Re: [rrg] [IRSG] IRSG Review: draft-irtf-rrg-recommendation-12.txt

[Robin Whittle <[e-mail]>]

Hi Tony,

You wrote:

>> This is the first time I am aware of you expressing interest in 
>> working on:
>> 
>> http://tools.ietf.org/html/draft-irtf-rrg-design-goals-01
>> 
>> since 8 July 2007 when you created this version 01.
> 
> Please read:
> 
> http://www.ietf.org/mail-archive/web/rrg/current/msg07262.html

OK - I missed this message last week.  My point remains that there has
been no revision to this document for over 3 years, and that my three
attempts to discuss improvements on it, in July and December 2007, did
not result in any significant discussion from you or anyone else.


>> As I wrote in March this year:
>> 
>> http://www.ietf.org/mail-archive/web/rrg/current/msg06160.html
>> 
>> this version 01 resulted from very few list messages.
> 
> This is correct.  There was a considerable amount of discussion
> offline and during meetings.

OK.


>> We should have worked much harder on this.
> 
> Actually, we did work really hard on this.  I realize that you
> didn't see it because you didn't attend, but we spent many hours
> discussing these issues.  It is grossly unfair of you to ignore
> that.

I didn't mean that you and others didn't work hard on the Design Goals
leading up to version 01.  I meant that in the three years since, we
should have worked harder on it.  I do not regard the version 01,
followed by three years of no further discussion, as fulfilling one of
the requirements of the Charter:

  Milestones

      It is hard to forecast exactly how the consensus process
      will play out, and therefore the precise timing to achieve
      various milestones. However, as a starting point, the RRG
      will produce a document enumerating the design goals for a
      new routing architecture.

I did my bit, starting a week after you released version 01 - and
there was no interest.  After Iljitsch van Beijnum chided me for the
lack of goals and non-goals in the first Ivip I-D, I did a proper job
of it in Ivip-arch-04, earlier this year.  Anyone could have used this
as a basis for discussion, or as input for revising the Design Goals.

My attempt at writing a Recommendation (2010-03-09):

   http://www.ietf.org/mail-archive/web/rrg/current/msg06219.html

contained 1830 words on design goals - about 80% more than version 01
of the RRG Design Goals I-D.  I don't recall anyone debating these or
suggesting they be used to revise the RRG Design Goals I-D.

If you, Lixia, and other RRG participants had continued to work on the
Design Goals from mid-2007 then I think we would have had a much more
comprehensive document as a basis for the rest of the RRG work.

This is all the more important now, since the RRG Report does not
contain any proper discussion of goals and non-goals, and the 500 word
limit on the Summaries makes it impossible for each proposal to have
its goals and non-goals clearly described.


I would support the publication of version 01 of the Design Goals I-D
as in informational RFC, provided it carries a disclaimer that it was
developed in the early months of the 3+ years of the RRG's scalable
routing work, and that it was not widely discussed, cited, or revised
since then.

If you want to work on a significant revision, my suggestions are
still probably relevant:

  http://www.ietf.org/mail-archive/web/rrg/current/msg00203.html

  http://www.ietf.org/mail-archive/web/rrg/current/msg00733.html
  http://www.ietf.org/mail-archive/web/rrg/current/msg00786.html

  http://www.ietf.org/mail-archive/web/rrg/current/msg06219.html 17.2

That would be a good project, albeit several years late.  I sense that
most people are keen to wind up their RRG involvement, so I can't
imagine that there will be significant discussion and revisions to the
RRG Design Goals.  If this does occur, and result in a better
document, then that would be great.  However, I think its publication
as an RFC should involve a clear statement of how significant
revisions were made after the proposals and Report was finalised and
so do not reflect a shared understanding between the designers of the
various proposed architectures.


>> I think the proposals in the Report are not based to any
>> substantial degree on the Design Goals version 01, since that was
>> a very early document, rarely discussed or mentioned on the list.
> 
> That was the choice of the designers of the proposals.  The goals
> were there so that the group understood the baseline.  They were
> also there for participants to use in critiquing the proposals.

Yes, but that was version 01, from July 2007.  Since then, more people
became involved and many ideas and proposals have been developed.


>> These people wrote about 39% of the list messages to date.  These
>> 18 people do not include anyone from the LISP team, or proponents
>> of about half the 14 proposals.
> 
> Consensus is only taken across those who choose to participate.

I understand this.  My point is that the people who seem to be
involved in the RRG now only represent a subset of those who have been
involved in the last few years, and who contributed to the proposals
in the Report.


>> Even if these 18 people, and whoever else is still engaged in the
>> RRG, did agree on a revised Design Goals document, it would not
>> be a document which was used in the preparation of the
>> proposals.
> 
> Again, you're inferring from your own personal experience.  It is
> grossly unreasonable for you to state what the remainder of the
> group was or was not thinking about.  You simply do not know.

OK - here is a clearer version:

It appears that at this late stage of the RRG, there are too small a
subset of the RRG contributors (as measured over 3 years by mailing
list messages, and by their proposed architectures in the Report) to
arrive at a document which could be properly regarded as any kind of
reference for the proposals in the Report.

The version 01 is so old, and has so rarely been discussed or cited,
that it cannot properly be regarded as a basis for these proposals, in
the absence of the proposals themselves explicitly citing it.  Any
substantial revisions to it now would have been made well after the
proposals and the Report was written.

  - Robin