Just to clarify, by asking "So I guess I'm curious - what problems are
we having that cause people to think we need to make recalls easier, and
could those problems be addressed in less disruptive ways?" what I was
really wondering was whether the interest in fixing the recall process
was indicative of actual problems with certain individuals (presumably
toward remote participants) that aren't being addressed.
In other words, are people interested in trying to use the recall
process to fix problems that are better addressed in other ways?
I do agree that it should be fair to remote participants.
On 3/29/19 6:31 PM, S Moonesamy wrote:
Is an appeal disruptive? Is the first step of a recall disruptive? I
would not describe them as such. Those procedures are usually part of
a process so that the process is fair and viewed as fair.
I have no direct experience with recalls, but I do have experience with
appeals. They are, in my experience, tremendously time consuming.
They take away time and energy, that could otherwise be applied to doing
technical work, from people who are already over-burdened. Everyone
understands that they're necessary, and that it's necessary to take them
seriously, so that the process is fair and viewed as fair. That's part
of why those handling the appeal spend so much time and energy - people
handling them are aware that the trust of the community is at stake, so
they want to do a good job. Even though an appeal can significantly
impede progress of unrelated IETF work.
But having seen how much work of many busy and talented people are
required to handle an appeal, it's also easy to see how the appeals
process could effectively be used as a denial-of-service attack on
various parts of IETF. I expect the same could be said of the recall
process.
Of course, an appeal can be a threat in other ways. If a working group
has spent years polishing a document, only to see an appeal threaten to
make that work irrelevant, that's a different kind of disruption. A
recall can be a threat also, both to the reputation and career of the
person who is facing it, and also to the community if the recall
threatens to displace a good leader. (Sometimes there are differences of
opinion about that.)
So there is a need to strike a balance. It's reasonable to require
substantial community support for a recall, to demonstrate that several
people think there's a problem with someone before significant time and
energy are invested in reviewing the recall request. And it's
reasonable to require that the people lending support to such a recall
are actually real people, who have participated in IETF, who have
experience with the individual in question, again to thwart
denial-of-service attacks.
If all someone has to do to initiate a recall, is get some number of
people sign up for a meeting at ietf.org (paying no money), it's fairly
easy to find enough people out of the billions of people on the Internet
who are willing to do that. So the threat of denial-of-service attack
is quite credible. So I believe the bar should be higher than that,
while not inherently favoring people who can get to IETF in person. I'm
just quite sure what the bar should be.
Keith
