Re: RRG thoughts (was [Bier] [pim] Q on the congestion awareness of routing protocols)

[Toerless Eckert <[e-mail]>]

Joel:

a) Purpose and opertions of IETF and IRTF are evolving both through new rules but also
   through new choices by leadership. I see this comparing some IRTH groups created
   recently vs. earlier. I see this in how we now have SIGs in OPS, i see this
   in AFAIK a growing set of groups having more interims or other in-between-full-IETF-meetings
   regular meetings since we have meetecho - just to name three examples. So i would
   appreciate if you could give concrete examples of what you think would be 
   "not fitting" in your opinion.

b) Some idea of how this could be structured:

   We use "routing-discussion" as the main communications forum.

   We designate some wiki page to capture proposed input (that has no other place
   in IETF, IRTF): drafts, presentations, external research papers/presentations.
   So the main work is to have folks who like chairs manage this.

   We designate a monthly slot 1..2 hours for the forum, but the slot is only
   active with a 2 week earlier invite when there is content (like a regular, but
   on-demand interrim). The content will consist of requested or invited talk/discussions
   (requested via mails/thread on routing-discussion).

  The main thing is to build the wiki up to be clear about procedures/goals:
    1. It's free (just create free datatracker account if you do not have one).
    2. Bring unfinished research / ideas to the forum, to present/get-feedback,
       can simply refer to your research paper (draft), no need for a draft
       submitted to IETF.
    3. Meetings will have also open forum section so folks who can't even figure out
       2. so far can ask questions.

   Of course, IETF note-well does apply (by mere use of standard IETF tools - 
   list, meetecho or webex for the meetings (webex has live transcription, maybe
   also translation ? may make it a better tool than meetecho !).
      
   There needs to be promotion of this idea to research community of course.
   Not really difficult when one puts together flyers and hands them out to
   colleagues attending research conferences dabbling in routing.
 
Its IMHO not really rocket science. And i don't claim above are the best ideas,
but retainly better than business as usual. Its can also double as a kind of dispatch
front-ending RTGWG . The main issue with RTGWG being that its a) more difficult
to figure out how to get into free, b) more difficult to be productive if you
attend remotely a primarily in-person event, c) some group of RTGWG participants
does not like to see the WG spend so much time on non-WG work

Cheers
    Toerless
   
On Sat, Dec 10, 2022 at 01:28:38PM -1000, Joel Halpern wrote:
> The idea of a suitable forum for informal conversation between researchers,
> vendors, and operators, seems useful.  I do admit that I tend to be
> skeptical of regular meetings without agendas or clear purpose, as such
> things tend to devlove (in various different unpleasant ways.)  More
> importantly, this does not fit well with the purpose and operations of
> either the IETF or the IRTF.  So figuring out structure and rules
> (preferably somehow making everything discussed subject to IETF note well
> and similar policies) seems tricky.
> 
> Yours,
> 
> Joel
> 
> On 12/9/2022 3:33 AM, Toerless Eckert wrote:
> > On Tue, Dec 06, 2022 at 12:54:57PM -0800, Dino Farinacci wrote:
> > > > path exploration? but consider the shadow pricing...
> > > Just something semi-formal where we can meet regularly. But first, please lets articuulate the problem very clearly. And bring operators in so they can confirm the problem that researchers are stating are real problems that need solving.
> > I actually would love to just have a standing forum to better bring routing researchers
> > together with industrial routing experts. Aka: not focussed on a specific issue
> > (as Tony's past RRG instance). E.g.: where researcher can ask questions to the experts, or propose
> > research and ask if/how this is useful to the industry, and where industry folks can
> > ask for researchers to look into specific issues (i think there are for example a lot of
> > simulations to investigate behaviors that we'd like to have from the undustry).
> > 
> > Not sure if "Research Group" is the right name for this. I think it would be a lot
> > closer to the SIG concept (https://wiki.ietf.org/en/group/iesg/specialinterestgroups),
> > except that that concept was built and specified around interaction between IETF
> > and operational considerations of a specific community. Not on bringing more researchers
> > back to the IETF.
> > 
> > We do have a subset of what i think such an RRG would do in rtgwg - researchers bring ideas -
> > and then they most often figure out that they can only proceed when they throw themselves
> > fully into the RFC process (which most of them cannot/wantnot do). And from what i
> > understand even this is not always welcome by rtgwg, because it does take a significant
> > amount of time that folks feel should better be spent on actual WG deliverables.
> > 
> > Of course, i am mostly interested in the ietf->research direction, e.g.: where the IETF
> > community can better raise the questions of interest to be researched because the way i
> > see it, there is no forum whatsover for this part.
> > 
> > Cheers
> >     Toerless
> > 
> > _______________________________________________
> > routing-discussion mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/routing-discussion

-- 
---
[e-mail]