Deciding between two choices

[Dave Crocker]

Margaret,

Thursday, December 19, 2002, 4:34:57 PM, you wrote:
>>When a working group:
>>
>>      1. Has a clear sense of what problem it is trying to solve
>>      2. Has a sense of urgency to produce something useful in a timely
>>      fashion
>>      3. Has a core of competent workers
>>      4. Has working group leadership (and an Area Director) able to manage
>>      the process with focus and timeliness -- ie, able to cut through
>>      distractions
>>
>>then no, we do not have a problem with making or sticking to decisions

Margaret> Which one of these factors was missing from the great SNMPv2
Margaret> wars?

I was not in the middle of that, so my understanding of its dynamics could
be fundamentally flawed.  But my sense is that #1 was the problem, in that
the very well-oiled core of experts split on the details of how to solve the
problem.  (Well, ok, that's a clear sense of solution, not just a clear
sense of problem.)

A different way of viewing it is that #3 failed, in that there developed
more than one strong core team.

Or, of course, we could go with Marshall's assessment that it was #4.  But
what is significant is that we are not finding ourselves needing to go
outside that list of factors.


Margaret> We somehow ended up with two useful, timely, competent proposals,
Margaret> and neither side was willing to compromise to reach consensus
Margaret> on the other approach.

Let's take your assessment as correct. I claim that the IETF should
virtually never try to resolve such differences. Publish both. If there is
any way to let the market make the choice, then we should.

For example, it worked quite well for SNMPv1 vs. CMIP.

d/
-- 
 Dave <mailto:dhc@dcrocker.net>
 Brandenburg InternetWorking <http://www.brandenburg.com>
 t +1.408.246.8253; f +1.408.850.1850