[arin-ppml] Draft Policy ARIN-2024-4: Internet Exchange Point Definition

Owen DeLong owen at delong.com
Thu May 23 21:41:13 EDT 2024



> On May 22, 2024, at 21:24, Martin Hannigan <hannigan at gmail.com> wrote:
> 
> 
> 
> On Wed, May 22, 2024 at 5:07 PM Tyler O'Meara via ARIN-PPML <arin-ppml at arin.net <mailto:arin-ppml at arin.net>> wrote:
>> Overall I support this change, but I have a few nitpicks:
>> 
>> 1) We should only include abbreviations/other names for the term if they're
>> actually used in the NRPM; I think future text that uses this definition would
>> be clearer if we selected a single acronym.
> 
> But that's not the way the real world works. All the acronyms are in use unfortunately.
>  
>> 2) I would remove the reference to Ethernet (or provide it as an example); we
>> shouldn't prescribe what L2 switching technology gets used by the IXP
> 
> Open IX OIX-1, an ANSI standard, prescribes ethernet for IX's and https://github.com/peeringdb/peeringdb/issues/1555
> There's pretty much a slammed door on the idea that a router is an IX as well.

IMHO, Open IX OIX-1 is in error here. I think that prescribing a specific transport technology is flawed. I agree that it should be a shared segment fabric (e.g. an ethernet switch or similar), but I do not agree that we should prohibit future alternatives to ethernet from being considered as valid IX candidates.

In another proposal, I mentioned the possibility of an (e.g. Infiniband) based exchange. Not saying I expect one of those next week, but I do feel like we aren’t in a position where we should be dictating people’s technical choices at that level and any valid shared fabric technology should be acceptable under the policy.

I would suggest that OIX-1 should be updated accordingly as well, though that’s orthogonal to the current discussion and this list in general.

>  
>> 3) I think Autonomous System should probably be capitalized, since we're
>> referring to a specific technical definition.
> 
> In that form "autonomous systems" should remain lowercase because it is a general term and not a proper noun.

Corner case. Often in legal texts, terms which have a specific definition (usually spelled out elsewhere in the document, (e.g. a contract)) are capitalized even though they are not proper nouns.

I’m OK either way in this case unless we are seeking to be certain that the use here is mapped specifically to an AS definition in section 2 of the NRPM (if there is one, I haven’t looked).

> The policy looks fine as it is, however let's see what the commentary on the 4.4 rewrite looks like to ensure they line up properly and as the community expects it will.

I think ethernet should be removed from both.

Owen

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20240523/3b3773b4/attachment.htm>


More information about the ARIN-PPML mailing list