Re: report: URN Architecture Meeting at University of Tennessee, Oct 30-31

Keith Moore (moore@cs.utk.edu)
Sat, 02 Dec 1995 11:13:30 -0500


Message-Id: <199512021613.LAA25800@wilma.cs.utk.edu>
From: Keith Moore <moore@cs.utk.edu>
To: "Roy T. Fielding" <fielding@avron.ics.uci.edu>
Cc: Michael Mealling <Michael.Mealling@oit.gatech.edu>, moore@cs.utk.edu,
Subject: Re: report: URN Architecture Meeting at University of Tennessee, Oct 30-31 
In-Reply-To: Your message of "Sat, 02 Dec 1995 00:39:08 PST."
             <9512020039.aa22589@paris.ics.uci.edu> 
Date: Sat, 02 Dec 1995 11:13:30 -0500

> No, the IETF was intended to create interoperable standards *based on
> practice*.  It does not create standards out of thin air.  

While I've heard the former statement enough times to make me think
it's probably true, the latter does not follow from the former.
The IETF does sometimes create standards out of thin air, if a standard
is needed and there is not already "current practice" or even an
already implemented protocol that appears to solve the problem.

> It does not
> ignore current practice just because someone thinks they have a better
> idea.  

No, but if there's consensus that there's a better idea, current practice
doesn't necessarily prevail.  Note that it's pretty hard to get consensus
for something that appears to do considerable harm to the installed base,
but there's a difference between not doing harm and following current 
practice.

> If it is a better idea, then practice will adopt it first, not
> the IETF.  

Doesn't follow.  Practice follows the path of least resistance.
IETF has at least some chance to introduce intelliegent thought
into the process.

> That depends on what the first "scheme:" is.  If the word chosen is so
> general that it is guaranteed to cause confusion, such as "scheme", "urn",
> "url", "name", "locator", "urc", etc., then it definitely does violate
> the Tao of URI.

How about Incompatible Nonuniform Resource Identifier (or a suitable
abbreviation thereof?)

Keith