Re: IAB comments on draft-baker-liaisons-00.txt

[Stephen J Trowbridge <sjtrowbridge@lucent.com>]

Leslie,
Inline.

On 8/17/2003 8:22 PM, Leslie Daigle wrote:
> I'm sorry that the comments have been heard as criticism of
> the effort, and not as input on what we believe is needed
> to call the work done.
> 
> Recap'ing what we have today:  we *have* a mechanism for submitting
> non-I-D liaison communications to the IETF -- posting to the
> statements@ietf.org mailing list gets the Secretariat's attention.
> N.B., many such liaison communications are accepted in non-ascii
> format (i.e., Word).  The problems we have had with that process
> are in determining who has responsibility for responding to those
> communications, and making sure that they are handled in a
> reasonable timeframe.  Scott will recall that I was on the IESG calls
> when various related issues were discussed:  liaison communications
> being ignored or otherwise inappropriately responded to.
Of course this does something for the incoming direction, but it has
become sort of a bit-bucket. Things get posted, but nobody looks at it
and nobody responds. Also, once posted, things are there forever.
Our document was intended to build a more meaningful process around
liaison statements than "the sender sends to statements@ietf.org and
the secretariat posts it". This by itself is not enough.


> 
> So there is no question that there is the need to improve our
> processes, so that we may communicate more effectively with other
> SDOs.
> 
> But, I see two major remaining pieces necessary to bring this
> work to completion:
> 
> 	1/  Addressing the very particular problem we have
> 	    today, which is independent of mechanisms for
> 	    exchanging communications:  identifying who is
> 	    responsible for responding to liaison communications,
> 	    and the scope of authority (within the IETF) for
> 	    individuals who emit liaison communications to other
> 	    SDOs.  I believe that is part of what Geoff was
> 	    the IESG's to sort out.
I thought that this IS addressed in the document. If you have
another suggestion, please propose text. Do you think that Geoff
is proposing to put IESG in the loop on every incoming liaison
statement? I fear that this will introduce delay, perhaps cause
reply deadlines to be missed, and create another black hole into
which things could disappear. Particularly in the case of an
established relationship with bi-directional communication, it is
best to assign responsibility immediately according to how the
incoming liaison statement is addressed (e.g., something addressed
to ccamp from ITU-T Q.14/15 gets immediately assigned to Kireeti
and Ron).

> 
> 	2/  Clarifying, within this document, how this proposed
> 	    process fits in the overall scope of our intended
> 	    communications with the rest of the universe.  That is,
> 	    there are still times when it is appropriate for
> 	    other organizations, including SDO's, to put their
> 	    contributions together in the form of an I-D, and
> 	    have it considered as individual input, even as we
> 	    recognize that there are times when that is *not*
> 	    the appropriate mechanism.
I agree that if another SDO wants something standardized through the
IETF process, and if that is the purpose of their communication, they
need to put it in the form of an ID and follow the same process as
everybody else. End of story. If you want, we can say this in the document.

But for other purposes, an ID format is not called for. Let me point
out that in the case where an individual is not aiming for a result of
a published document (xx standard, informational or experimental RFC, BCP)
they don't put their input in the form of an ID either! They just write
an email to the list. But as we have discussed earlier, an email to the
list is a lousy interface for an external SDO. IETF should decide what
a consensus reply is and tell them rather than asking the other SDO
to watch our list and decide for themselves what the IETF position is.

> 
> Again, understand that I chose to review the document from the
> standpoint of someone in J-Random-self-proclaimed-SDO desiring
> to communicate with the IETF.  I can't imagine how the IDN WG
> would have come to closure had the chairs had to spend their time
> explaining to various organizations how this document ought to be
> interpreted or was not applicable to the liaison communication they
> had just submitted for discussion on the agenda.  (I cite IDN
> as an example where I know there were various "standards development
> organizations (SDOs), consortia, and industry fora" who were
> desirous of participating through communication means more like
> statements and less like IETF process).   I know that
> to you it is crystal clear that these communications would only be for
> organizations with which we have established formal liaison
> relationships, but you might note that the document does not
> in fact SAY THAT.  Hence my concern about this document leading
> to DoS scenarios.
This is back to the question of whether we need something along
the lines of an IETF version of ITU-T Recommendation A.4 to indicate
how we decide who to communicate with. As I said before, my own
opinion is that while this is a worry, there is no urgent need
for such a process at this point in time, and that we can hold off
on creating additional machinery here until (and unless) it ever
gets out of hand with the number or types of other organizations
we end up communicating with. Do our IESG and IAB friends think that
we need such a process now? If so, we can try to create a seperate
document for this purpose (probably a lighter weight and IETF-ized
version of ITU-T Rec. A.4). (BTW, we DO have a VERY lightweight process
for this now that can be described in 3 words: "The IAB decides".
Let's be sure that this is not sufficient before we go off the deep
end with additional process).
Regards,
/Steve

> 
> To reiterate -- these are critiques of the document.  Clearly,
> in at least some cases (e.g., the last point above), the
> document isn't documenting what's in your thinking, so we ought
> to be able to agree that there's room for improvement in the
> document?
> 
> If not, then I, too, am at a loss as to how to proceed.
> 
> Leslie.
> 
> 
> 
> 
> Fred Baker wrote:
> 
>>At 06:37 PM 8/15/2003 -0400, Leslie Daigle wrote:
>>
>>
>>>This seems to open the door to suggesting to other SDOs that they 
>>>don't need to interact with the IETF the same way as all individual 
>>>participants of the IETF -- through I-Ds.
>>
>>
>>Something you might want to carefully consider and have a good answer to:
>>
>>The position of some in the IETF has consistently been that any 
>>communication from another body to us should use our communication 
>>mechanisms. Any time the ITU wants to say something to us, they should 
>>use an internet draft. That draft should be expressed as a flat ASCII 
>>file, in order to avoid proprietary formats, and have certain headers 
>>and trailers describing copyrights, status for usability as a source of 
>>text, and so on.
>>
>>In all fairness, it would be perfectly reasonable for other bodies, such 
>>as ITU, to insist that any communication from the IETF to them use their 
>>communication mechanisms. Any time we want to say something to the ITU, 
>>including such memoranda as a reply to an internet draft posted by a 
>>person claiming to represent the ITU, we send a meeting contribution. A 
>>meeting contribution is a file in Microsoft Word 2.0 format, on the 
>>letterhead of the company that is a sector member of the ITU which we 
>>represent, representing the collective opinion of that company on a 
>>certain topic. The exact contents of the contribution are largely 
>>irrelevant, however, as it will not be directly referred to and cannot 
>>actually be counted on has having been read. Rather, the contribution 
>>serves as a ticket for a person attending the meeting to get an 
>>opportunity to make a presentation on the concept contained therein and 
>>participate in the midnight editing sessions from which the ITU document 
>>actually results.
>>
>>Yes, I have worked in certain ITU processes to develop documents; I do 
>>know whereof I speak.
>>
>>If we would not like ITU and others to make what we would consider 
>>outrageous demands on how we might communicate with them, it might serve 
>>us well to consider being flexible on the manner in which they 
>>communicate with us. Inflexibility on our part does not help build a 
>>productive relationship.
> 
>