Dear Nick, Remco, address-policy-wg, The suggested changes Nick put forward appear sensible, and should result in a lasting framework with a less irregular process. By keeping the hostmasters in the loop, the requirement "For the purpose of evaluating, the request will be treated as if it were filed by a regular LIR." is met by the most competent authority. This, in turn, allows the Pool of Arbiters to concentrate on the merit of a request rather than policy technicalities. Therefore, I support these changes. -- Respectfully yours, David Monosov Nick Hilliard wrote:
On 25/03/2009 08:41, Remco van Mook wrote:
There won't be a contract because it's an impossibility - that's one of the main rationales behind the proposal. We could have the NCC sign twice but that's a bit ridiculous and won't be a legal contract anyway.
A couple of comments here:
1. It may be a good idea to explicitly note in the proposal that the RIPE NCC is formally exempted from signing any of the normal contracts required for number resource assignment / allocation.
2. I don't want to sound like the language fascist here, but as the proposal deals with both assignment and allocation of resources (to use the RIPE NCC terminology), it might be a good idea to use the two terms in the policy proposal. Currently the word "allocation" is used, but if the NCC is going to apply for a PI ipv6 /48, then that's an assignment. Similarly for ASNs.
3. I have a slight preference for using the Pool of Arbiters instead of the WG Chairs for the approval mechanism, purely on the grounds that smaller committees are better than bigger ones. I don't have a problem with Remco being proposer of this policy change and also being on the pool of arbiters. Just out of interest, the pool of arbiters is described here:
http://www.ripe.net/membership/arbitration.html
4. Regarding the function of the approval group, there are two important requirements for RIPE NCC number resource assignment / allocation process:
a. consistency with the assignment / allocation guidelines b. process transparency
As it stands, the proposed process delegates the entire process of number resource approval to the approval group, with no obligation to explain or publicise their decision. I just wonder if it wouldn't be better to have a process like this:
4.1: assignment / allocation request received and processed by RIPE NCC hostmasters who will give a formal written opinion on whether the request is consistent with current assignment / allocation guidelines 4.2: this opinion is evaluated by [pool of arbiters / WG chairs] who are entitled to approve the request only if the hostmaster team find that the request is consistent with current rules. 4.3: either way, the request, the hostmaster recommendation and the reasoning of [pool of arbiters / WG chairs] is published. 4.4: if both hostmasters and [pool of arbiters / WG chairs] decline request, then petition can be made to RIPE Plenary.
There are a couple of reasons for this alternative proposal. First, it's the purpose of the RIPE NCC hostmaster team to evaluate whether number requests are consistent with the rules. The [pool of arbiters / WG chairs] are not hostmasters. Secondly, the [pool of arbiters / WG chairs] should not necessarily be given sole authority to decide whether a number request is consistent with current RIPE rules; 4.2 above implies that if the RIPE NCC hostmaster team believes that a request is not justified, the [pool of arbiters / WG chairs] have no authority to override that decision. Thirdly, the reasoning and decision of the [pool of arbiters / WG chairs] should be published so that this mystical ideal of transparency is achieved.
Nick