Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!asuvax!ncar!elroy.jpl.nasa.gov!sdd.hp.com!zaphod.mps.ohio-state.edu!caen!uflorida!kluge!scs!acmfiu
From: acmfiu@scs.fiu.edu (ACMFIU)
Newsgroups: comp.sys.apple2
Subject: Re: TeX
Message-ID: <2217@kluge.fiu.edu>
Date: 4 Feb 91 23:10:37 GMT
References: <439@generic.UUCP>
Sender: news@kluge.fiu.edu
Organization: Florida International University, Miami
Lines: 49

In article <439@generic.UUCP> ericmcg@pnet91.cts.com (Eric Mcgillicuddy) writes:
>64k arrays are not a problem, if are willing to go to the trouble of breaking
>up your data into segments. 64k is a HUGE amount of memory, what could you
>possibly store that requires more than this present at any one time? 2Megs
>would be nice, but design it to work on 1Meg if you want anybody to use it.
>
>BTW, Some Major obstacles have been crossed with ViM and I hope to have it
>ready soon. Although I call it Virtual Memory, it is more of an automatic
>overlay system. This help you out a bit, but you MUST break up your data (and
>code to some extent) t otake advantage of it. I had posted a preliminary list
>of restrictions a few months back, I can repost if interested.

Have you ever taken a look at the source for TeX. There are several
"types" of TeX. By this I mean that there are certain people that distribute
TeX with small/big memory constraints. The one I am using at present has
TeX using 1meg+ of data. Yes, just data, no code. As a matter of fact, I 
don't even think the program can fit on an 800K disk (i can't remember as
it's been awhile since i worked on it on the GS, but continue to work with
it on the Sun). Anyway, TeX keeps everything in arrays. Due to the fact
that TeX is to be extremely portable, Don Knuth decided not to make any
assumptions on the system TeX would be ported to. That is one reason why
TeX produces the same output on every machine it exists on (else it is
not formally called "TeX").

Once the conversion is complete, I plan on making some major modifications
to the way TeX code looks. Primarily, I'd like to get rid of the TeX.pool
file and make all memory dynamically allocated. I haven't taken a look at
this extensively yet because I first want to get this thing done.

As to the speed consideration some people have been talking about, well I
don't know yet how fast it will run. I don't have any accelerator on my
GS so if I use it then you can probably rest assured it runs adequately
fast. I've had certain TeX documents on the IBM version I work on that
just screech TeX to a half for seconds/minutes. I'll be sure to test times
on both the IBM and the GS once things are done.

Oh yeah, one thing i forgot to mention above. I am not using the default
memory constraints defined in the original tex.web program. This is far
too small for those of you wishing to use LaTeX, BibTeX, AMSTeX, etc.
I forget now the difference but I'll be sure to have a version of TeX with
just these memory constraints (the original TeX, that is), and others
with big, bigger memory constraints. Of course this will be soon after
TeX is complete because removing the tex.pool file and making all memory
dynamically allocatable is not an overnight project.

I can't vouch for the speed of the previewer. All that stuff will come later,
if not by me then by other interested parties who want to extend TeX.

albert