--On Monday, January 11, 2010 10:21 +0100 Shane Kerr
<shane(_at_)isc(_dot_)org> wrote:
John,
On Wed, 2010-01-06 at 17:13 -0500, John C Klensin wrote:
I am extremely concerned about getting into a situation in
which the IETF spends time debating issues that are basically
minutiae, designing (or fine-tuning) procedures or naming
schemes in a committee of a few thousand.
*Getting* into a situation? The IETF spends a great deal of
its time in such activities, and has at least for the last 10
years or so (there may have been a "golden age" before that,
although it seems unlikely to me).
Well, yes. I should perhaps have said "open up yet another set
of opportunities".
I think the reality is that there are certain technical issues
that:
1. Are simple enough that everyone can understand them
s/can understand/believes that they understand/
2. Have no completely obvious solution
s/$/ and, even if they do, that obvious solution, while
consistent with general understanding, raises complications or
issues that are not so easy to understand/
These issues invite long, pointless threads discussing them.
Things that require a lot of time to understand the details
tend to only have a small handful of experts who discuss them,
because there aren't that many people who have the time and/or
motivation to get up to speed.
Things that have a clear solution tend to only have a few net
kooks who argue (loudly, repeatedly, rudely) against this
solution.
My modifications above are intended to suggest that there is a
third case: a simple approach that seems to be a "clear
solution" to those who actually don't understand the issues (but
think they do) and that appears to raise other issues (if not be
outright wrong) to those who think they have a more in-depth
understanding of the issues and possible side-effects. Where
we get into trouble isn't just the ranting, it is the
fundamental disconnect about what constitutes an adequate
understanding of the problem.
For this case, the community has traditionally given IANA a lot
of discretion about selecting particular names and numbers.
There are some important reasons for that, as well as just
avoiding noise. Moving into a situation in which we require
IETF consensus for name selection or alteration is actually a
fairly big deal (having IANA consult the community to see if
anyone sees bad side effects is another matter entirely and both
reasonable and precedented).
I think a number of process and other efforts have been made
over the decades to try to improve the signal-to-noise ratio
of IETF discussions. I don't know how successful these have
been - perhaps the IAB has metrics on this? I have no further
proposals, except perhaps less wringing of hands over the fact
that we seemly waste a lot of time arguing over
not-so-important stuff.
Let's save that discussion for another time, but I think it
would be safe to say that, typically, the community waits until
there is actually a crisis before making any significant changes
and then tends to get in a hurry and apply partial solutions
that end up causing new sets of problems.
Oh, and we might also remind people who have drafts caught up
in a vortex of endless nitpicking that it's not their fault,
and there is nothing wrong with their drafts. (Good luck Joe!)
:)
right.
john
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf