sob(_at_)harvard(_dot_)edu (Scott Bradner) writes:
But we *often* take straw polls in f2f meetings,
but we do not count hands - we look to see if there is a clear
difference between hands one way and or the other
I agree that this is exactly how we should be using hums/polls.
But I'm sure many of us have also seen cases where a consensus call is
made and it is not immediately/obviously clear (to all) just how clear
the difference between the two sides really is. Or to someone outside
the room who wasn't there.
Having hand counts makes it possible to have a more honest discussion
(both during and after the fact) about just how close the call really
is, how much participation there was, etc.
If folk are unhappy about a particular call, but they are not really
able to challenge it, it can reinforce the perception (rightly or
wrongly) that the IETF processes are unfair.
But as others also note, there are also downsides with "voting". And
there have been cases where "packing the vote" have taken place. The
most obvious example I recall was in the case of IDN WG. It is exactly
in these cases that we need to have good/strong chairs that recognize
when this is happening and factor it into the consensus call.
Ietf mailing list