Without having spoken to Ian, I'm more than happy to move forward
with protocol work, as long as there's sufficient diversity in the
input. A second protocol candidate is a great start at this.
Patrick's comment that such input is contingent on webi beginning
"useful protocol work with an expectation that the work would be
accepted" is on the mark; my impression of the lunch last week is
that we're more on the way to getting a decent cross-section of
implementors involved. It would be great if we could also get people
who will *create* invalidations involved (e.g., content
generation/management apps like Vignette, large Web sites, etc.), but
that may be too much to hope for, as they don't seem to participate
in the IETF too much (perhaps some of the rumored coordination and
liason work will help with this in the long term).
Before doing this, we'd need to determine the disposition of the
requirements docoument; we need to decide whether to continue working
on it, whether to publish it, and whether to use it as a yardstick to
judge candidates against. I'm not comfortable publishing it or using
it in its current state, but I agree that it doesn't seem likely that
it will mature much on its current course.
We were chartered to produce a requirements document. Good
requirements documents aren't busy-work, they assure that the WG
agrees what it is going to do, and gives a means of measuring its
output. It also helps scope the work. IMHO a requirements document
would be particularly useful for us, because some parties that will
implement and use RUP aren't represented well (as discussed above).
Since there doesn't appear to be interest in doing the requirements
document now, yet there still seems to be interest in our output, I
personally think a good way forward would be to keep working on it as
we concurrently start protocol development. It would be a living
document, not set in stone.
(Ian and I would need to discuss whether this would require
re-chartering; off the top of my head, I think it just involves
changing our milestones, which need to be updated anyway).
However, having more than one candidate (if indeed we do enter that
happy state) and no ready requirements document means that we'd need
to get consensus on the list about which should be used as the basis
of RUP. If we go down this path, this will be our next big hurdle.
Thoughts?
-- Mark Nottingham http://www.mnot.net/
This archive was generated by hypermail 2b29 : Thu Nov 18 2004 - 11:23:00 MST