Re: [apps-discuss] Updating the status of SPF

[Scott Kitterman <[e-mail]>]

On Friday, August 12, 2011 12:19:03 PM Andrew Sullivan wrote:
> On Fri, Aug 12, 2011 at 11:38:57AM -0400, Scott Kitterman wrote:
> > My view is that the experiment was largely successful and only minor
> > changes based on this experience are necessary.
> 
> This is the problem I'm focussed on, however: once you say, "My
> changes are minor, $someone's are not in scope," you basically open up
> the protocol by having to debate the values of "my" and "minor" (and,
> too often I find, "changes").
> 
> I really don't have a dog in this race: I don't care about SPF, and
> the TXT battle was lost a long time ago.  I'm just saying that, if the
> goal really is supposed to be, "Just document what's happened," you
> don't need a WG and you don't need to move tracks.  You need a
> document that outlines what happened, period.  That will more readily
> achieve the goal of documenting actual existing practice and getting
> the document out in a reasonable time frame.  Everything else
> effectively means that you have to deal with all the issues Dave
> Crocker correctly identified.
> 
> At the same time, I am strongly opposed, on process grounds, to a WG
> whose task it is to move a protocol from the experimental track to the
> standards track without facing any of the questions about whether the
> experiment worked, the extent of that working, and the trade-offs that
> might have been acceptable in an experiment but that are not
> acceptable for anything we might call a standard.  If there is to be
> any distinction at all between these tracks, then those questions must
> be acceptable when changing.  Otherwise, we get an experimental track
> that is just a new level in the 3 (soon to be (3-n)) level standards
> track we already have.

I see three possible conclusions a working group might reach:

1.  The experiment was successful and with minor adjustments the protocol is 
suitable for standards track.

2.  The experiment had some good points, but identified significant issues that 
require substantial change, but we can make them and move on to the standards 
track.

3.  The experiment is not yet mature and after revision we need to repeat the 
experiment some more as an update experimental protocoal.

The evidence I'm aware of points to number one as the most likely result of 
doing this work.  I would prefer an initial charter that is oriented on this 
path to minimize the amount of time that needs to be spent on bikeshed 
discussions.

That does not mean that I think the WG should be blind to data that points to 
outcomes two or three.  If that's where the rough consensus goes, then that's 
where it goes.  In the interests of being efficient about the most likely path 
I'd rather come back and recharter for that type of work if it proves to be 
necessary.

Scott K