As is moderately obvious from the stream of commentary on this
thread and there companions, there is no *one* problem at
the root of all this. One way to draw this is:
Issue: Documents are too slow in achieving the first rung of the
standards process
Contributing issues:
--------->WG formation is slow, as there are now often 2 BoFs before work begins
--------->Working group activity is slow, as it pulses to physical meetings
--------->Late surprises arrive from late cross-area review (often
from teams) or the IESG
--------------->Because there is little early cross-area review after a BoF
Resulting issues:
------->Little energy remains in working groups to advance documents
once they do complete
-------->The IESG sees that few documents get early re-review as part
of advancement, and
raises the document quality requirement for the first
rung to prevent impact on the
rest of the ecosystem
Results of the results:
------->Things get slower
------->More work is done outside the IETF and brought in only to be blessed
------->More of the internet-runs on Internet Drafts.
Results of the results of the results:
--------->It's harder to tell which documents are actually the ones
you need, both because
some actual Standards documents are obsoleted by drafts
and because some sets
of drafts have functional consensus and some don't.
Results of the results of the results:
---------->ADs and others want more tutorial data added to the RFCs,
which makes
producing them slower.
Try to find one place to tug on this and the actual results of your
tugging won't
really be seen until a full document cycle, and there will be odd
states in between.
That causes debate and discussion, and worry with all the nice people
we've tasked
to worry about these things (and many others besides). That burns energy that
could be going into working groups and, well, you get the picture
(things get slower).
Should we do nothing? No. But we need to accept that no single thing we do
is going to solve all of the problems. Changing the document labeling
will not increase
early cross-area review. But if the top-line issue is "Documents achieving the
first rung of the standards track do so too slowly" we may have to tackle
it, the WG creation problem, and the WG "pulsed activity" problem at once
to make real progress.
If the problem we want to tackle is "The first rung is set too high", then there
are other possibilities (including simply recognizing that the first rung is
really WG draft, marked or unmarked as it may be). I personally don't
think that's quite enough, as the value of the IETF (as opposed to its
working groups) is that it can and does cross-WG and area review. But I
see the attraction--if the first rung goes lower, the documents may be produced
faster, which can mean that there is enough energy to go up the track plus
the cross-area review is functionally earlier. My worry (yes, I worry)
is that if we
re-use a label for the first rung after lowering its bar, we create a
confusion that we
can't easily solve *especially if the energy does not magically appear*.
As we stare down this rathole one more time, let's at least be certain
that there is more than one rat down there, and be realistic about the
energy we have on how many we can tackle. Russ's draft tries to
do two things:
Restore the 2026 rules for Proposed as the functionally in-use bar for the
first rung.
Reduce the bar for Standard to the old bar for Draft.
Listening to the discussion, I think we have focused a great
deal on point two, but have either not really noticed point
one or didn't believe it. I think this addresses a marketing problem
(long an issue, though now commonly explained away) and it
focuses on the first two "resulting issues" in the quasi-chart above, and
thus may have some cascade effects on the other two. It doesn't
tackle any of the contributing issues, but this is not really a defect
in my eyes, as those can't really be addressed by document issues.
Are there other ways to tackle this? Sure. But if the community
accepts that this restores the 2026 bar for the first rung *and holds the IESG
to it*, then I think this is one useful place to tug on the tangle of issues.
Just my two cents,
regards,
Ted
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf