On 2/17/12 11:59 AM, Noel Chiappa wrote:
> From: Pete Resnick<presnick(_at_)qualcomm(_dot_)com>
> We do need to make sure that the folks evaluating consensus know
> that "voting doesn't count" and that their decisions are made by
> consensus on the technical issues, not the number of people speaking.
Yes, but how do you tell where the consensus is if 97% of the people in the
'room' haven't expressed an opinion?
Condensing part of my unfinished essay to a few sentences: You decide
consensus based on open issues, not on number of voices. If folks have
brought up unanswered objections, there's not consensus yet (rough or
otherwise). If all objections have been answered (even if the answer is
simply a well-reasoned, "We understand that that is an issue, but for
these other reasons, we're not solving that problem", and there is not
significant objection to dismissing the issue), then the presumption is
that there is at least rough consensus.
If the 97% haven't expressed an opinion, you presume that they are not
filing objections and are therefore consenting. Consensus is all about
consent, not expressed agreement. Objection is the only way for there
not to be consensus.
The 'me too' posts do serve a purpose in
giving a larger sample size (provided, of course, that they are from long-time
Not to me. I don't see what they add.
Qualcomm Incorporated - Direct phone: (858)651-4478, Fax: (858)651-1102
Ietf mailing list