From: Juha Virtanen (jiivee@hut.fi)
Date: 05/21/93


From: jiivee@hut.fi (Juha Virtanen)
Subject: Re: Let's write a wordprocessor.
Date: 21 May 1993 17:12:39 GMT

Hi,

This sounds really a good idea to implement a word processor or
word processing environment--get existing applications like
LaTeX, TeX, ghostscript, xdvi together and write on frontend for
editing and lauching other tasks.

In article <mjr.737924559@ursa> mjr@ursa.calvin.edu (Matt Ranney) writes:
   Because LaTeX/TeX are popular and some people write that code
   directly. I'm assuming that in addition to feeding these files to
   TeX, these people are also going to want to modify them and do other
   crazy macro definitions and stuff like that. How can we read this in
   again? Should we have a complete TeX parser built in to this new
   program? No, we shouldn't. We should have a SAVE file format that is
   not nice to edit, but is a 7-bit ASCII file that can be read in again
   by this program. Then we should have several OUTPUT file formats that
   will give you LaTeX, PostScript, WP5.1, whatever you want.

From my wiev, it's not worth of the time of implementing a new
text format if there will be no conversion programs from and to
that file format. For example, if I get a LaTeX source for a doc
and print it and then find out that I need to change some things
(typos, errors) if there is no conversion from LaTeX to this new
format. Why not to use LaTeX/TeX sources and borrow code from
there to conversion programs?

In general, I agree also that this word processor needs it's own
file format with space to expand. It would be somehow simpler to
implement than a word processor using LaTeX format files.

From other side, if the file format exists for now (like LaTeX),
this word processor should be written so that it can be expanded
with new packages. This sounds like a macro expander (well,
macros are needed in any case, as this should look like a w4w).

Just some random thoughts.

Juha