FWIW, with one qualification, I tend to agree with Brian. While
Scenario C seems to contain more detail and definition, a great
deal of it appears to be handwaving that tends to conceal more
complexity and more risk. I suspect that both schedules are too
optimistic (which one is "more" too optimistic is something I'll
try to generate a note exploring when I have a bit more time)
and that, while O is missing a risk analysis, as a subsequent
note from Brian points out, the risk analysis of C seems to
underestimate some of the risks. More on that in a bit too.
And, as you can guess from previous conversations we've had, I'm
not wild about the degree of control and involvement both
scenarios give the IETF standards leadership over financial
matters. More on that soon, too.
--On Tuesday, 21 September, 2004 12:04 +0200 Harald Tveit
Alvestrand <harald(_at_)alvestrand(_dot_)no> wrote:
I've seen some argument that Scenario C, being more
well-defined, is actually less complex than Scenario O.
Also, I was surprised to find that of the two timelines in the
writeups, the one for Scenario C was the shorter one. (That
may reflect the writers' degree of optimism, however!)
So the outcome is not settled in my mind yet.
I'd like your comments on why you find Scenario O simpler, and
VERY much like your comments on where the details of Scenario
O need "hacking", if that's the conclusion of the community.
--On 21. september 2004 11:10 +0200 Brian E Carpenter
Thanks, to you and the various authors, for this effort
and for reducing the choice to a binary one.
To me, this clarifies that it's a one-horse race. I just can't
see any argument for the extra complexity, overhead cost, and
risk of Scenario C.
Obviously we would need to hack at the details of Scenario O,
but for me there is no doubt that it's the way to go.
(My only regret is that Scenario O text doesn't include a
short risk analysis, like Sceanrio C.)
Ietf mailing list
Ietf mailing list