4.0551 TEI, SGML, and Real Word Processing (3/103)

Elaine Brennan & Allen Renear (EDITORS@BROWNVM.BITNET)
Tue, 2 Oct 90 21:26:29 EDT

Humanist Discussion Group, Vol. 4, No. 0551. Tuesday, 2 Oct 1990.


(1) Date: Tue, 2 Oct 90 16:11:34 PDT (21 lines)
From: cbf@faulhaber.Berkeley.Edu (Charles Faulhaber)
Subject: Re: 4.0545 TEI Workshop Trip Report

(2) Date: Tue, 2 Oct 90 9:52 GMT (20 lines)
From: Don Fowler <DPF@vax.oxford.ac.uk>
Subject: RE: 4.0547 The WP Saga's Markup Moral; Real Word Processing?

(3) Date: Tue, 02 Oct 90 10:57:59 +0100 (62 lines)
From: Dominik Wujastyk <ucgadkw@ucl.ac.uk>
Subject: Re: 4.0547 The WP Saga's Markup Moral; Real Word Processing?

(1) --------------------------------------------------------------------
Date: Tue, 2 Oct 90 16:11:34 PDT
From: cbf@faulhaber.Berkeley.Edu (Charles Faulhaber)
Subject: Re: 4.0545 TEI Workshop Trip Report (1/129)

I can't remember whether I added my impressions of the
TEI draft or not. Needless to say, I was very impressed
by the amount of work that went into them and the incredible
variety of issues that needed to be addressed.

The appendices with sample DTD's are very useful. I would also
like to suggest a second group, with a "starter set" of tags
for the most common document types, i.e., the minimal level of
encoding that the committee would recommend. My only problem
with the draft was that I found it somewhat difficult to put
the various pieces together. I think we need something that
we can hand to people to get projects started, then give
them the guidelines to consult when specific problems arise
not covered by the starter set.

Charles Faulhaber
UC Berkeley
(2) --------------------------------------------------------------29----
Date: Tue, 2 Oct 90 9:52 GMT
From: Don Fowler <DPF@vax.oxford.ac.uk>
Subject: RE: 4.0547 The WP Saga's Markup Moral; Real Word Processing? (2/45)

Lou Burnard is of course right about procedural markup, but the problem
is that for most people it is an awful bind to set up their system so
that explicit SGML type tags can be used (you have to write the macros,
however rudimentary). What is necessary is to get the major commercial
WP programs to provide an SGML import/export facility, so that one could
use the easy facilities within the program for footnotes, font changes
(esp. between languages - the problem with moving Greek or Hebrew
between WPs) etc but then choose to have the text output with tags. Of
course there would have to be sensible (and changeable) equivalencies
between typographical and logical markup, which could never be perfect
(is that underlining emphasis, a title or what?), but it would help a
lot and would me more use to most people than large expensive SGML
editors (which I gather exist). How about putting pressure on Microsoft
and WP Corp to provide something like this?

Don Fowler
(3) --------------------------------------------------------------74----
Date: Tue, 02 Oct 90 10:57:59 +0100
From: Dominik Wujastyk <ucgadkw@ucl.ac.uk>
Subject: Re: 4.0547 The WP Saga's Markup Moral; Real Word Processing? (2/45)

Lou says,

> Personally, the thing I find most annoying
> about WP footnotes is the arbitrary distinction it makes between foot-
> and end- notes. And the fact that you can only have one sequence of
> foot- (and one of end-) notes. And ...

Many HUMANISTS use Nota Bene (aka XyWrite, which is also embedded in many
turnkey systems, especially for professional typesetting, like ATEX),
and this program allows three separate layers of footnotes, tagged as
follows: <<FN1This is the text of a first-level note>>, <<FN2While this
will be a second.>>, <<FN3And this is a thirt-level note.>>. This
looks pretty SGML-like to me. ("<<" and ">>" are char codes 174 and
175.) And you can command the notes to be put at the bottom of a page,
not to be printed at all, or to be dumped out at the end of the main
text. You can control each level of notes separately in this respect.
That's one of the reasons I like XyWrite.

And James O'Donnell asks

> Subject: Why do we call them word processors?

> Couldn't we imagine
> a `word processing program' that would function chiefly within a
> computer environment, manipulating and presenting words in ways that
> take advantage of the possibilities of the computer environment and only
> secondarily worrying about how they will look on paper? Are there such
> programs, only less well-known and less commercial? (Probably: any
> recommendations?)

Such programs exist in profusion, as one might have guessed, and are
called "editors" as opposed to "word processors". In essence, a word
processor is an editor plus a formatter. The editor allows you to
input characters. Some of these characters are your text, others are
codes that will later control the formatter. When you press "print",
the codes tell the formatter what to do with the text. Some clever
word processors can do this in real time, printing the output on the
screen. That is WYSIWYG. Others go part way in real time, but offer
full WYSIWIG only as a preview facility: look but don't touch.

The great grand daddy of all editors is EMACS by R. M. Stallman (MIT
and FSF). EMACS wasn't the first, and it stood on the shoulders of
TECO, but I think it could count as the one that taught the world what
an editor should be like. EMACS is still one of the best editors
around, and is installed on about half of all mainframe computers and
minis in the world (pity the other half). FREEMACS by Russ Nelson is a
version of EMACS for DOS.

The great grand daddy of all formatters is harder to decide, but it is
probably Osanna's roff/troff/nroff. Reid's Scribe has also been hugely
influential. But most word processing programs try to hide the
formatting codes to a greater or lesser extent, so one is less
explicitly aware of this side of the program.

Garn! I managed to say all that without mentioning TeX -- oops!

Dominik