[address-policy-wg] 2014-08 New Policy Proposal (Language Clarification in “Contractual Requirements for Provider Independent Resource Holders in the RIPE NCC Service Region”)

On Thu, Oct 23, 2014 at 03:15:24PM +0200, Marco Schmidt wrote:

> A proposed change to RIPE Document "Contractual Requirements for Provider 
> Independent Resource Holders in the RIPE NCC Service Region" is now available 
> for discussion.

>     https://www.ripe.net/ripe/policies/proposals/2014-08

# Any resources allocated/assigned to the RIPE NCC will be registered in the
# RIPE Database. All policies set for allocating or assigning resources to
# LIRs apply equally to the RIPE NCC. The RIPE NCC as a resource holder should
# fulfil the same basic requirements also expected of normal LIRs, such as returning
# unused resources. As an exception from normal LIRs, RIPE NCC as a resource holder
# is exempted from signing any of the normal contracts required for number resource allocation/assignment.

RFC 2119, as mentioned in the introduction, is not relevant to this document
(see earlier mail).
In this case, the change appears straightforward even if should vs must was
interpreted in a natural language context. On second thought, it remains
undefined what the "basic requirements" are (leaving another ambiguity)
and it isn't clear whether the list of exemptions is exhaustive.  As an
example: is the NCC undergoing a regular or case by case audit and
who is supposed to execute that?  Arbiters are supposed to evaluate
requests, but how would the NCC's adherence to the "basic requirements, [...]
such as returning unused resources" be supervised?  There's no vox populi
appeal to the arbiters.
But wait, we have a general community involvement in the NCC's operation
and my perception was it was considered functioning.  So, I'm not
really convinced that an s/should/must/ would prevent the NCC from running wild
if it wanted to.  On the other hand, I might be looking forward to
future policy proposals and further should vs must (or similar) discussions
except that the PDP rarely provides the right slot for this level of detail.
In other words: what's being tried to fixed here is likely to happen again.

Back to 2014-08: it's either unnecessary or incomplete.

-Peter

PS: Obviously in this case it is prudent for the NCC as the submitter to
    err on the side of caution