I think this is a nice document, with many useful suggestions and insights. I think it would make a great ION if we still had IONs, a fine IESG statement, or perhaps an I-D that was reissued every 5 1/2 months to keep it active. The more I think about it, the less I like the idea of publishing it as an RFC. The problem is that RFCs are forever. RFCs subjected to IETF Last Call and published in the IETF Stream --especially ones that advise on IETF processes-- are also official, at least in the sense of representing some level of community consensus and IESG approval. Publishing this document in the RFC Series runs a considerable risk of causing (or at least reinforcing) exactly the situation the it argues against -- creating a permanent structure of meetings for which "informal" is as much of a misnomer as "bar BOF" has become. I also note that we hold, or try to hold, a lot of "side meetings" that don't fall anywhere on the "Bar BOF" - "pre BOF or pre Charter" spectrum. They include some directorate meetings, ADs meeting with WG Chairs, WG Chairs meeting with authors, various design team meetings, extra IAB, IESG or IAOC meetings, meeting related to parts of the RFC Editor function, and so on for what is probably a rather long list. Those meetings draw on the same resources as the types of meetings discussed in the document and those resources are not unlimited. Let me give two examples of the problem with the understanding that there are more: * If we have an official-looking, cast-in-RFC-stone, discussion of what "success" means for an informal meeting, it means that the IETF has established success criteria for a meeting type that we simultaneously claim is not part of the WG-formation or consensus processes. One really can't have it both ways and, in that sense, the document reinforces the "first you need a mailing list, then a Bar-BOF (sic) or two, then a "real" BOF (or two), ..." trend. * The draft says "Quiet restaurants are not hard to find and many offer private dining rooms..." (Section 3, paragraph 2). Well, that depends on the venue. We've certainly met in places where quiet restaurants _are_ hard to find (at least anywhere near the meeting location) and in places (not necessarily the same ones) in which the number of private dining rooms near the meeting location is not nearly sufficient to accommodate the number of side meetings (see extended definition above) that people want to hold. If anyone wants to add whining about how the IAOC didn't choose a venue with a sufficient nearby supply of quiet restaurants with private dining rooms to the regular cycles of complaints about venue choices, this document's recommendations would be a good place to start. Presumably we don't want to put the IESG or the Secretariat into the business or allocating, or refereeing, priorities for informal meeting spaces either... but, again, that is exactly the direction in which this draft seems to take us. Again, this doesn't make the draft any less useful as good advice; it merely argues against publishing it as an RFC, especially in the IETF Stream. If the IESG wants to stop the worst parts of the problem, it needs to verify community support and then put its collective foot down. IMO, any of the following would be of help: (1) Set an application cutoff for allocations of IETF space for pre-charter informal meetings before the cutoff for BOF requests, thereby eliminating one issue that does not appear to be mentioned in the I-D: use of Bar-BOFs by people who couldn't get their acts together to make a timely and satisfactory BOF request. (2) Just say "no". Once upon a time, we had a relatively strict rule against any IETF-related open (or semi-open, or publicly announced) meetings in the IETF venue unless those meetings were either on the agenda or a formal IETF group. That was in the time when Bar-BoFs (and Bar-Design-Teams, and Bar-other-things) were held in Bars, with Bar-sized groups, and those were not considered meetings of any type. We could do that again -- forcing pre-charter meetings into other venues would at least reinforce the message that they weren't WG meetings of unchartered WGs. (3) Get rid of the "two BOF" rule, replacing it with criteria that simply made it harder to get a BOF approved, and harder to get a WG proposal taken seriously, the more face to face meetings it took to get it organized. One of the causes of "WG in disguise"-style pre-charter meetings appears to be a combination of fear of hitting the two BOF limit and deciding to hold regular meetings despite not being able to get a WG charter or BOF proposal together that would satisfy the IESG. (4) Make it clear that any claims of consensus arising from an informal meeting would not only be ignored, but would cause BOF or WG proposals from the leadership of that information meeting to be discarded on the grounds that the people were clueless and incapable of working within the IETF. Taking these "unofficial BOF" discussions seriously merely reinforces their misuse. best, john _______________________________________________ Ietf mailing list Ietf@xxxxxxxx https://www.ietf.org/mailman/listinfo/ietf