Re: Fwd: FYI: Site Local

[Thomas Narten <narten@us.ibm.com>]

Rob,

Your analysis is right on the money. But there is a bit more as well,
I think. What happened in SF was that consensus developed during the
meeting, and the subsequent consensus call had a clear outcome. But
note that there was no consensus at all going into the meeting.

The problem for those not at the meeting is that they missed the
discussion. They are now being asked to ratify something for which
they have no context and no buy in.

So, we may have built consensus among those in the room, but that has
not happened for those only on the list.

Thomas

Rob Austein <sra@hactrn.net> writes:

> I think part of the problem is the distinction between

> a) confirming whether an already determined consensus is correct, and

> b) determing whether a consensus exists in the first place.

> Mailing lists are ok for (a), much harder for (b).

> That is: if you already think you know what the consensus is, it's
> often possible to use the mailing list to confirm what you think you
> already know, and the answer you get back is either "yep, that's the
> consensus", or "no, that's not the consensus".  The test here is
> fairly straightforward: if you get back a a nontrivial number of
> objections (or a bunch of new issues), you don't have consensus.

> If, on the other hand, you're trying to establish whether a consensus
> exists, the mailing list is only a useful tool if everybody is willing
> to play, and to play nicely at that.  Refusal to play nicely on the
> mailing list is one of our ongoing problems (and that, in turn, stems
> at least in part from a lack of shared understanding of what "play
> nicely" would mean).

> Finally [Rob's standard rant on prerequesites for consensus]: no
> consensus process can cope with significant numbers of participants
> who hold opposing entrenched positions that they are not willing to
> change come hell or high water.  The process requires the participants
> to have a shared goal of reaching a group answer than that is more
> important to them than having their own favorite win.  Without that
> shared goal, what you're left with is (at best) some form of voting.