Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!mit-eddie!genrad!decvax!savax!elrond!anson
From: anson@elrond.UUCP
Newsgroups: comp.graphics
Subject: Re: PHIGS standard available for comments
Message-ID: <1192@elrond.CalComp.COM>
Date: Mon, 31-Aug-87 09:27:51 EDT
Article-I.D.: elrond.1192
Posted: Mon Aug 31 09:27:51 1987
Date-Received: Tue, 1-Sep-87 04:27:52 EDT
References: <3765@well.UUCP> <2162@xanth.UUCP> <6813501ADC@MSU> <1176@elrond.CalComp.COM> <8510@utzoo.UUCP>
Reply-To: anson@elrond.UUCP (Ed Anson)
Organization: Calcomp Display Products Division, Hudson, NH, USA
Lines: 37

In article <8510@utzoo.UUCP> henry@utzoo.UUCP (Henry Spencer) writes:
>
>  One major reason for refusing
>to provide machine-readable forms of standards is to make *sure* that the
>reader is seeing the complete and unaltered original, not somebody's half-
>assed idea of an "improved" version.

Of course, this is a major reason for their policy.  Unfortunately, it has
the (unintended) effect of reducing the quality of documentation that can be
supplied with a PHIGS implementation.  As an implementor desiring to abide
by the copyright laws, I will probably be forced to issue three separate
documents to application programmers:
1. The PHIGS Functional Spec (purchased from ANSI, of course)
2. The language binding spec (separate document, also purchased from ANSI)
3. A third document, detailing the specifics of the particular implementation,
   specifically stating our resolution of those areas that ANSI explicitly
   made implementation dependent, and admitting to any implementation
   limitations.

The net result is that an application programmer will have to look at three
separate documents to learn how to use a single function.  I would prefer not
to do this, but the alternative is to produce a "half-assed" attempt at
paraphrasing the ANSI specs without violating their copyright.

What I would rather do is start with the original ANSI spec, and fold into
it the specifics of my implementation, as well as the binding information.
There is no reason to alter the text of any of ANSI's specs.  Indeed, I
strongly prefer to reproduce them exactly.  I just want to get related info
together on the same page.  Typographical distinctions between the various
items on a page should be enough to avoid reader confusion.

But, alas, such a solution doesn't appear to be feasible at the moment.  If
anyone has any other ideas, I would be eager to know them.
-- 
=====================================================================
   Ed Anson,    Calcomp Display Products Division,    Hudson NH 03051
   (603) 885-8712,      anson@elrond.CalComp.COM