Re: [dnsext] afasterinternet.com trial and draft-vandergaast-edns-client-subnet-00

[Ted Hardie <[e-mail]>]

On Wed, Aug 31, 2011 at 7:25 AM, Andrew Sullivan <[e-mail]> wrote:
> On Wed, Aug 31, 2011 at 02:43:59PM +0100, Wilmer van der Gaast wrote:
>
>> Which draft are you reading here exactly?
>
>>
>> For a new option code, only expert review will be required: "Expert
>> Review is required for allocation of an EDNS Option Code."
>
> Oh, phew, and stupid me.  I was really really surprised when I
> (thought I) saw what I did when I checked it earlier.  That'll teach
> me to do such readings in a hurry.
>
> You're quite right.  And (again no hat) expert review is, IMO, better
> than RFC required, but I'd still prefer FCFS.
>
> A
>

I think FCFS works fine for unbounded spaces, but I'm nervous about
a blended model where there are easy registrations up until exhaustion
looks likely, then
it gets restricted.  We have lots of evidence that this causes grumpiness
in the IP allocation side of the house. Expert review, with clear
instructions to reviewer
(e.g. that this is not duplicate functionality) seems more likely to
get what you want.

The trick is in the clear instructions to the reviewer, especially
when the reviewer
is actually a list with changing membership....

regards,

TEd