From: Theodore Ts'o [mailto:tytso(_at_)mit(_dot_)edu]
Ted-
Sorry for taking so long to respond - I wanted to give some thought to your
questions before replying (comments in-line)
On Tue, Sep 27, 2005 at 06:47:36PM -0700, Nick Staff wrote:
2. An IETF "netiquette" committee, to offload list banning
procedures from the IESG.
I'm a big fan of the netiquette committee. I'd like to
suggest that
volunteers be allowed to "throw their names into the hat" and that
members be selected blindly from that pool. This would of course
avoid any stacking or favoritism, but we would need a
"qualifier" that
prevented interlopers from submitting their name. Though I hate to
suggest it as it would exclude me from selection, having
attended an
IETF meeting in the last x years could possibly be a good filter.
Maybe. I see two potential problems:
1) Serving on this committee is going to be no fun at all.
Getting qualified people to sign up for what will only be
seen as a sh*t job is going to be difficult.
I figure if Brian was able to get multiple volunteers for the IESG scribe
position (of which I was one), then this should be a cakewalk ;)
And how do you
exclude certain known
(repeat) troublemakers from throwing their hat into the ring?
Or maybe you don't, but then if they get selected, they
would then have the opportunity to practice their own unique
form of DOS on the netiquette committee?
Here are some general design points I've been thinking about to help prevent
the DOS you speak of as well as some other pitfalls:
1. 7 or 9 member committee
2. Members selected blindly from pool of volunteers
*Let's not forget that no matter who you are, there is someone out there who
thinks you're a troublemaker, that you're dumb, mean, etc. This is why it's
open to all volunteers, to prevent the tainting of the committee and the
stacking towards one point of view.*
3. Majority can close discussion and force vote
3a. Unanimous minority can stay vote for max of 2 days
4. Verdicts are made up of 2 separate votes
4a. In the first vote, the committee members vote whether to sustain or
refute the petitioners claim.
4b. In the second vote (which immediately follows the first) the members
vote on the punishment. One of the choices MUST always be to issue a
warning. The other choices will vary depending on the petition.
4bb. Anyone who is issued 3 warnings in less that a years time, on
subsequent "punishment" votes there MUST NOT be the choice to issue a
warning. This will be for a period of 1 year beginning on the day their
third warning was issued.
4c. Note that when a petition is sustained the committee votes on a
PUNISHMENT FOR THE ACCUSED, and when a petition is refuted the committee
votes on a PUNISHMENT FOR THE ACCUSER. This should help curtail frivolity.
5. Any sentence suspending someone's posting rights due to abusive/off-topic
posts is required to pass with no greater than 1 dissenter. This is to
enforce the idea that if there can be sensible disagreement about whether a
post's off-topic, then it's too subjective for such a serious punishment.
5a. When 2 voting choices differ only on length of time, then their votes
may be added together to reach the needed majority - however in those cases
the shorter of the two sentences MUST be imposed. For example if 6 members
vote for a 1 year ban and 2 vote for 30 days (with 1 voting for a warning)
then even though there is not sufficient majority for a ban, the six votes
and the two votes can be added together which means the ban will pass -
however it can only be a 30 day ban and can never be the greater of the two.
6. In all cases the dissenting minority is allowed to publish their
dissention along-side the majority verdict (in fact, one MUST NOT ever be
stored, displayed, or considered without the other.
2) Unless discussion of the decisions of the netiquette
committee, during the committee is considering a request, and
after the committee has rendered a decision, is ruled out of
scope, it's not going to help the very long discussions such
as this one which plague the IETF list.
In the worst case, we can assume that the mailing list abuser
will immediately appeal any decision of the netiquette
committee, which means that after inventing this entire
mechanism, it may not have any effect other than prolonging the agony.
I know personally, if I feel a process is fair, then even if I hate the
decision I can accept it and move on. That's another reason why I think it
should be an unmanipulated membership.
I also think the dissenting opinion will help here. Sometimes just hearing
someone agree with you is enough to calm the whole situation down and give
someone a sense of justice or understanding - even if the majority verdict
is against them.
thanks,
Nick
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf