IRTF summary & pros/cons

["Phil Rzewski" <philr@inktomi.com>]

On a recent CDI editor's call, we discussed contingency plans in the event
that the IESG does not approve us as a working group. One possibility
offered was that we could take our work to the IRTF. I took on the task of
summarizing the IRTF and offer some pros/cons of going that route.

The www.irtf.org web site is pretty empty. Everything I learned I got from
RFC2014 (BCP8) "IRTF Research Group Guidelines and Procedures", so if you
want the full story, just read that. For those that want the reader's
digest version, I've tried to summarize.


The following are similarities/differences/parallels between the IRTF and
IETF:

- Instead of "Working Groups", they have "Research Groups".

- Participation in RGs is based on "individual contribution", just like the
IETF.

- RGs are expected to be "long-term" in nature, unlike WGs that are usually
pressured to produce their results and close up ASAP.

- RGs have "one or two chairs", much like WGs.

- RG chairs have similar roles to WG chairs, either doing or delegating
tasks such as:

    - Managing group processes

    - Moderating the group mailing list

    - Preparing face-to-face meetings

    - Documenting/distributing group results

- RGs do NOT have the concept of "consensus", since they are producing
research, not standards.

- RG results are typically published as research papers or
Informational/Experimental RFCs.

- RGs can have open or closed ("limited") membership, unlike IETF WGs where
it's always open.

- RGs are formed in a similar way to WGs

    - Individual/group submits charter


    - Reviewed by IRSG

    - There's several consideration criteria that must be met before it
gets formed

- There's no formal "IRTF meeting" where all RGs come together. Rather,
they encourage the RGs to co-locate meetings with IETF.


Now, for some pros/cons... feel free to add your own or debate these...
they're opinions...

Cons:

- One editor advised: "Unless you form a closed group, you'll get a lot of
new participants that end up introducing new things in the name of
research". This could be seen as a "con" if it derails us from our original
purpose of wanting to quickly facilitate content internetworking in the
real world.

- We would not be on a standards mission anymore. Rather, we'd be on a
research mission that is one step removed from standards. This is arguably
a step back and may not be looked on favorably by the market forces that
saw this as a problem to be solved in the first place.


Pros:

- The first "con" above could also be a "pro". Perhaps some of the new
minds could help us better understand the fundamental issues and make it
ever-more-obvious how to scope the work that eventually gets pushed down to
IETF.

- An editor gave the example of AAA and how having the "IRTF umbrella kept
AAA in the game", since work filters down to IETF WGs as it becomes more
concrete. In other words, it would be a way we could keep content
internetworking alive, as opposed to, say, publishing our current drafts as
individually-submitted Informational RFCs and then giving up.

- Judging from the wide scope of some of our work, a research forum may
well be the ideal home for this topic. Perhaps we're not ready for
standards track with our current efforts. Perhaps an IRTF group would help
spawn off smaller, more-focused IETG WGs that would have a more
crisply-defined charter.


I now submit the issue to the group to seek consensus on this as a possible
contingency plan.

--
Phil Rzewski - Senior Architect - Inktomi Corporation                  
650-653-2487 (office) - 650-303-3790 (cell) - 650-653-1848 (fax)