--On Sunday, August 10, 2014 11:25 +0100 Adrian Farrel
<adrian@xxxxxxxxxxxx> wrote:
>...
> The authors and shepherd are still working on synthesising the
> comments received during last call and preparing their answers.
> i don't believe this document is at the top of their list of
> priorities - it will not lead to shipping code that resolves
> to revenue!
>...
I had hoped to sit this one out, but the most recent discussion
(including the notes from John L. and Brian as well as this one)
suggests two observations.
First, to the degree to which this constitutes procedural rules
(as a BCP, it would seem to), it once again shows that
* We aren't very good at that
* There are always edge cases that get us tied in knots
or force us to waste time figuring out how to make
exceptions and/or drive square pegs through round holes
to allow something that common sense would consider
obvious.
* If something is important enough to justify revising
BCP 9 (RFC 2026 and its many updates) or BCP 25 (2418
and/or 3934), we are probably at the point where we
should just open and revise them, not incrementally
create more and more confusions about who has the
authority to do what.
In particular, any time a procedural document, or discussion of
it, heads the path of "do this, unless those circumstances
apply, in which case do that, unless these other circumstances
apply, in which case do that other thing, unless..." I think it
should be taken as a pretty good indication that the correct
rule should be more like "the WG Chairs and AD should, after
consulting each other for calibration and advice, use good
sense".
Second, this document is a mixture of significant procedural
changes and a tutorial on a particular role. The mixture seems
unwise. Language like "...include a number of additional
functions and responsibilities which are critical to the smooth
operation of IETF WGs..." could be construed as requiring that
every WG needs a secretary as well as at least two co-chairs.
Neither is true. Under the right circumstances, a secretary may
be a good idea and a huge help (just like one or more
co-chairs). The document provides good guidance about some of
the ways in which secretaries might be used when they are
appropriate. I don't read anything in 2418 as preventing those
roles (in addition to those it describes). If others read it
more narrowly, a discussion of IETF procedural flexibility and
the application of good sense might be a better treatment than
the attempt to compile a list that "is intended to be as
complete as possible".
Given the above, application of the same principles that led to
moving the Tao from the RFC Series to a living web page where
changes were subject to opportunities for IETF review and
subject to IESG approval would seem to make a lot more sense,
especially so because we really don't have a "WG Chair Handbook"
or an "AD Handbook" and it seems odd to be compiling what
amounts to a "WG Secretary's Handbook" in their absence.
The analogy to the Tao also suggests that the effort to compile
this document write things down, and get community review has
been worthwhile and constructive. What is less clear is
whether the result should be a BCP in the RFC Series or, like
the Tao, an evolving document.
best,
john
john