Re: [address-policy-wg] 2005-12 New Policy Proposal (4-Byte AS Number Policy Proposal)

[Geoff Huston <]

At 02:11 AM 29/12/2005, Joćo Damas wrote:

> On 26 Dec, 2005, at 21:28, Geoff Huston wrote:
>
> > b) that the proposal replace the term "2 byte" with "16 bit" and "4
> > byte" with "32 bit" due to the somewhat imprecise nature of the
> > definition of a "byte"
> why not use octet, which is the language used in the i-d?
no particular reason - I'm sure in any room half would prefer "bits" and 
half  "octets"

I'm personally inclined to use "bits" in preference to "octets"




> > Other topics of discussion on the ARIN list have been
> > - whether the terminology and nomenclature sessions should be
> > included in the policy proposal
> If policy is what decides if a requester gets the resource or not,
> then no, as this sounds more like local implementation (procedure),
> like most of the text (except for the dates, perhaps, which set what
> gets assigned and when)
However it does make the policy proposal clearer to read!


> > - whether the specification of dates are reasonable in this context
> is this a question of whether the suggested dates are appropriate or
> whether any dates are appropriate?
the latter - the ARIN discussion has appeared to head towards a conclusion 
that the dates are the critical part of the proposal, and to remove them 
from the proposal would make the entire exercise somewhat meaningless. I 
concur with that view.

> > - whether the policy alters the current sequential number
> > allocation registry practice
> > - the criteria (if any) that should be applied to a request for an
> > AS number of the "other" type
> > - the desireable size of the private use AS number pool
> Would the policy need a reference to the IANA as the ultimate
> caretaker of the registry?
No I do not think so. Nothing else changes in the policy proposal.

regards,

    Geoff