Re: [rrg] Recommendation and what happens next

[Robin Whittle <[e-mail]>]

Short version:      TL:   Again, we've been trying to have that
                          debate for three years.

                    RW:   I believe you have not contributed much to
                          the debate.

                    TL:   The feeling is mutual.

                 Tony gives no explanation for his dismissal of my
                 efforts - including 5 of the 15 critiques in the
                 RRG Report.

                 I give reasons for my negative assessment of his
                 contribution to debates - as briefly as possible.

Hi Tony,

You wrote:

>> Maybe your vision of an enlightened debate doesn't accord with that
>> of others.  I have been trying to do the same thing and it seems you
>> and I rarely agree about how to discuss things or about how to solve
>> the routing scaling problem.  Other people appreciate my efforts to
>> understand and discuss their proposals - so maybe my idea of how to
>> debate things is shared more by other people than yours is.
> 
> That may well be.  Doesn't make it right, however.

My impression is that an IRTF research groups is a less formal and
more free-wheeling arrangement than an IETF WG.  In msg06204 I quoted
some pertinent parts of RFC 2014.

Responsibility is generally given to the Research Group.  But the
Chair has wide discretion about how everything is done and is
ultimately responsible for the Group making progress.

By your own account, the Group hasn't made much progress.

We have achieved consensus on almost nothing.  We have no agreed
statement of the the problem or of the goals for a solution. We have
no agreed taxonomy of potential solutions.  We can't agree on whether
IPv4 is worth working on.  We can't agree whether all hosts should be
required to move to IPv6 and do more work, suffering more delays when
establishing communications or changing the (Locator) IP address (CEE
= Loc/ID Separation) rather than add some things to the routing
system (CES) and leave IPv4 and IPv6 host stacks and applications
untouched, and with no further burdens and few, if any extra delays
in establishing communications.

If you are right about how the Group should be run, are right about
Locator / Identity Separation being the way forward, and are right
about IPv4's fate:

   I want nothing.  IPv4 is done.  Over.  Cooked. Fully toast.  It
   will either enter a black market where we deaggregate and no
   proposal will help, or we shift to v6 and v4 is irrelevant.  In
   either case, we're not in time to do anything significant for v4.

then in the fullness of time this will be widely recognised and all
will be well.


>>> I'm not expecting anyone to develop a recommendation.  I'm not expecting
>>> anyone to be happy with it.
>>
>> Does this mean you are going to develop the recommendation?
> 
> Yes, that's on the co-chairs.

OK.

>> My interpretation of this is that you and Lixia will write the
>> recommendation and while you will read or listen to what people have
>> to say about it, you won't promise to take notice of any of it or to
>> alter your text according to the views expressed by RRG participants.
> 
> Correct.

OK.

>> In the past, you have stated your intention to achieve consensus on a
>> recommendation.  Now I understand you don't intend to achieve or even
>> test for consensus on the recommendation you and Lixia will prepare.
> 
> Yes, that changed.

I would appreciate it if you announced such changes clearly on the
mailing list.  The only way I and I guess other people discovered
this was by me questioning you in recent days.

I think both you and Lixia failed to inform those who rely on the
mailing list what your plans are and how they have changed.


>>> Again, we've been trying to have that debate for three years.
>>
>> I believe you have not contributed much to the debate.
> 
> The feeling is mutual.

I have explained my reasons for thinking you made little contribution
- firstly a simple absence of you debating things at all, or in any
depth and secondly you banning or discouraging debate on "proposals"
(AKA "candidate architectures") while failing to lead by example by
debating whatever it is you wanted us all to debate, which you
described as "architecture".

Your negative assessment of my contribution is without any such
explanation.

I wrote critiques of, and discussed, every proposal but my own and
ILNP.  (I also discussed ILNP on the list and offlist with Joel, but
I never wrote an actual critique.  Still, ILNP is covered by my
critique of all CEE architectures.)  I recall that the proponents of
all these 13 proposals wrote on the list that they appreciated my
efforts and found the critique and/or the discussions which followed
helpful.

The draft RRG Report contains 5 critiques I wrote.  In the case of
LISP, an alternative was offered by Noel:

  http://www.ietf.org/mail-archive/web/rrg/current/msg05747.html

If I hadn't worked on these 4 proposals: hIPv4, Compact Routing, TIDR
and the formidable IRON-RANGER (which I devoted a week of my life to)
the RRG Report, for which you and Lixia are primarily responsible,
would not have critiques for these proposals.

How many of the proposals have you read?  Since you accepted the
proposals, you have only commented on ILNP.

I wrote about the architectural distinction between CEE and CES
(msg05865) and you dismissed this (msg06221) without any explanation
for why the things I mentioned (including Loc/ID Separation) are are
not both architectural and important.

I tried to describe the constraints we face due to the need for
widespread voluntary adoption:

  http://www.firstpr.com.au/ip/ivip/RRG-2009/constraints/

and you showed no interest - not even a criticism.


I regard your unexplained dismissal of my contributions to the RRG -
and to the report you and Lixia are primarily responsible for - as
ungracious, unfounded and unbecoming of the position of IRTF Chair.


>> You frequently requested we debate "architectures" - as if this was
>> somehow different from proposals.  Yet I don't recall an instance of
>> you leading by example with any such messages.
> 
> Not everything happens on the mailing list.

Can you cite where in the meetings you lead by example?  The audio
archives are available.

Since you exhort us all to debate things on the mailing list the way
you want them debated - something to do with "architecture", which I
think we are doing very well despite your disapproval - then I think
it is your responsibility to lead by example, on the mailing list.


>> I don't see how you could be sure of this if you haven't tried.
> 
> We've tried several times to simply agree on what terminology we would use.
> All attempts failed.

Can you point to any messages where you proposed some terminology we
should adopt?

See your recent msg06221 for how you dismiss, without giving any
reasons, a serious attempt to discuss terminology and architectural
distinctions between proposals.

My impression of your position is:

   Loc/ID Separation is doing it right.

   IPv4 is dead, toast etc. etc.

   ILNP is the only current proposal you support.

   You don't like reading or writing things longer than a few
   paragraphs.

   You are generally uninterested in what you regard as "engineering
   details" - but find that many other folks think that what you
   regard as such is "architectural" or at least important.

   You frequently don't feel it is necessary to explain the reasons
   behind your disapproval of other people's work - at least in a
   level of detail which allows the other people to debate those
   reasons.  (See note for a recent example.)

   When someone points out that you don't seem to be leading by
   example in the mailing list debate you want us to have, rather
   than admit this, you feel it is adequate to assert that you
   are doing this in some unspecified venues other than the list.

   You are highly confident that you know what is best for the future
   of the Internet.

   You have not encountered much, or anything, in this scalable
   routing phase of the RRG which has altered your view of what is
   "right".

Would you disagree with this, or add anything to it?

Can you mention other people who share your vision of how the
Internet should develop?

 - Robin



Here is an example of your lack of explanation when disagreeing with
someone - in this case refusing to debate something because you
consider it to be "engineering" rather than "architecture":

  http://www.ietf.org/mail-archive/web/rrg/current/msg06221.html

You don't explain why "engineering details" are "mechanism
trivialities".  Nor do you explain why the CEE/CES distinction I
believe is architecturally important is not so.  You explain the
reasons why you believe forcing all hosts to adopt the Locator /
Identifier Separation naming model is not "architectural".

I think this is a stark example of a pattern of dismissing views
contrary to your own in a manner which is uncommunicative - and which
I find hard to distinguish from being disrespectful to the people who
work hard to contribute these ideas to the Group you co-chair.