Re: WG process etc. [Re: WG last call on tunneling scenarios]

[JORDI PALET MARTINEZ <jordi.palet@consulintel.es>]

Hi Jim,

See below, in-line.

Regards,
Jordi


> De: "Bound, Jim" <jim.bound@hp.com>
> Responder a: owner-v6ops@ops.ietf.org
> Fecha: Mon, 1 Nov 2004 15:28:18 -0500
> Para: <jordi.palet@consulintel.es>, <v6ops@ops.ietf.org>
> Asunto: RE: WG process etc. [Re: WG last call on tunneling scenarios]
> 
> Jordi,
> 
>> On the other way around, I perfectly understand that industry is
> leading and some non-standards are
>> becoming de facto standards, which is not good.
>> That's why we need to work faster and in parallel instead of in serial
> and slow mode. If the WG is not
>> commenting or providing inputs, but there are no objections either, the
> work should be standardized.
> 
> This is your confusion and the IETF changed I think about 18 months ago
> or when we killed A6.  Specs cannot go forward from silence and that is
> now true in all IETF WGs I know of in the IETF.  I first ran into this
> in DHCPv6 working with Ralph as Chair many years ago.  I did not like it
> at first and did not get it.  Then we were able to get the engineers to
> comment on DHCPv6 and made the spec 10 times as strong and solid
> consensus.  So now I am a firm believer in silence is no good as metric

This happened in this WG already, and is going to happen again with the last
calls send a couple of days ago.

In an ideal world, I will agree, that no WG comments is not good, but world
is not ideal, and even if a document is bad (or good), may be there are no
inputs. This is happening.

> to move a spec forward.  Zero conf work had lots of mail thread
> discussions and appears to be valid to accept as work item within the
> IETF.  Bottom line is the Chairs have not broken any rule but enforcing

Other documents also got lots of mail threads, and they aren't WG items.

And anyway, is difficult to measure if a document is or not a valid WG item
depending on the number of emails or threads, because it can be a simple
analysis which is clear, and not necessarily have comments. In our case, the
only comments that we got were from Brian mainly, regarding the missing SLP
analysis, which has been already done.


> the rule.  Also sometimes the WG is just maxed.  For example we did not
> get input as fast as we needed it for Enterprise Analysis but then we go
> so much I am still parsing it as Ent Analysis editor.  I don't think
> there is any scientific method to this at all the longer I am around.

My metric to see if a document is a valid WG item is the WG charter.
Otherwise, why we need the charter at all ?
And if the charter is not clear, then we should change it, but not start
discussing if this or that document is or not within the charter. You don't
think so ?

> 
> There is also no secret discussions that is just absurd.  Is it possible
> the ADs and Chairs individually don't support specific work, sure, but
> that's another matter and their right, and fair too.  The objective is
> to get the WG excited technically about specific work, and that makes
> the Chairs and ADs get a buzz.

Yes, you say it "individually" but that's not with their respective hats ...

Again the point is the charter is there for something.

> 
> Reqarding industry doing defacto standards.  Yes this is happening now
> with IPv6 Transition and several mechanisms are being deployed now that
> are way ahead of the IETF.  That will correct itself between the market
> and the IETF.  At times the market leads, but usually the IETF is in
> synch with the curve, but not on time. But, v6ops is doing everything it
> can to meet time-to-market and I for one applaud all of us here for that
> we are getting real work done and on time.  As you know I am very pro
> defacto standards and solutions when the IETF don't get it and a large
> number of implementers do. We just move forward in industry and keep
> sending data to this body called the IETF.  The IPv6 Forum is exactly
> from the IETF moving to slow and in 1999 implementors took matters into
> their own hands and now the IPv6 Forum is a world wide deployment body
> that clearly can support defacto standards and with task forces that are
> part of the IPv6 Forum across the planet.  That is what happens when any
> standards body is to slow and does not meet the needs of the market.
> 
> I read every mail on this list and a few others and you have not been
> treated unfairly at all, but your work has not reached consenus on this
> list that I can see as working group items.  That does not mean it is

Some of our works had been requested by the chairs, for example
tun-auto-disc, and we decided to take over it, because actually we already
were working on that direction.

Is not fair then that this should be WG item (specially because is an
analysis of an operational issue, which is part of the charter) ? Otherwise
is not fair that chairs ask for it ;-)

> not good work but maybe not work in the IETF, as a question?  Ask your
> self is it a protocol, operational tool that can be standard without
> forcing implementation of protocols through configuration, a best
> current practice, etc.?  And most important "what problem does your work
> solve"?  

All the answers to those questions bring me to the same conclusion ... It
should be a WG item, unless there is objection in the WG.

> 
> What I have found with my work in the IETF when it stalls it is usually
> there was no consensus on the problem it solves or there needs to first
> be discussion of everyones assumptions.  For example, I believe many
> customers will simply shut off IPv4 on a dual IPv4/IPv6 subnetwork and
> cascade that policy expediently as a transition strategy across all
> their other subnetworks until the entire customers Intranet or Internet
> is IPv6 dominant with only pockets of legacy IPv4 for transition.

Fully agree, you know, specially with only IPv6 ;-)

> Educating all why and how is what I am doing now and once they see that
> then solving the problem can move forward.  I also think most customers
> now will go get IPv6 prefixes and 6to4 is highly questionable as widely
> used for the transition and that is a new change in the market some of
> us have learned directly.  These are just examples of others who have
> the same problem and I don't think it is because of secret meetings in
> the IETF.
> 
> I don't think the chairs or ADs warrant your mail and its unfair as one
> working group members input these chairs work their ass off and do what
> they can to keep things moving.  Now if they don't listen or ignore
> consensus I will be the first to throw tomatoes, but I don't see that in
> this specific case.
> 
> P.S. Pekka - I still do not agree with you about 70% of the time :--)
> 
> Regards,
> /jim
> 
> 
> 
> 
> 



**********************************
Madrid 2003 Global IPv6 Summit
Presentations and videos on line at:
http://www.ipv6-es.com

This electronic message contains information which may be privileged or confidential. The information is intended to be for the use of the individual(s) named above. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information, including attached files, is prohibited.