[Top] [All Lists]

Re: Scenario C prerequisites

2004-09-22 06:19:14

Harald opines:
re: 1/ Considering the level of participation in this discussion on the
IETF list I do not see how one could assert that there was IETF
consensus without an explicit discussion at an IETF plenary - I do not
think that just issuing a last call (as envisioned by the Scenario C
document) would be seen by about anyone as an adequate involvement of
the community.

I am not at all certain of that. In what way is 20 people arguing in front 
of a thousand people in a room more "community involvement" than the same 
20 people arguing in front of a thousand people on a mailing list?

because this way you are sure that the thousand people at least know
what is going on and have a chance to express their view if they think
things are going wrong (in person or on the list during last call)

Our tradition as IETF has been to declare that mailing list discussion is 
the final arbiter of consensus. If we need to abandon that principle for 
organizational matters, we leave ourselves in a situation where we can only 
make significant decisions at 4-month intervals; that is a theoretically 
defensible position, but sharply limits the scope of what we can hope to 
accomplish in any given timeframe.

I trust we will not be reorganizing at this level all that often - I
think that adding or subtracting an area does not need a plenary session
but changing the legal basis of the endeavor is a rather important
step, one that should not be done on the basis of 10 people expressing
their opinion on a list.

i.e., I think some things are important enough to require the full
measure of due diligence - this level of reorganization is an example as
was the ravin discussion - waiting until the next face to face meeting
for a public discussion does not seem to be too big a issue when we 
are talking about a change that (historically) comes around every 
18 years.


Ietf mailing list

<Prev in Thread] Current Thread [Next in Thread>
  • Re: Scenario C prerequisites, scott bradner <=