[rfc-i] Comprehensive review of draft-iab-rfc-editor-model-v2-02 - Protocol Parameter Registry Operator

Brian E Carpenter brian.e.carpenter at gmail.com
Tue Jul 12 15:08:49 PDT 2011

On 2011-07-13 06:24, SM wrote:
> I am unfortunately unable to provide references to supplement some parts
> of this message as the www.iab.org move to Wordpress broke the old "links".
> Many protocols defined in RFCs make use of identifiers consisting of
> constants and other well-known values.  The IETF uses a registry
> function to record these protocol values and their associated semantic
> intent.  Historically, this registry is referred to as the Internet
> Assigned Numbers Authority.
> A proposal about defining IETF Internet Assigned  Numbers Authority as
> it applies to the IETF Standards Process was submitted in December
> 2002.  The publication of the proposal through the IAB Stream in April
> 2011 resulted in RFC 6220.  For what it is worth, this is done under
> contract but I prefer not to get into that.
> In the text quoted above, John Klensin discussed about the relationships
> between various bodies and IANA.  I'll steal a few words from him: "In
> practice, it doesn't make any difference".  Where it makes a difference
> is when the IAB takes a position on the IANA NOI.
> The previous IAB was of the opinion that:
>   "The IAB has the responsibility to appoint an organization to
>    undertake the delegated functions of the Protocol Parameter Registry
>    Operator for each IETF protocol parameter."
> If the current IAB would like to maintain that opinion, I suggest that
> it uses "Protocol Parameter Registry Operator" instead of "IANA" in some
> parts of draft-iab-rfc-editor-model-v2-02.  The roles are clearly
> defined in RFC 6220.  I note that RFC 6220 states that:
>   "Any intellectual property rights of IETF protocol parameter
>    assignment information, including the registry and its contents, and
>    all registry publications, are to be held by the IETF Trust on behalf
>    of the IETF."

That's all very interesting, but the IAB Charter (which is a BCP, unlike
RFC 6220) only charters to the IAB to appoint one IANA, which takes
technical instructions from the IESG:
  "The IAB must approve the appointment
   of an organization to act as IANA on behalf of the IETF. The IANA
   takes technical direction on IETF protocols from the IESG."

RFC 6220 is written in more general language, but the IAB doesn't today have
authority to create mini-IANAs. That being so, I can't see why the present
draft needs to use more complicated terminology than just saying "IANA".

> I suggest removing the following sentence from Section 5:
>   "The IAOC will facilitate the establishment of the relationship
>    between the RFC Production Center and IANA."

Why? I certainly hope that the IAOC will continue to do this.


More information about the rfc-interest mailing list