On Tue, Feb 10, 2009 at 08:41:50AM -0500, Melinda Shore wrote:
Under classical consensus decision-making there's
a prerequisite that the participants have some
investment in the process itself and that they
actively participate. Drive by "I'm against it!"
posts almost certainly don't qualify as
participation - there's absolutely no opportunity
there for negotiation and compromise.
Sure. But under such classical consensus decision-making, one knows
who's in "the group" for the consensus. The IETF doesn't, because the
answer to "Who's in the group?" is supposed to be "Who replied on the
The roughness in consensus partly comes from the person who evaluates
consensus deciding that certain replies on the list, if opposed,
perhaps don't carry as much weight.
There's no question that this sort of process lends itself to nasty
attacks by people who aren't invested in the IETF culture. One way to
solve that, of course, is to give up on the way we do things. Another
way is to tolerate the occasional attacks, on the not implausible
grounds that most such attackers(*) will go away eventually. I think
what is not a good idea, however, is to decide in advance whose
opinion counts. In my view, that would merely lead to "comfortable
consensus" rather than rough consensus: only the opinion of those who
already agree will ever be considered as relevant. (There are those
who suggest we've already arrived at that eventuality, but I don't
think we have, and I'd like it not to happen.)
(*) I'm sure some of us can think of a counter-example or two.
Ietf mailing list