On Tue, Jun 11, 2013 at 11:20 PM, Ted Lemon
<Ted(_dot_)Lemon(_at_)nominum(_dot_)com> wrote:
On Jun 11, 2013, at 6:03 PM, Dave Cridland <dave(_at_)cridland(_dot_)net>
wrote:
... and how would we judge IETF consensus on a document that doesn't get
done under a charter (which would in turn have been granted consensus
without any IETF comments?)
I would expect that you'd start with a mailing list, see if there is
interest, come up with a proposal for a BOF, come up with a working group
charter, get IESG review on the charter, then get IETF consensus on the
charter, and then start working on the document. That's how it's usually
done.
I think you misread entirely.
But anyway, let's assume we follow those steps.
I suspect the closest we get to getting an idea of IETF consensus is the
interest gauging at the beginning of the process, though interestingly this
is only positive interest - objections to doing the work at all aren't
really relevant here. The IETF consensus on the charter is handled by
Apathy Is Assent rules, so claiming that this consensus call becomes the
default is an interesting argument to make.
BTW, the fact that a few people think the process ought to work
differently does not mean there is consensus for it to work differently.
Also, what there may not be consensus on among the people who have weighed
in on the topic is whether positive statements in favor of a document are
relevant in IETF last call, but I don't really know how to reduce that to
practice, because in reality I think it is rare for a quorum of IETF
participants to read a document as a consequence of a last call
announcement. Without that, I don't see how you can have any other last
call process than the one we currently have.
OK, so we don't have voting but we do have a quorum? How wonderful. And
given that the majority of people are silent means they agree (with, mind,
everything - not just that they're not reading), so any number of people
are therefore "few".
But to loop around:
I strongly feel that positive statements have value, as they allow the
community to gauge the level of review and consensus, and I suspect that
human nature means that we get more reviews if people get to brag about it.
I suggest that if more than one bit of data is required, it's simply asked
for. Given that the text of IETF Last Call announcements is not governed by
any process RFC that I'm aware of (feel free to correct), I suggest simply
putting a set of optional questions there. I note this practise has served
the XSF very well. I do not think this needs an endless bikeshed discussion
on what questions; the IESG can pick what it wants to know.
If, on the other hand, only objections are sought, then the text (which
simply asks for "comments") also needs changing. And the GenArt, AppsDir,
and SecDir reviews should only be send when they have objections to
publication, of course.
If you feel that the only way to make either change is to form a working
group and publish an RFC to change something undocumented in the series,
then I think we're stranded in a bureaucratic quagmire with no chance of
escape, but I'll be happy to send "comments", as requested, nonetheless.