG> <xanthian@well.sf.ca.us>