Hi -
To sort of close off this thread. Thanks for all the commentary.
What I sort of got was "been there, done that, didn't work" for various
reasons.
I'm not completely convinced we couldn't do better going forward, but
I'm more convinced that my vision of this is probably impractical as stated.
I did note one thing missing from my original proposal - a "carrot" to
go with the "stick" of "you will die in 1 year". In otherwords,
consideration the participants wouldn't get in sucking it up in a normal
WG. I don't know what that might be.
Thanks again for the discussion - Mike
-- 30 --
On 8/11/2025 07:07, Jeffrey Haas wrote:
On Aug 6, 2025, at 12:20 PM, Bob Hinden <bob.hinden@xxxxxxxxx> wrote:
Stewart,
On Aug 6, 2025, at 12:42 AM, Stewart Bryant <stewart.bryant@xxxxxxxxx> wrote:
A reality check. BFD WG was supposed to be done in two WG meetings.
Good point. It’s a lot easier to create a w.g. than have it complete on time. Getting things done on a schedule is not an IETF strength.
I also think that “tiny” isn’t a good description, the focus should be getting something done in a few IETF meeting cycles, not keeping it small (aka tiny).
As the poor soul whose first chair job was that "short" BFD working group... yeah, it didn't work out that way. It is only this month that the last bits of work are finishing in a way the WG could be closed - but just as likely is going to be left open and in maintenance mode.
The learnings I would offer from the BFD experience:
- You only move as fast as the desire of the WG as a whole to actually push things out as RFCs.
- Your core team can move things fast enough if you can keep focus tight for the ideally < 2 years of work. If you can't pump out the RFCs in that time, people simply get distracted. I.e., there is finite attention you can muster.
- Once you've lost people's attention, especially if the draft is shipping code, people stop caring about process work.
I wasn't a well enough trained/strong enough chair to herd the cats fast enough. The ADs tried.
The larger lesson is that IETF is very good at starting work, and getting it far enough along to have running code. We're not always great at *finishing* the work and publishing the RFC. Certainly some of the recent charter focus is on trying to do better about that.
-- Jeff
