Re: [mpowr] Mailing List Management

On Mon, 29 Dec 2003, John C Klensin wrote:
>
> The bottom line, it seems to me, is that we need to function, as
> Dave Crocker has been pointing out (I hope I've correctly understood
> him), as a collaborative community.  No amount of rule-making is
> going to cause either a sense of collaboration or a sense of
> community to happen.

Agreed. Similarly, good (simple, clear, flexible, etc) rules do not
hurt and often help with collaboration we all want. Community
collaboration can exist in an anarchy and in a dictatorship, but is
more likely to exist and flourish somewhere in between those two
extremes.

> If we can't keep (or recover)  that sense, then we need a rather
> different set of rules and structures, to the point that debates
> about mailing list participation is a waste of time.  Within such a
> community, it seems quite rational to me to tell WG Chairs that they
> have lead responsibility for WG progress.

While the above is indeed perfectly rational, it is not the only
rational solution. That is what causes, in part, the debate in this
thread. There is a whole spectrum of rational solutions (all within
the collaborating community realm) with different proc and cons:

	- no banning whatsoever; use personal filters if needed
	- WG can ban
	- IESG can ban based on AD request based on Chair request
	- AD can ban based on Chair request
	- Chair can ban

Even if we manage to agree on the last scheme, there is still a
set of rational subrules:

	- N formal warnings are required (N >= 0)
	- minimum and maximum ban time is enforced (0 <= T <= oo)
	- repeated violators suffer more severe penalties or not
	- etc.

> If something or someone is interfering with that progress
> --including disruptions on mailing lists-- then they should be
> encouraged to stop it and, if necessary, pushed out of the way.

Agreed. The devil is in the details: exactly how interference is
defined, detected, and stopped. Again, there are many rational
solutions here.

> ... the right way to deal with problems of these sorts _within a
> community_ is to catch them earlier and push back (which, IMO,
> should include involving the whole WG in a "shunning" process when
> that seems useful), not to try to solve them by creating
> ever-more-rigid and complex procedures.

Still, there are enough flexible and simple alternatives people on
this list seem to argue about. See above for a rough outline of the
problem space.

> Fewer formal procedures, more exercise of good sense, more
> unambiguous negative reinforcement for failure to exercise good
> sense....

The above is an ideal that works fairly well for small communities of
similarly minded individuals. IMHO, the diversity, internal conflicts,
and visibility of the current IETF, make that ideal impractical.
"Good will" is not enough when "good" is essentially undefined and
subgroups interpret "good" differently.  You need a formal structure
with well-defined and "transparent" interfaces to encourage, educate,
and direct participants and their sponsors (if any) while resolving
conflicts.

The only working alternative to basic formalism is probably a
totalitarian dictatorship with IESG as superpower, dispensing pieces
of that power to individual ADs and Chairs. Simple, efficient,
flexible. Many of us would probably not want to live in that kind of
community though.

Formalism and interfaces do not have to be complex or rigid. They need
to give WGs as much control/freedom as possible while implementing
cross-WG quality checks and Oracle appeal process. The devil is, of
course, in the details.

Alex.

_______________________________________________
mpowr mailing list
mpowr@ietf.org
https://www1.ietf.org/mailman/listinfo/mpowr