Re: IAB comments on draft-baker-liaisons-00.txt - Need for otherthan IDs

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

Scott & Leslie,
Clearly we need to go one by one or we will get hopelessly entangled
in an ever growing mess of comments and discussion. I'll add what I
think this issue for this email is to the subject line so we can keep
them straight.

A couple of extra thoughts on why IDs can't be the only thing:
If another SDO wishes something to be standardized in IETF, certainly
they should send it in as an ID. Nobody gets a free pass to not follow
the process to produce standards in the IETF.

But while this may be one purpose of a liaison statement from another
organization, I think that there are many other purposes which are not
served well by this model. Two obvious cases are:
- Another SDO wishes to solicit comments or reaction from the IETF on
  a standard they are developing. I think that in IETF we would like to
  participate in the larger community of global standardization bodies,
  but to require that another organization convert their documents to
  ASCII with stick figure diagrams before we will look at it and offer
  comments (pdf or html, fine, but not ASCII) will only serve to obstruct
  this kind of valuable communication. If we make it too difficult for
  another organization to solicit comments from IETF on their documents,
  they won't do it.
- IDs are not a satisfactory interface for communication in the direction
  from IETF to another SDO, either in the case of a reply or in the case
  of a spontaneously generated communication from IETF. For an individual,
  they can submit an ID and see how others react to it on the email list
  to know how well it is received. But it is not such a practical interface
  to expect another SDO to watch the IETF email list to know what the
  reaction was to something they sent to IETF. For most SDOs, it could
  hardly be considered any kind of official input to their process to
  extract something from another organization's email list. And even if
  it were, do we in IETF want the other SDO to make their own judgement
  about what the IETF consensus response was by watching the email list?
  I would think that we would rather make the judgement about what the
  consensus is within IETF and tell the other SDO (officially) what our
  judgement was. Internally, we expect WG chairs and ADs to judge consensus
  out of a mass of contradictory emails for internal decisions - we do
  not think it should be different for anything requiring an external
  response.

Regards,
Steve

On 8/15/2003 4:54 PM, Scott Bradner wrote:
> there are a number of different issues raised - it may be better
> to not hit them all at once so I'll comment on the ones that seem
> to need comment one by one
> 
> 
>>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.
> 
> 
> I disagree that "all individual participants of the IETF" communicate with 
> I-Ds
> 
> fwiw - I see two rather different cases
> 	1/ a note that says 'you might wantto look at X' or
> 	   'just to let you know the X SDO SG 3 is working on that
> 	   topic' or 'if you use value X for varialble Y it will 
> 	   be compatable with what we are doing'
> 
> 	2/ here is a proposal for technology we would like you to 
> 	   consider
> 
> I do not think that its reasonbable to insist that the other SDOs use
> I-Ds for #1 - these sorts of things are normally done in the 
> IETF with messages to mailing lists
> 
> for #2 the normal way is with I-Ds if the idea is for the WG to
> do something with the text - but its also not that uncommon to
> get someone sending a URL to a list and suggesting that the WG
> folk take a look at this paper by Craig.
> 
> this particular assertion has been made before (tell them to use I-Ds)
> because we do
> 	1/ I don't think "we do" for most of the cases this would
> 	   deal with
> 	2/ insisting on I-Ds sounds like an effective to make sure the
> 	   IETF working groups stay uninformed about what others
> 	   (in other SDOs and researchers) are doing
> 
> note that RFC 3356 (interaction with the ITU) says:
> 3.3.2 ITU-T to IETF
> 
>    A Study Group or Working Party may send texts of draft new or revised
>    Recommendations, clearly indicating their status, to the IETF as
>    contributions in the form of Internet Drafts.  Internet Drafts are
>    IETF temporary documents that expire six months after being
>    published.  The Study Group or Working Party must decide that there
>    is a benefit in forwarding them to the IETF for review, comment and
>    potential use.  Terms of reference for Rapporteur Group meetings may
>    authorize Rapporteur Groups to send working documents, in the form of
>    Internet Drafts, to the IETF.
> 
>    In these cases, the document editor would be instructed to prepare
>    the contribution in Internet Draft format (in ASCII and optionally
>    postscript format as per [RFC2223]) and submit it to the Internet
>    Draft editor (email internet-drafts@ietf.org).  Alternatively, the
>    Study Group, Working Party or Rapporteur Group could agree to post
>    the document on a web site and merely document its existence with a
>    short Internet Draft that contains a summary and the document URL.
>    The URL can point to a Word document as long as it is publicly     
>    available and with the understanding that it will not be eligible for
>    publication as an RFC in that format.
> 
> so the intent is not to say that I-Ds are not a good thing, they are just
> not the only thing
> 
> Scott