[rfc-i] On two committees

Brian E Carpenter brian.e.carpenter at gmail.com
Mon Nov 29 12:42:03 PST 2010

On 2010-11-30 08:48, Russ Housley wrote:
> Brian:
>> In brief:
>> In the model I have in mind, the RSE could choose
>> to appoint an advisory committee to advise him or her,
>> entirely at the RSE's discretion. Actually I reckon that
>> anybody, doing any job, can do this. No need for BCP text
>> about this.
>> The other part of my model is to introduce community-based
>> oversight in some form, just as we did when reorganising
>> IETF administration, which is why I suggested an Oversight
>> Committee.
>> My not-hidden concern here is to ensure that there *is*
>> oversight while offering the IAB the chance to get out
>> of the details. Note, I am *not* criticising the IAB for
>> stepping up to the mark over the last couple of years;
>> somebody had to. But isn't it odd for a group with
>> "architecture" in its name to be responsible for documents
>> such as RFC 5741?
>> All details TBD of course, but that is my top level concern.
> I agree that there needs to be community-accountable oversight of the
> RSE, and I agree that the IAB does not want to perform that role.
> When I started reading this section, I imagined a relationship similar
> to the IAOC oversight of the IAD.  That is, policy decisions are make by
> the IAOC, and then the policy is implemented by the IAD.  I like this
> model because the community-accountable body is making the policy decisions.
> I do not think the current document has this clarity.  Also, your
> earlier message about the Transition Team further confused me.  Maybe it
> is just a matter of finding the right words.  If this is your intent,
> then I support it.

Yes, the words I sent were drafted very quickly under the influence of jet
lag. Basically what you say above is what I'd like to see, although (as
in the case of the IAD) we have to be careful to leave enough responsibility
in the job description to make the job attractive.


More information about the rfc-interest mailing list