Path: utzoo!attcan!uunet!tut.cis.ohio-state.edu!cs.utexas.edu!uwm.edu!rpi!batcomputer!cornell!beck
From: beck@hermod.cs.cornell.edu (Micah Beck)
Newsgroups: comp.text.tex
Subject: Re: Graphics in TeX
Message-ID: <37578@cornell.UUCP>
Date: 21 Feb 90 14:22:50 GMT
References: <1990Feb20.235243.21897@Neon.Stanford.EDU>
Sender: nobody@cornell.UUCP
Reply-To: beck@cs.cornell.edu (Micah Beck)
Organization: Cornell Univ. CS Dept, Ithaca NY
Lines: 18

In article <1990Feb20.235243.21897@Neon.Stanford.EDU> rokicki@Neon.Stanford.EDU (Tomas G. Rokicki) writes:
>It is becoming increasingly obvious that the future of graphics with TeX
>(and virtually anything else for that matter) is in PostScript graphics.

Well, you're half right.  PostScript is half of a solution - the back end.
However, a document described in PostScript cannot necessarily be manipulated
in any way except to be printed.  For instance, it is not in general possible
to edit a PostScript file in a structured way (ala MacDraw, Fig, etc..)

The worst case: text which has been written in TeX, translated to PS, and
included as part of a figure cannot be edited in LaTeX, or even roughly
translated back to TeX.  Same for structured graphics objects.

What's needed is a standard application-level representation for graphics,
and a good standard translation from this to PostScript.  Representing
all graphics in PostScript is like writing all programs in assembly language!

/micah