|
I have been somewhat troubled at the discussion about the
SG draft and proposed experiment, and I think part of the reason is that there's
a range of leashes being envisioned, with everything from "study groups are
where the villagers riot" to "the process for forming a study group is
indistinguishable from forming a working group".
If the SG process gets sufficiently close to the WG
process, I don't see the point. If you have to be able to convince a reluctant
AD, why would you not request a BOF? At least within RAI, the running code is
"meet as a SIPPING ad hoc, either at lunch or at 10:30 PM", and there doesn't
seem to be much process wrapped around that.
There are enough positive things about Gab's suggestion
(to use the IRTF as a home for WG explorations, in addition to research) that
I'd like to see a proposal for this alternative.
Dan (and Gab) have been around the IEEE a lot more than I
have, but one of the translation issues I'm wondering about is that the go/no-go
decision in IEEE seems a lot more straighforward in IEEE than in IETF - if you
have a plausible PAR and reasonable answers to the Five Criteria,
doesn't the decision tend toward "go"?
At least in RAI, we've been approving new working groups
fairly frequently, but there have been times, especially in specific areas,
where there was a lot of resistance to chartering new working groups, based
solely on the number of existing working groups and the load that places on ADs
(http://www.watersprings.org/pub/id/draft-klensin-overload-00.txt might
have been outside the mainstream, but I don't think John and Marshall were WAY
outside the mainstream in their 2002 draft).
I am intrigued by the idea that we can't charter new work
because we have too many working groups now, because a quick examination of my
own area shows about five-or-so working groups that are amazingly close to
closing down, but they do not seem to close down, IETF after IETF. Is this
different in other areas?
I would ask people to spend a little time thinking about
the idea that the number of slots in an IETF week should determine the number of
chartered working groups. That seems like the tail wagging the dog.
Spencer
----- Original Message -----
Sent: Monday, October 08, 2007 11:19
AM
Subject: RE: Comments on
draft-aboba-sg-experiment-02
I have seen the functioning of SGs at the IEEE and agree
that they can be useful, but I'm not sure about how it is being "translated"
into the IETF>
It occurs to me that we don't need to invent a
new process here. The IRTF houses different types of "research" groups: some
are meant
to be long-lived, some are meant to meet during IETF, some never
meet, etc. There also are some RGs that have operated in a manner
similar
to the study groups being proposed: NSRG (name spaces research group), for
example. And some that have been started as an
alternative
to petitions to form a WG, and which would seriously benefit
from
having a tighter charter with specific milestones and expectations
(e.g., p2prg RG).
RGs are created with all sorts of different
goals in mind. All that the IESG needs here, I think, is to start an RG to
probe further into
a given issue, and keep it on a short leash along the
lines stipulated for the SG: e.g., milestones, meetings during IETF, explicit
IESG liaison, etc.
But the point is that these conditions need not be the
same for each such RG/SG.
I also think this is something useful
the IRTF could do, as most often than not, it actually doesn't do any
research. The IESG wins, the IRTF wins,
the IETF
wins.
-gabriel
> Date: Mon, 8 Oct 2007 15:45:44 +0200
> From:
dromasca@xxxxxxxxx
> To: lear@xxxxxxxxx;
ekr@xxxxxxxxxxxxxxxxxxxx
> CC: jari.arkko@xxxxxxxxx; ietf@xxxxxxxx;
iesg@xxxxxxxx
> Subject: RE: Comments on
draft-aboba-sg-experiment-02
>
> The way I see it the problem
that this proposal tries to solve is about
> helping the IESG and the
community to make a better decision when the
> forming of the working
group us discussed. It is not about bringing more
> work to the IETF, it
is about making sure to a better extent that the
> right work is being
brought into the IETF. In the absence of such a
> process what we see in
many cases is the formation of ad-hoc groups,
> which is not necessarily
bad - but why not charter them with a set of
> clear questions which may
help the IESG and the whole community reach a
> more educated decision?
>
> Regarding terminology, the term 'study group' is used in
this proposal
> in a way similar to how the IEEE is using it.
>
> Dan
>
>
>
>
> > -----Original
Message-----
> > From: Eliot Lear [mailto:lear@xxxxxxxxx]
>
> Sent: Monday, October 08, 2007 3:30 PM
> > To: Eric
Rescorla
> > Cc: Jari Arkko; ietf@xxxxxxxx; iesg@xxxxxxxx
>
> Subject: Re: Comments on draft-aboba-sg-experiment-02
> >
> > If I understand the purpose of this experiment it would be to
> > provide ADs some indication of level of interest and ability
> > to succeed. I see no reason why we need to formalize this
> > within the IETF. Furthemore, the terminology is problematic.
> > We are overlapping a term that is commonly used by the ITU
> > the way working group is used by the IETF.
> > Let's
not make the process any more confusing than it already is.
> >
Finally, milestones for such "study groups" seem to me inappropriate.
>
> It may be that a topic is uninteresting for quite a while and
>
> then picks up. ANY way to demonstrate that interest and
> >
ability to succeed should be sufficient, regardless of how
> > much
time has passed.
> >
> > Eliot
> >
>
> _______________________________________________
> Ietf mailing
list
> Ietf@xxxxxxxx
>
https://www1.ietf.org/mailman/listinfo/ietf
Help yourself to FREE treats served up daily at the Messenger Café. Stop by today!
_______________________________________________
Ietf mailing
list
Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf
|