Here is my take on the facilitators experiment, from one of the volunteers
who tried doing that for 7 months. Did we make the IETF list better? Hard to
say. We have seen a couple big threads on the IETF list, and we tried
various ways to drive them towards productive outcomes. SM tended to use
formal responses "from the facilitators" through the list; I mostly used
private messages to the key participants in the thread; Carlos used both
methods. We probably helped things a little bit, but not very much. The
basic assessment is that this was probably not very useful, while requesting
quite a bit a bit of work. It also requires holding your keyboard, and not
participating in the debates that you have to facilitate.
There were limits placed on the facilitators role. We were supposed to not
intervene in matters like last calls, for which the AD in charge is supposed
to manage the debate. Then there are gray areas like discussion of IETF
tools for which there is a person in charge, but not-very-well-identified,
and there is a doubt whether facilitators should intervene. But if we
exclude last calls and tools, the scope of the facilitators is restricted to
threads about general policy like the Singapore debate, or thread about
recurring issues, like DMARC, RFC formats, and the like. As a community, we
want to have the debate. It can easily become loud and somewhat random, not
to mention repetitive. You wish someone would channel the energy and make
the debate more productive, but handling contentious issues as formal
facilitators is very hard.
For these type of threads, the best tool is the summarization message, as in
"this is the debate so far, please progress from there". The problem is that
such summarization works better if completed by an action list, "and here is
what we could do." The facilitators have no particular authority to put up
an action list, it has to come from the IESG, the RFC editor, or the IAOC
depending on the subject. Falling an action list, the summary can easily be
ignored.
The "facilitators" find themselves stuck between two failure modes:
concentrate on a polite process and overlap with the "sergeant at arms", or
try help solve the issue and overlap with the IESG. For example, during the
Singapore debate, I was very concerned that an attempt to steer the debate
will paint the facilitators as "a tool of the IETF establishment", trying to
smooth the path for the IESG and chair. But between the sergeant at arms and
the IESG there is only room for the occasional nudge, and that could be done
just as well by responsible individuals, without a facilitator title.
So, if you want my opinion, after trying this facilitator role, I would
rather keep things simple and not create another semi-official role in the
IETF.
-- Christian Huitema