On Wed, 18 Aug 2004, Hallam-Baker, Phillip wrote:
By longstanding IETF precedent a charter only states the problems
that a WG MAY solve. There is no requirement that the eventual
solution MUST meet all the criteria that might be claimed to be in
the scope of the charter.
The IETF has a longstanding precedent of having running code that
demonstrates the validity of the solution. From the Mission
Statement:
Code is only required for progress from Proposed to Draft. In this
case we already have a lot of independent code bases that more than
meet the criteria that would be applied on the Proposed to draft
transition if they were conformant.
I suggest you look at RFC 2026, and tighten your terminology. A "draft" is
what we have now. A "draft" is worked on, edited, and finally moves to
"Proposed Standard". A "Proposed Standard" moves to "Draft Standard". It
does not move to "draft" ordinarilly.
4.1.2 Draft Standard
A specification from which at least two independent and interoperable
implementations from different code bases have been developed, and
for which sufficient successful operational experience has been
obtained, may be elevated to the "Draft Standard" level.
The multiple implementations and experience referred to in RFC 2026 is
different from the concept of obtaining _some_ running code to ensure that
a draft is complete and workable. Indeed, the mission statement was
drafted to make it clear that the IETF does not deal in hypothetical
protocols.
But we aren't moving from proposed standard to draft standard, we are
moving from draft to proposed standard, and I'm not calling for multiple
codebases or successful experience. Unless I miss the purpose of the last
call or the track this draft is on.
The point of several people posting is that there are many different
problems that need to be solved before this draft should be considered for
last call. _some_ testing and codebase should be available before a draft
is put to last call. Otherwise, we are dealing in hypotheticals.
Several days ago it was proposed that some testing be done before last
call. I thought there were many people in favor of that. Indeed, I would
have thought there was a consensus on that direction. Is that not the
case? So what is the rush? I'm still not understanding what the rush is.
--Dean