[standards-jig] Last Call standards

Peter Saint-Andre stpeter at jabber.org
Fri Dec 13 16:41:22 UTC 2002


On Fri, 13 Dec 2002, Dave Smith wrote:

> DTCP is at Last Call stage because a couple of people proposed it as 
> such. That does _not_ mean it _should_ be at Last Call stage -- indeed, 
> I would argue that the barrier to "Last Call" should be significantly 
> higher. 

IMHO the old system (essentially a JEP Editor autocracy) was not broken,
but we fixed it anyway because we wanted to make sure that there was some
level of community interest before proceeding to Last Call. Now, however,
we've lost the previous check of the JEP actually being ready for prime
time (which was negotiated between the JEP Editor and the author, with the
JEP Editor often consulting with Council members).

During the tenure of the first Council we didn't have the Last Call
process, but there was an effort to get the Council members more involved
earlier on to make sure that a proposal was sound before proceeding (e.g.,
Dave Smith proposed that a Council member write a "report" on each JEP
before it could given a number and then published). That idea never took
off, but I do think it would be good to get some feedback from the Council
before moving to Last Call.

One possible solution is that only Council members could propose the Last
Call. This would force Council members to read and review the JEP before
it goes to Last Call, since in essence the proposing Council member would
act as a sponsor for the JEP. I don't think Council members would do that
without first consulting with other Council members, because their
technical reputation would be on the line. This would also force JEP
authors to communicate with Council members earlier in the process (IMHO
such communication has been lacking). Once a JEP was proposed for Last
Call by a Council member, 5% of the JSF membership would still need to
second the proposal as we are doing now (thus retaining the "democratic"
aspect of consent from the community). I feel that this would introduce
some checks and balances into the process -- it's not autocratic as in the
past (much as I yearn for the good old days of absolute power :) nor fully
democratic as we have now, but a good mix of both (the Council is like the
Senate and the JSF members are like the House of Representatives).

A system such as this would be my strong preference going forward.

Peter

--
Peter Saint-Andre
Jabber Software Foundation
http://www.jabber.org/people/stpeter.php




More information about the Standards mailing list