Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!mcsun!ukc!warwick!nott-cs!bhamcs!mjh
From: mjh@cs.bham.ac.uk (Mark Hewitt)
Newsgroups: comp.sys.xerox
Subject: Re: Interpress
Message-ID: <1991Jun27.125039.25031@cs.bham.ac.uk>
Date: 27 Jun 91 12:50:39 GMT
References: <1991Jun20.025104.22967@minyos.xx.rmit.oz.au> <9106241553.AA12380@Basilisk.ansa.co.uk>
Sender: news@cs.bham.ac.uk
Distribution: inet
Organization: Birmingham University Computer Science
Lines: 22

If you are going to develop an Interpress -> PostScript converter, be
aware of the following:

	a) Some of the maths is rather tricky
	b) Allow the order of a man year to do a good job
	c) It will be very slow
	d) It will be difficult to support some INTERPRES operators (but
	   I can't remember which, it's a couple of years since we did
	   this, CORRECT requires two passes over the evaluation stack
	   as I recall, which makes it very slow if it is used to
	   perform justification.)

The present Interpress -> postscript enging (written byy us where us =
Kernel Technology NOT Birmingham University where I post this from) will
only support phototypesetters, and will not work on Apple LaserWriters
or other PostScript printers.  This is an explicit trap because Xerox
clearly don't want to loose sales of their printers to the competition.

(Excuse the bouncy keyboard - I'll get a new one some day)

I speak for me alone.
					Mark J. Hewitt