On 6/15/24 01:18, Brian E Carpenter wrote:
(Switching lists as we are off the Last Call topic)
For what it's worth I thought posting my responses to the Last-Call list
was probably less likely to contribute to the disruption than posting
them to the IETF list.
On 15-Jun-24 16:49, Keith Moore wrote:
On 6/15/24 00:30, Adam Roach wrote:
For avoidance of doubt: would it be fair to infer from the second
half of your message (the portion below the dividing line) that you
would decline to support /any/ PR-Action based on the process
described by BCP 83?
no. I haven't tried to do such an evaluation, and I can think of at
least hypothetical cases in which I'd support use of BCP 83, at least
in an emergency.
A four-week Last Call can hardly be called an emergency solution. An
emergency would surely be somebody sending, say, grossly illegal
content. I think (and certainly hope) that people with the appropriate
admin passwords would take immediate action on that, regardless of any
BCPs.
To me an example of an emergency would be an overt threat of harm,
perhaps even reputational harm, especially one with immediate effect.
But I agree that a 4-week last call would be too slow for such an occasion.
But I think we have enough experience with BCP 83 by now to see its
considerable downsides,
I'd like to see an impartial summary of all PR-actions to date, to
form an opinion about that.
How could an impartial summary even exist? Everyone projects their own
experiences and ideas onto such things.
and I find much of the community's response to BCP 83 PR-actions
(including this one) to be unprofessional and disruptive to IETF's
purpose.
Disruptive, yes; ironically enough, this (the Last Call) part of the
cure is worse than the illness.
Unprofessional? I'm not sure (with one or two exceptions).
It may be that I have associations with the word "professional" that
others don't share. For me, "professional" would include an effort to
be fair, for charges to be precisely determined in advance (so that
those defending themselves don't have to answer to new charges that
their accusers or others might have added), for the infractions to be
well-defined in advance of the actions the person is accused of (nothing
so vague as "unprofessional"), etc.
More generally I don't think a "rough consensus" process similar to
that we use to approve protocols, is an appropriate way to consider
punitive actions.
It isn't intended to be "punitive". It's meant to protect the IETF. If
it also changes the behaviour pattern of the individual concerned,
that's a win too.
I doubt that it has that effect. I suspect what it does far more often
is to drive potential contributors away, whether those contributors
empathize with the accused, with the accusers, or both.
People can contribute to consensus to support a protocol for
their own reasons; they don't need to supply any clear or precise
justification, and I'd argue that that's a feature. But when
considering punitive actions it seems to encourage everyone in the
self-appointed "jury" to come up with their own charges, their own
evidence, their own rationale to support the punishment. It's
basically a witch trial, or a mob trial, in which all accusations (no
matter how poorly founded) are considered to be valid without
support, and there's peer pressure within the mob to start
(metaphorically) throwing stones.
I really don't think that's right. In this particular case, people
have been adding instances, and that was necessary since the IESG
message really didn't establish a pattern. But it's a well-defined
process with an end point.
That the IESG message didn't really argue its case, and cited only a
single brief message as justification, is part of the problem with this
PR-action. In general it seems like the action was hurried. But I also
don't think that a person so accused should have to answer for messages
they've sent arbitrarily long ago. People can and sometimes do improve
their behavior over time, for various reasons. I think BCP 83 was
intended to deal with an ongoing pattern of disruption, not acts in the
distant past. And it's a stretch to claim that a single simple message
with ambiguous intent, however obnoxious, is indicative of an ongoing
pattern of disruption.
Keith
