So from participants, chairs, IESG - are there any lessons to be learned
here that should apply to other WG so that other WG don't have process
failures like this one ?
There's a veritable process failure epidemic going on. It seems to me that this specific sort of problem is inevitable from time to time and that it's not sufficiently pervasive to think that this (where "this" refers to a difference of opinion as to whether or not an agreement has been reached) is not a spot problem. We do tend to have a problem with people who simply will not accept "no" for an answer but I don't think this is one of those cases.
Although, come to think of it, to the extent that this represents a broad category of process failure it would be around our inability to do consensus well, whether it's as simple as learning how to ask questions better or publishing the decisions that have been reached, or as subtle as how we identify decisions (this aging hippie looks at "80% of the people in the room said 'yes' so we've got rough consensus" and thinks "oy vey") and actively try to create consensus. These issues are under discussion, etc., elsewhere (the problem working group) and approaches to solving them are, as well (the "solutions" mailing list), and it may be worthwhile to take a look at the documents and mailing list archives to see if you feel there's more we could be doing around this specific problem area.
Melinda