[discuss] Roadmap for globalizing IANA
jcurran at istaff.org
Tue Mar 4 03:55:48 UTC 2014
On Mar 3, 2014, at 5:52 PM, Milton L Mueller <mueller at syr.edu> wrote:
> Today IGP released an innovative proposal to resolve the 15-year controversy over the United States government’s special relationship to the Internet Corporation for Assigned Names and Numbers (ICANN).http://www.internetgovernance.org/wordpress/wp-content/uploads/ICANNreformglobalizingIANAfinal.pdf
> The proposal, which involves removing root zone management functions from ICANN and creating an independent and neutral private sector consortium to take them over, will be presented at the Singapore ICANN meeting March 21, and has also been submitted to the “NETMundial” Global Multistakeholder Meeting on the Future of Internet Governance in São Paulo, Brazil, April 23 and 24.
> We propose four basic principles to guide the reform of the IANA functions: 1. Keep the IANA function clerical; separate it from policy; 2. Don’t internationalize political oversight: end it; 3. Align incentives to ensure the accuracy and security of root zone maintenance; 4. De-link globalization of the IANA function from broader ICANN policy process reforms. Even if there are quibbles about the details of the proposal, we look forward to gaining agreement on those principles, and are willing to entertain any proposals that embody them.
Thank you (and Brenden :-) for submitting this proposal; if nothing else, its principle-based
assessment (and proposed solution) may lead to better understanding of some of the more
interesting aspects of the present IANA function arrangements...
I have a few comments and suggestions regarding the proposal; feel free to use or discard
as desired -
a) You describe the IANA functions in the context of the NTIA contract, but don't mention
the MOU between IAB/IETF and ICANN (RFC 2860) by which ICANN agrees to
perform IANA registry updates (for IETF protocols) in compliance with IETF RFCs;
further noting that assignment of domain names and the assignment of IP address
blocks are also subject to "policy issues" for which ICANN may follow its adopted
policy so long as it does not prevent RFC compliance (less the IETF terminate)
Given that the IETF defines the protocols which create the actual registry spaces
(DNS, IPv4, IPv6) for the general purpose identifiers that are being administered,
it might be worth seeking similar acknowledgement from the IAB/IETF of any new
arrangements for clerical administration of the DNS identifier space. Some may
argue that formal recognition isn't needed for various reasons, but even nominal
acknowledgment might help prevent potential confusion in future DNSA / IETF
relations (e.g. related to DNS protocol & technical requirement changes) as well
as providing a framework for those entries in the DNS root zone which are purely
present due to technical requirements (e.g. .arpa)
b) Your proposal suggests that "the IANA functions related to protocol parameters
would be moved to the IETF", where believe it might be more appropriate to
suggest instead that "the IANA functions related to protocol parameters should
be performed per authority of the IETF (who may perform them directly or instead
contract them to a third-party as they choose)" The IETF may not necessarily
have the desire or capacity to perform these tasks directly, so it might be better
for the proposal to simply note the authority by which they will be performed.
c) The proposal also suggests that "the IP address-related functions would be retained
by ICANN until such time as new global policies regarding their disposition could
be developed"; that is interesting considering that the address aspects of the IANA
functions are something today which is very clearly defined, both by ICANN Address
Supporting Organization (ASO) MoU and IETF RFC 7020. It might be best to simply
note in a parallel fashion per above that "The IANA functions related to IP addresses
should be performed in a manner agreed by the IETF and the Internet addressing
community." rather than distract from the real DNSA-focus of your proposal.
d) From points b and c above, it is fairly clear that it would be useful for the IETF (at
some point) to set some broad guidelines for how they wish to handle administration
of general-purpose identifier registries, so that if we ever get a new protocol (or new
version of an existing protocol) which has globally-unique identifiers for general
purpose consumption, we'd at least have some guidelines that we'd all know would
apply such as: authority for general purpose identifier registries should only be
delegated to parties which the IETF believes are reasonably representative of the
communities dependent upon the registry, that do policy development in a manner
open to all interested parties and via transparent decision making processes, etc.
This also may not be a matter directly for inclusion in your proposal, but still worth
considering as may imply a source for some of the criteria for both the DNSA and
any party performing policy development activities for the general-purpose DNS
e) As a general comment it would be worth noting the irony of the proposal, in that
it finishes the inversion of ICANN which started with the absorption of the original
DNSO into ICANN. ICANN was intended to be the administrative coordination
body for all of the IETF registries (protocol numbers, IP address, DNS names) and
was to make use of external bodies for primary policy development functions. Per
your proposal, DNS policy development is finally recognized as ICANN's "reason
for being"; a very significant task which deserves a body dedicated to that mission.
Thanks again! Feedback on the above thoughts would be welcome.
Disclaimers: Read carefully - My views alone. In particular, while I have some
awareness of the various organizations mentioned in this message,
that does not imply in any manner any overlap between them and
my particular hallucinations on how things might work, and in fact,
any proposal following my suggestions may be become less (rather
than more) credible as a result... caveat emptor. No refunds.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the discuss