ietf
[Top] [All Lists]

Re: WG management

2005-06-23 12:41:19
The open door policy might call for some creativity in resource management, but it in no way guarantees that we are unable to manage our aggregate resources

It pushes us into a significantly less predictable situation than with
committed resources. That's guaranteed.

Ahh. I think I see where we differ. You are focusing on whether the IETF can make guarantees about when it will deliver things. Whereas I am concerned with placing limits on the folks who seek to consume IETF resources.

No, we cannot guarantee when things will complete. Yes, we need reasonable flexibility.

But if some folk come to the IETF and ask to consume IETF resources, we have more than a right to require that they consume those resources reasonably and productively. We have a responsibility.

They consume IETF secretariat time, IESG time, meeting rooms, etc. They congest our queues. They draw from our talent pool (such as it is.) So we definitely have a responsibility that they consume our resources productively.

I've heard rumors that ADs -- and maybe even IETF Chairs -- sometimes feel as if their workload is a bit high. Perhaps it would be less so if the groups that were trying to work here were actually responsible for doing timely productive work, rather than dragging things out for hefty fractions of a decade?

And we have plenty of history of refusing such folk or of terminating such folk, when we feel that their consumption is not (any longer) warranted. We refuse to charter groups who lack focus. We shut down working groups that fail to produce.

What WE have failed to do is to apply this performance demand with much rigor.

My suggestion is merely that we be more strict in applying exactly those demands that we have applied many times in the past: We need to require that a group that is chartered show enough commitment -- by virtue of participation -- and enough focus -- by virtue of crisp clear charters, documents, etc -- to make it likely that their work will be timely and useful and that they then are.


If there is a strongly dominant constituency in favor of something, then that something gets done.

Of course. But I don't see the connection with rough consensus. Exactly the same is true in voting standards bodies - if there is no majority constituency,
nothing gets done.

Actually it is often quite different. A voting body can allow itself to make a decision when virtually nobody participates, depending upon their rules for a quorum. More importantly, voting usually makes a decision based on whether more than half of those involved say yes.

By contrast, the IETF pattern is to require that there be a body of seriously active folk and that their decisions have at dominant coherence (somewhere in that mystical space that is much more than half and maybe less than unanimous).

How many people compose "a body of seriously active folk"? Beats the heck out of me. But let me suggest that if we make a point of looking honestly at an effort during its formation and as it is prosecuted, we will typically see very clearly whether the activity is of a few folk pursuing a personal -- albeit possibly appealing -- whim or is of a seriously active industry constituency.

Asking the question honestly is the hard part.


Exactly the same in a voting organisation. Except (and I don't
personally like this) they tend to meet deadlines by holding
votes on whatever text they have at that time.

Real projects meet deadlines by delivering what they have committed to. And when they fail to do that, their existence is legitimately threatened. Real projects that succeed to not get to commit to 2 years of effort and then linger for 5 or 8, producing no measurable benefit to their customers.


Require that an effort begin with -- and continue to demonstrate -- a serious constituency in terms of numbers and activity, and most of our problems disappear.

Exactly the same in a voting organisation.

Brian, you seem to be resorting to the voting model as if it somehow relates to my point. I think it doesn't.

First, there is nothing that says that the IETF model must have nothing in common with membership/voting organizations. It merely says that there are important differences. Second, no matter how a group is organized, it has to be productive.

Frankly the tone of your responses to this topic seem pretty consistently to be one of a hopeless shrug, as if to suggest that there is nothing proactive (or even reactive) that can be done to *require* timely productivity.

This, of course, is a good way to keep a pleasant tone in the group, but it also seems like it risks letting the organization simply die from lack of relevance.

One might even predict that there will be less and less participation and that it will take longer and longer to produce work that has less and less impact.

Oops.  That is what is already happening.


Ignore that requirement and we are, instead, we are left with congestion, individual idealism and vetoes... and a belief that there is nothing we can do about it.


So if you have a WG with only 6 active participants whose only job is a large
MIB, do we deem it too small and too inactive? Whereas a WG with 100 active
partcipants and 30 drafts in progress but no drafts reaching the IESG is OK?

Who is expected to use that MIB? Why do we think that these 6 people represent those people well enough to make the use likely? Is the work progressing productively?

"no drafts reaching the IESG" would seem to answer the question of productivity rather directly.

/d

Attachment: dhc2.vcf
Description: Vcard

_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf
<Prev in Thread] Current Thread [Next in Thread>