>>>>> "Harald" == Harald Tveit Alvestrand <harald@xxxxxxxxxxxxx> writes:
Harald> Sam, let me put it this way:
Harald> Changing the rules in the middle of the process is Just
Harald> Plain Stupid. We've done that too many times to count.
Hi. I've been perplexed by this comment.
I'd like to understand why changing the rules in the middle of a
process is a bad idea.
I can think of cases under which changing the rules might be problematic:
1) Many legal and process systems have as a part of fairness the idea
that actions that people should not be punished by actions that are
prohibited/illegal unless the actions are prohibited/illegal before
the action took place. In my own experience, the US constitution
has a prohibition against ex post facto laws--laws that make
something illegal after it happened. There are some fairly obvious
reasons why you want this particularly when the goal of punishment
is to discourage behavior from happening. Our process is such a
process so I agree we should not sanction people for actions that
were not sanctionable when the action took place.
2) It is undesirable to make a sanction significantly harsher than
initially expected during a process. If I expect that what I'm
doing is OK, but I might be wrong and get a warning I'll be very
surprised and annoyed if I end up getting kicked off the mailing
list. It would be unacceptable to change the process to allow
this to happen in the middle of the process.
3) Our standards process has an idea that we don't want to change the
bar too much while a standard is in progress and that we want
consistency. We don't want to apply one process to one standard
and a different process to a similar standard. We phase in new
requirements over time. If people believe they have met the
requirements for publication then it is frustrating to surprise
them with new requirements they could not have anticipated. This
is true even if the new requirements are approved through process
changes. Now clearly there is a balance here.
Note that reason 3 to avoid rules changes during a process does not
directly apply to PR actions. I don't want to view PR actions in the
same way as a standard. It should not be the case that if you collect
sufficient evidence you can get someone banned from a mailing list.
You have a right to expect that if you collect sufficient evidence of
an administrative problem like a problematic individual on a mailing
list, this problem will be solved in some way. You don't have a right
or expectation to demand a particular solution. If for example the
IESG successfully managed to convince the individual to clean up their
act, you don't have a right to be disappointed that a PR action was
not approved. (If the IESG claims they have convinced the individual
to clean up their act, you may well be dubious about whether this
claim is valid.)
In particular I'm having a hard time finding an ethical or logical
reason why we would not want to approve a process change that allows a
lesser sanction for behavior that is already prohibited.
Can you help me understand why that specifically would be a bad idea?
Now, there is one case where I can see a concern. If we are concerned
that the behavior may not be sanctionable today then what we are doing
might be problematic. We could make an explicit determination that
the behavior was currently prohibited before deciding to apply the
lesser sanction. Some people might question whether we could isolate
the two calls enough to make that decision.
So I agree that a solution open to less question is to refuse to apply
a sanction, create a process change and wait for prohibited behavior to
happen again. I'm concerned that in the case of Pr actions that may
be unfair to those trying to get work done. I could accept a decision
not to apply sanctions and to change the process though if the
community feels that is necessary.
I am concerned though that such a decision may lead to what I consider
to be a major problem. If we decide that we're going to apply a large
sanction because that is the only tool available to us, I believe we
have done something that violates the spirit of our process and that
does not meet the standard of fairness we hold ourselves to.
The fourth goal of the standards development process in RFC 2026 is
fairness:
The goals of the Internet Standards Process are:
o technical excellence;
o prior implementation and testing;
o clear, concise, and easily understood documentation;
o openness and fairness; and
o timeliness.
In particular, section 6.5.3 provides a path for resolving the
situation where we follow procedures that do not meet the standards of
fairness and openness to which we have obligated ourselves:
Further recourse is available only in cases in which the procedures
themselves (i.e., the procedures described in this document) are
claimed to be inadequate or insufficient to the protection of
the
rights of all parties in a fair and open Internet Standards
Process.
Claims on this basis may be made to the Internet Society
Board of
Trustees.
I think that if there is general agreement in the community that a
lesser sanction, were it available, would be adequate to solve a
problem, but we apply a greater sanction because that is the only
tool our process permits, there would be a claim for relief under
section 6.5.3 of RFC 2026. I hope that ISOC would grant relief in
such a situation. The IAB's response today gives me confidence that
such a claim might never reach ISOC.
So, if the community decides that we need to avoid a sanction in some
specific case so we can change the process, I can agree with that
decision. If we choose to apply a sanction we agree is too great
simply because it is the tool we have, I look forward to a successful
appeal of our foolishness.
--Sam
_______________________________________________
Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf