[rrg] RIRs' choice to offer PI space

[Brian E Carpenter]

On 2008-11-27 07:20, Stephen Sprunk wrote:
> Iljitsch van Beijnum wrote:
>> On 25 nov 2008, at 0:40, William Herrin wrote:
>>> The RIR membership, composed primarily of network operators and
>>> engineers with a remarkably firm grasp on routing *operations*,
>>> perceived a critical failure in leadership by the IETF and reached
>>> consensus that the RIRs should step up and fill the void.
>>
>> Well, if they decide that PI in IPv6 doesn't pose any problems, why
>> are we still here?
> 
> It wasn't so much that PI didn't pose any problems, because it obviously
> does, but that _not_ having PI causes other problems that were/are
> perceived to be more serious.  If nobody adopts IPv6 because it doesn't
> meet their needs, it doesn't matter how scalable routing it is.  Lacking
> any IETF action to meet those needs, the RIRs took the only action
> available to them: PI.

Actually, the action open to them that they didn't take was to write
an I-D setting out in non-emotive language the needs they concluded
were unmet and the possible directions for a solution. Or to put
it another way, the amount of operator participation in the MULTI6
WG was probably less than ideal.

However, look at the resonance between current RRG discussion and
http://www.ietf.org/proceedings/03jul/slides/multi6-13.pdf

   Brian