Re: [mpowr] Re: Getting Bad Ideas to Fail Early

Greetings,

Apologies for my silence...

I recognize that this has morphed into an icar discussion.
Perhaps we should move this thread there.  I've tried to cut out
stuff where I didn't really have any more to say.

On Tuesday 13 January 2004 21.49, Alex Rousskov wrote:
> On Tue, 13 Jan 2004, David Partain (LI/EAB) wrote:
> > Who will a working group listen to?
>
> From process point of view, the working group does not have to listen
> to any reviewer. They can, technically, mark each submitted review
> with "we disagree" and let IESG resolve all the conflicts.

This is certainly one world view.  Another possible world view is
that the IETF decides to create a review process that a WG must
listen to.  Yet another world view is that we should depend upon
WGs to do the Right Thing in dealing with constructive criticism
and take the proper technical decision.  I'm sure there are a
plethora of other possibilities as well.

> > They currently must listen to the IESG.  Is the IETF willing to
> > invest reviewers with similar weight?
>
> IETF should be willing and able to design exit criteria that allow
> some documents to get published without IESG involvement.

Why is that the case?  I'm unconvinced that that would be a good
thing.

> In other words, reviewers can "clear" a document (if all reviewers and
> the WG agrees and there is sufficient coverage). However, reviewers
> cannot kill the document. Only IESG can.

Again, isn't that what has to be decided?  I'm unconvinced one
way or another.  I see real value in "review early, review often"
and with a sledgehammer that can be 

One of my big concerns, however, is that I very seriously doubt
that there are going to be very many cases where reviewers are
going to be able to point to clear "principles of the Internet"
in using their sledgehammer.  I suspect that those wanting to
weild a sledgehammer are going to want to do so more because of
gut reactions, some overriding sense of unease.  I'm certain that
they'll be able to point out technical reasons, but I don't know
if it'll be possible to point to particular clauses in our
ephemeral Internet architecture.

> > What do we do if the WG refuses to acknowledge that an idea is
> > "bad" and forges ahead?
>
> Wait for IESG to decide. It would be nice to design some procedure for
> IESG to decide sooner (early) rather than at PS stage.

Might be a very good idea, but at the risk of making them
interrupt driven, which I personally find to be a hard way
to work.

Cheers,

David

_______________________________________________
mpowr mailing list
mpowr@ietf.org
https://www1.ietf.org/mailman/listinfo/mpowr