Re: AWEB; Enhanced Document Encoding

["Nick Roberts" <Nick.Roberts@dial.pipex.com>]

Well, I think to quite a large extent Sven and I are arguing at cross
purposes, and to quite a large extent I agree with what he is saying anyway.
I also suspect, perhaps, he is not being deeply serious, and I am, after
all, only throwing an idea into the wind!

Forgive me for not requoting his post as appropriate, but I think the size
of our articles could start getting a bit too big (does that sound funny?
You know what I meant! :-).  Again, I hope I'm forgiven for cross posting.

Firstly: MS Word is clearly not the market leader through technical
superiority alone.  I don't think anyone's going to disagree with that!
It's neat in some ways, and it's certainly 'feature-rich', but it sure has
its problems.  I wouldn't want to use it for any large work (for fear of it
crashing or just being too slow); I might well use TeX for such a project,
instead.

The example I gave with the secretary was, of course, not a very serious
one, and I agree with Sven on every point he makes about it (apart from the
paperclip, which I find abominable ;-).  He is particularly right about
secretaries not knowing enough about maths to typeset formulae correctly (of
course, some secretaries will, but I think the truth is most will not).
(And as for Windows crashing: my Windows 95 PC crashed yesterday, losing me
a large e-mail I was just about to send.  It's a fairly regular occurrence.
'Nuff said.)

The typesetting problems that Unicode solves are only a few; this was a very
minor point.

Sven points out that, for a lot of programming tools (e.g. sed, awk, perl,
shells), outputting normal text will indeed be easier than numbered codes.
He is, of course, quite right.  I tend not to use such tools, so this was
simply a case of biased thinking on my part.  He also questions why
outputting codes would be 'much easier' than outputting, say, LaTeX
directives, and on reflection, I must admit, there would probably be no
great difference between the two (in most programming languages, and an
advantage to LaTeX in some languages).

I think Sven must misunderstand my point about the control characters.  Yes,
you need them (and even then, not all) for old-style character
communications, but you don't need them, except a few, for anything else.  A
text document file could certainly get away with using the control
characters (or most of them) for other purposes.  If you needed to
communicate such a file somewhere, you'd send it as a binary file.  No
problem!  (If you need to process it as plain text, you must put it through
an appropriate filter; also not really a problem.)

Finally, Sven confirms his hatred of all things WYSIWYG.  Might projects
such as LyX change his mind?  I used to be pretty contemptuous myself of
typical WYSIWYG word processors*, most of which were suitable only for
letters and trivial documents, once.  But I observe and listen to many other
computer users in my job (roving consultant), and most of them love WYSIWYG,
and the immediacy of typical GUIs.  As a programmer, I feel I must bow to
the needs of the majority.  Also, I must admit, I like it a lot -- if I am
trying to use a program I have never used before and will never use often --
when I can get the program to do what I want straight away by just pulling
down a menu or two.  (Of course, only the best 5% odd of GUI programs are
this well designed, in reality; but the principle stands.)  Anyway, what I
am proposing is not to replace TeX, but to augment it (and maybe, one day
long from now, supercede it), as a standard format for text documents.  I'm
not trying to take TeX away from anybody!

-------------------------------------
Nick Roberts
-------------------------------------

*apart from the wonderful WordStar [sheds a sad tear], way back when
graphics screens were unknown, when WordStar was the only serious
semi-WYSIWYG word processor for microcomputers, and the world worshipped at
its feet.  Whatever happened to WordStar?  Why wasn't there an enquiry?