[pcap-ng-format] [OPSAWG] draft-gharris-opsawg-pcap.txt --- IANA considerations

[Adrian Farrel]

Hi,

I'm pitching in here as Independent Submissions Editor (ISE). I really should use the proper email for this (rfc-ise at rfc-editor.org) but I'm sticking with this address as it is already subscribed to the OPSAWG list.

Responding to this particular email because (I think) it is the first one that mentioned me by name.

Let's take a few things:

1. Find out more about the Independent Submissions Stream by looking at https://www.rfc-editor.org/about/independent/ and reading the referenced RFCs.

2. Michael is right that RFC 8726 will govern how I handle drafts that make requests for IANA action. Precedent will not carry any weight: 8726 describes the process in place at the moment.

3. While it *is* true that 8726 describes situations under which IANA registries can be created by Independent Submission Stream documents, be aware that the conditions are very limited. Also be aware that the assignment policies that you can set for such registries are also very limited.

4. The Independent Submissions stream is not a short-cut to RFC publication. If there is a proper home within the IETF then that path must be followed. If the IETF process is not functioning to your satisfaction, then you need to work with the IETF processes to resolve that issue.

A previous email on this thread said...

>> So, for FCFS, does that mean anybody who wants a linktype can just grab one?
>
> Yes, but it's not quite the free-for-all one might imagine.
> They have to email IANA and there are some records kept.
> IANA does do some amount of abuse control.

FWIW, you can impose a little bit of control by specifying a form that must be used to make a FCFS request to your registry.

> I've re-read RFC8726, in case we want to send pcap via the ISE, as OPSAWG is
> very slow to adopt documents.   The short of it is that we *CAN* create
> Registries on the Independent Stream Editor Q.
>  https://www.rfc-editor.org/about/independent/
>
> But, we can not create Specification Required registries as that requires
> assignment of Designated Experts.  FCFS streams are fine.
> Options are there:
>  1) get OPSAWG to adopt this document.
>  2) ask an AD to sponsor the document.
>  3) not split the Registry as per above, use FCFS for it all.
>  4) move the LinkType Registry to pcapng, send pcap via ISE as
>     Informational (might get hung up waiting for pcapng, however, depending
>     upon how we write the text, and whether or not it results in a MISREF)

I strongly suggest that you pursue options 1 and 2 because the work appears to be in scope for the OPS Area.
If the OPSAWG declines to adopt the work, and if you can find no AD to sponsor the draft, then it is worth coming to the ISE (me). Additionally, in the (very unlikely) event that progress is procedurally blocked and appeals are not producing results, I can also consider publication.

Obviously, as already noted, if the document does wind up with the ISE:
- IANA actions will need to conform to 8726
- normative references will need to be cleared before an RFC can be published

Cheers,
Adrian
--
Independent Submissions Editor (ISE)
rfc-ise at rfc-editor.org

    >> -----Message d'origine-----
    >> De : OPSAWG [mailto:opsawg-bounces at ietf.org] De la part de Michael
    >> Richardson
    >> Envoyé : mardi 22 décembre 2020 17:36
    >> À : Guy Harris <guy at alum.mit.edu>
    >> Cc : Pcap-ng file format <pcap-ng-format at winpcap.org>;
    >> opsawg at ietf.org; tcpdump-workers <tcpdump-workers at lists.tcpdump.org>
    >> Objet : Re: [OPSAWG] [pcap-ng-format] draft-gharris-opsawg-pcap.txt
    >> --- IANA considerations
    >>
    >> But, we can not create Specification Required registries as that
    >> requires assignment of Designated Experts.

    > [Med] I'm not sure if the rules changed since then, but experts can be
    > designated for registries created by ISE documents. See for example:
    > https://www.iana.org/assignments/iax-parameters/iax-parameters.xhtml#iax-parameters-7

Well, this document is ten years ago, and RFC8726 is Nov. 2020, and I think
that the confusion on this question is what generated the need for RFC8726.

Adrian: wrt:
  https://mailarchive.ietf.org/arch/msg/opsawg/m4I4vMrYyAxhv18k3LcUk-_Te94/

--
Michael Richardson <mcr+IETF at sandelman.ca>   . o O ( IPv6 IøT consulting )
           Sandelman Software Works Inc, Ottawa and Worldwide