Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!husc6!bbn!uwmcsd1!marque!gryphon!richard
From: richard@gryphon.CTS.COM (Richard Sexton)
Newsgroups: comp.sys.amiga.tech
Subject: Re: A Modest Proposal (IFF QuickDraw)
Summary: Graphics standards
Message-ID: <4033@gryphon.CTS.COM>
Date: 16 May 88 06:15:33 GMT
Reply-To: richard@gryphon.CTS.COM (Richard Sexton)
Organization: Trailing Edge Technology, Redondo Beach, CA
Lines: 88

In article <5251@xanth.cs.odu.edu> kent@xanth.UUCP (Kent Paul Dolan) writes:
>
>I suggest, rather than using a non-standard such as Postscript (what

Non standard ? Just get back from Mars, Kent ? Did you bring me anything ?

>Good does it do you going to a flatbed plotter, for example), you
>choose the actual international structured graphics standard from ISO
>and ANSI, the Computer Graphics Metafile, and let a CGM to Postscript
>cconverter (there are sure already to be several available) take care
>of those microcomputer add-ons that grok Postscript.  I'm not terribly
>familiar with Postscript, but I don't have the impression that it is
>designed to be invertible back to a useful form for internal display
>list processing, as is CGM.

I've looked at CGI/CGM and PostScript. The sheer size of the CGI
spec almost virtually guarentees nobody will read it.

It's SOOOO comprehensive, it almost requires being split up into
a dozen volumes.

Graphics 'standards' are fun. You have your formal standards, like
SIGGRAPH core, GKS, CGI/CGM, which cover everything, but almost
nobody uses.

Look at CORE. It was agreed upon by all of SIGGRAPH, and you would
think that would guarentee all the industry graphics weenines would 
use it.

How many graphics applications were written for CORE ?

Now look at how many graphics applications were written for IBM PC EGA.
Not much of a standard, but people have it and people use it.
It's a de facto standard. Just like PostScript. EGA is dying, PS is
doing quite well, and doing better all the time. 

>"Standardization" on a
>non-standard like Postscript is just a way for the Amiga community to
>paint itself into a corner and isolate itself from the mainstream
>workstation market toward which the hardware line seems to be aiming
>itself.

PostScript is virtually the only reason the Mac has done so well.
Lots of peripherals speak PS, from lowly 300 dpi lasers to 2500+
dpi phototypesetters and now there are even color thermal and
color laser PS printers. Given that an IFF/PS file could be generated
on an amiga and transported to any of these devices, I just cant
see how we would be "painting ourselves in a corner". Maybe you
could explain this to me, Kent ?

>One of the players with whom we must compete eventually, Sun, has
>itself poised squarely in the mainstream, with product efforts in GKS,
>CGM, CGI, PHIGS, and perhaps PHIGS+; I'm not aware of any IGES work

Thats great, Kent, but how many packages under $5000 use any of those
standards ? Bugger all. How many devices speak those languages ? Bugger
all.

>mention my too often mentioned IRA! ;-) Right now this looks like a
>trireme that finally hired on a drummer, so all the oars are stroking
>in rhythm and the vessel is moving ahead fairly quickly, but has not
>yet gotten itself either steersman or lookout.

You sound like you spent too much time in a submarine, Kent :-)

My main objection of CGI/CGM (IMHO the most managable of the standards
Kent mentioned) is that its a "standard by comittee", and has a function
for everything - lessee, whats the opcode dor set a transparent dashed
border for convex pink polygones on surface 1, priority 4, warp 6, captain
to enterprise, 2 to beam up Mr. Scott.

In covering ALL the bases, they built something so goddamn bloated you
need a Sun 4 to run it. Combine that with the fact that the standard
isn't finished yet, and there are ZERO devices that speak it, and
ZERO screen display systems that use it directly it seems like a poor
choice.

What is attractive about the PS imaging model is that it embodies a rich
set of primatives upon which anything can be built. 

Several people seem to think I'm advocating NeWS and/or incorperating
postscript commands into IFF files. No. Just developing an IFF format
that has a direct 1 to 1 correspondence to postscipt primitives, such
as line type, clipping paths, translation, scaling, rotation and the such.

-- 
               And other people run red lights, too.
richard@gryphon.CTS.COM                          rutgers!marque!gryphon!richard