Just piling in after Barry here ... and speaking as the AD who had the most recent ¨draft on a telechat agenda with consensus = unknown¨, but I'm hardly the only one.
The field in the tracker is labeled ¨consensus¨, which is ambiguous but actually means ¨IETF consensus¨. So, it should be set after IETF Last Call.
A fair number of the documents I've processed already had it set to ¨yes¨ when they were publication-requested, so that means the shepherd/working group chairs thought it meant ¨working group consensus¨.
I believe - but Barry would know - that we've requested that the field label be changed to ¨IETF Consensus¨ in the datatracker.
(I know we talked about that, but I don't know whether we've made the request yet)
Spencer
On Tue, Oct 7, 2014 at 11:47 AM, Barry Leiba <barryleiba@xxxxxxxxxxxx> wrote:
> It's not true that no consensus is needed for a document just because it's
> not a WG product. Anything that comes through the IETF stream (including
> AD-sponsored documents) need to reflect consensus.
That's generally true the vast majority of the time, though there are
exceptions for some Experimental or Informational documents, which is
why the flag is there in the first place.
We do occasionally produce documents that describe proprietary
protocols or that republish outside documents in the IETF stream. We
try to do those in the Independent stream instead, but it's not always
the best or right thing. When they're published in the IETF stream,
the point is that we have consensus to publish them, but we might not
have consensus on the protocol that's described. In those cases,
we'll use "No" for the "Consensus" flag.
Barry