The IETF's problems

[todd glassey]

Melinda - All -
I have a question, and that is "do you consider the IETF's not formally
allowing more than one standard track item per/WG per/initiative-scope" a
problem?, because I do, and I consider it a serious one at that. What I am
referring to is the fundamental issue of whether WG's should be allowed to
limit the number of initiatives on any one track to one, and my take is that
the answer here is NO.

My reasons for asking this are that RFC2026 and 2223 as well as 2418 all
talk about the scope and process of the standards model, but none of them
really have any process for one effort to supersede another unless that
effort is done by the people "owning" that spot originally. There are no
"hostile takeover guidelines" or processes, and that means that any
evolution of challenge of a commercially used protocol has to have the
blessings of the WG Chair and those that already own that standing in a WG
(This of course pertains to existing protocols predominantly).

So look - here is an example, Say I have a protocol for "Whiz-Banging" and
there is already a protocol on track inside a WG that Whiz-Bangs but in a
different manner. So the question is whether the IETF can support two or
more Whiz-Banging implementations, or only one, because if the answer is
"only one" then we need to also have a method for how would anyone introduce
the second Whiz-Banging protocol to the IETF to compete with the other?

The problem is that the answer is today that functionally that this will
never happen. And it is because it is very very unlikely that the folks with
their Whiz-Banging Protocol are going to want to roll-over and die...

So then what happens??? The challenging Whiz-Banger is submitted to the I-D
staff as a disclosure. The I-D staff go to the WG Chair who makes a judgment
that they will not allow that protocol to harm the one that has already
gotten investment from them and the others in their WG, so now the WG Chair
takes an active role in suppressing the challenging effort. But say the I-D
issues get past the WG Chair, so then perhaps with the current model the I-D
staff also decide that they don't want to publish this draft because it
competes with an already in-process or existing standard or standards-track
participant. So this is yet another hurdle to publication and an issue. So
now they as well are part of a restraint of trade development process as
well.

So the question simply is where does this end?  Just how does the IETF allow
for a competitive effort, since anything else seems to have serious legal
issues with restraint of market development (the building of standards is a
key part of this). Or is this just not of importance to this IETF?

Todd Glassey
----- Original Message ----- 
From: "Iljitsch van Beijnum" <iljitsch at muada.com>
To: "Melinda Shore" <mshore at cisco.com>
Cc: <problem-statement at alvestrand.no>
Sent: Tuesday, July 22, 2003 6:35 AM
Subject: Re: The IETF's problems


> On zondag, jul 20, 2003, at 19:28 Europe/Amsterdam, Melinda Shore wrote:
>
> > The discussion itself isn't doing much to distill the topic
> > down to document text, and we need to stay focused.  Do you
> > feel that existing text regarding scope in the problem
> > statement document is inadequate?
>
> Not sure what you mean by scope.
>
> However, many of the "root" problems in the problems draft aren't that
> fundamental. The three most fundamental problems I see are:
>
> - the IETF doesn't know what it wants to be: a "real" standards
> organization or some kind of a grass roots movement
> - inability to make decisions other than just wait until there is
> agreement or come up with convoluted processes that only work because
> people are forced to interact until they somehow solve the problem
> - inability to manage resources effectively and efficiently
>
> Iljitsch
>