W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > September to December 1995

Re: The IETF process

From: Balint Nagy Endre <bne@bne.ind.eunet.hu>
Date: Tue, 26 Sep 1995 05:29:24 +0100 (MET)
To: Shel Kaphan <sjk@amazon.com>
Cc: http WG <http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com>
Message-Id: <82.bne@bne.ind.eunet.hu>

My father said [while talking about design of electronic devices]:
Every particular device can be designed in seventy-seven different ways and
seven out of seventy-seven solutions are equivalent. Which solution is choosen
out of those seven makes the difference between electronic engineers.
I add the following phrases to my fathers:
[axiom]
Only Good is enough capable to list all 77 possible solutions.
Only Good is enough capable to decide, which 7 of them are the best.

How my fathers words apply to our work [protocol design in context of
the internet standardisation process]?

[applicability statement]
Being humans, we can only aim the goal of finding one solution out of
possible 77 (and hope that the choosen solution is enough near to one of
the 7 "optimal" ones.).

In other words (or being more practical)
We can hope, that the solution choosen by WG can be turned into a working
solution.

We have an other option [among open standardisation process]:
Let independent developer teams to implement as much as possible working
solutions, and let the open market to choose the best.

HTTP/1.0 demonstrates, how this option works.
(I think, I should not give more arguments explaining the benefits of
open standardisation process vs. concurrent implementation of similar 
but different protocols.)

Understanding this means:
Including un-proven ideas into standard zeroes our hopes in achiving a
working solution.
In other words:
Our believe in a proposal being possible to work isn't enough to promote
that proposal as a standard.

Improving the IETF process itself is a job of the poised95 WG.
(poised@tis.com, subscribtion: poised-request@tis.com)

The purpose of the HTTP WG is to design the HTTP protocol to be
standardised in the framework of the IETF process.

According to this, we can start implementing promising ideas to see,
are those ideas working or not.

Andrew. (Endre Balint Nagy) <bne@bne.ind.eunet.hu>
Received on Monday, 25 September 1995 21:37:53 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:32 EDT