Path: utzoo!utgpu!watserv1!watdragon!lion!ccplumb
From: ccplumb@lion.waterloo.edu (Colin Plumb)
Newsgroups: comp.std.internat
Subject: Re: ISO 639
Message-ID: <20700@watdragon.waterloo.edu>
Date: 8 Feb 90 21:37:26 GMT
References: <3109@paperboy.OSF.ORG> <1091@tuminfo1.lan.informatik.tu-muenchen.dbp.de> <15894@haddock.ima.isc.com> <1990Feb8.164921.22689@utzoo.uucp>
Sender: daemon@watdragon.waterloo.edu
Reply-To: ccplumb@lion.waterloo.edu (Colin Plumb)
Organization: U. of Waterloo, Ontario
Lines: 26

In article <1990Feb8.164921.22689@utzoo.uucp> henry@utzoo.uucp (Henry Spencer) writes:
> Implement?  You think ISO cares about implementation?  Couldn't prove it
> by me, not considering some of the things they put in those protocols.
> You don't understand:  ISO protocols will gain widespread acceptance by
> being rammed down our throats by politicians, not by being enthusiastically
> accepted by implementors.  That's how ISO works.

I have to agree.  Why the U.S. DoD ever decided that they were going to
take a well-working Arpanet and redo it using ISO/OSI protocols is
beyond me.  TCP/IP is relatively simple, understandable and works
well.  Yes, if you need to, reformat the packets, change the checksum
algorithm, or whatever, for your hardware-assisted 1 Gbit gateway, but
please find serious problems before fixing them.

(I've just been reading the SCSI "standard" and have wondered where
standards teams get such sick ideas.  I like the Futurebus standard,
although they introduce some wierd terminology for cache snooping, I
like the ANSI C standard.  They both know what they're trying to do
(and it's *not* create a standard so vague that everyone's
implementation is standard, as crept into IEEE 754 and the extended
precision disaster), and it's clear to anyone who understands the field
reasonably well.  I think a good sign of a bad standard is when they
need to invent a lot of terms not commonly in use.  Maybe they're
solving problems not commonly encountered?)
-- 
	-Colin