----- Original Message ----- From: "John C Klensin" <[email protected]> To: "t.p." <[email protected]> Cc: "ietf" <[email protected]> Sent: Thursday, January 31, 2013 2:44 PM > > --On Thursday, January 31, 2013 12:04 +0000 "t.p." > <[email protected]> wrote: > > > Time to publication of an RFC can be reduced by cutting out > > the pauses, which could achieved by highlighting when they > > occur. > > > > This can be done by a tool which, for every active Working > > Group, runs monthly and, for every draft adopted by the > > Working Group, records whether or not there has been a change > > and posts this list as an e-mail to the list of the Working > > Group. Where a new version has been submitted to Tools, then > > this new version is listed with the date of submission. Where > > the status has changed, as recorded in the tracker (AD Review, > > IETF LC, Publication Approved etc), then the new status is > > listed with the date of change. Where nothing has changed, > > then this is listed with the date of the last change. > >... > > Tom, > > This is interesting but could also introduce a pathology in > which drafts are generated too frequently to encourage (or even > permit) healthy discussion. If one were going to do this, also > collecting some statistics on how much (or whether) a given > draft was being discussed on a WG mailing list might be very > important. New drafts and indications of motion like status > changes may be good clues but what really counts for measures of > progress and consensus is whether real discussion is going on.
Yes, drafts might be produced more often just to 'game' the system but we have something like that anyway. I often see drafts produced with no changes simply because the previous version has expired. Usually the WG Chair or editor posts a note to that effect; if not, someone on the list may ask and so the reason emerges. Yes, a measure of discussion would be better, that might be the gold standard, whereas my idea is only silver or bronze. But I don't know how to measure discussion in an automated way, one that places no additional burden on such as WG Chairs or ADs. So I see this as a step in the right direction upon which others might build. Tom Petch > john >
