<<<
Chronological Index
>>>
Thread Index
>>>
Comments on Framework of Interpretation Working Group Interim Report
- To: foiwg-interim-report-2012@xxxxxxxxx
- Subject: Comments on Framework of Interpretation Working Group Interim Report
- From: Antony Van Couvering <avc@xxxxxx>
- Date: Sat, 4 Feb 2012 02:11:02 -0800
Dear Working Group,
Please accept my comments on your interim report.
I was formerly very engaged in ccTLD redelegation issues, but have not paid
attention in recent years. Reading your report, I am pleased that the
definition of "interested stakeholders" has passed from a pitched battle
between government claims, which war not much different in tone "all your base
are belong to us" and advocates of strict non-interference. In particular, I am
heartened to see a continuing reference to RFC 1591, which apart from its
historical claims still strikes me as the best brief description of how to run
a top-level domain, and by far the most humane. RFC 1591 is now 18 years old,
and is not a holy script. Yet it does contain the best of the spirit that gave
birth to the Internet and guided it as an open and truly global medium -- so
its recommendations deserve consideration and respect.
I do have a couple of comments, however, that you may be interested in
considering. First, in answer to your question, "Is the approach used by the
FOIWG satisfactory?" While the report does a good job in identifying what it
calls Significantly Interested Parties, I have a few concerns and comments in
this regard:
1. Decisions for redelegating a ccTLD continues to turn on political support,
rather than performance on the job. While there is an undeniable political
dimension to the administration of a ccTLD, and the administrator needs to be
responsive to its stakeholders, I urge the working group to consider not just
who has support, but how well they are doing their job. In the case of what I
will call a "hostile" redelegation, where the incumbent does not wish to be
removed, there should be a burden on those advocating redelegation to show that
the administration of the ccTLD has some other flaw other than the dislike of
the manager by certain stakeholders. This might not be the only factor, but
in my view it is important. If the manager is performing its duties from a
technical, managerial, and financial perspective, they should have a
presumption of incumbency so long as they are responsive to inquiries and
suggestions from stakeholders.
2. Notably missing from the list of Significantly Interested Parties are two
groups whose interests have long been recognized: the local and the global
Internet community. RFC 1591 speaks about the stakeholders served by a
top-level domain:
"These designated authorities are trustees for the delegated domain, and have a
duty to serve the community. The designated manager is the trustee of the
top-level domain for both the nation, in the case of a country code, and the
global Internet community."
In RFC 1591, the phrase "significantly interested parties" is not defined or
enumerated. Occurring as it does in the context of a dispute between
"contending" parties, it strongly suggests that at least one of the interested
parties is the current manager. Otherwise it is silent.
My reading of RFC 1591 is that the local community (or "nation") and the global
Internet community should be considered to have a "direct, material,
substantial, and demonstrable interest in the operation of the ccTLD(s)." Any
other interpretation risks shutting out those whom ccTLDs are supposed to
serve.
3. A revocation or seizure of a ccTLD as a political act puts at risk the
requirement (again from RFC 1591) that a top-level domain be run by a
"responding to requests in a timely manner, and operating the database with
accuracy, robustness, and resilience." Whenever the "significantly interested
parties" are from a territory where communications are poor, Internet usage
nascent, or free expression of ideas is curtailed in a significant way, there
is the possibility of "capture" by just one group (often the government) which
will claim to represent all the stakeholders. This party may or may not have
any knowledge or ability to manage a ccTLD, and that puts "accuracy, robustness
and resilience" at risk.
From an operational and security perspective, therefore, the definition of who
is a significantly interested party should include domain name holders,
Internet users, and the global Internet community as parties with legitimate
interests. If the definition of who is a significantly interested party is too
narrow, it runs the risk of capture by a manager who is not capable of
operating the ccTLD in an acceptable manner.
4. The fact it can be demonstrated that the local and global communities are
significantly interested parties does not mean that they should be forced to do
so on every occasion. In other words, there should not be a burden on these
parties to prove their bona fides. They should be explicitly enumerated in the
way that governments are.
In summary, I encourage the Working Group to consider job performance, whether
actual or prospective, and to explicitly include the local and global Internet
community as "significantly interested parties." It's not easy to say what the
"Internet community" is, or how to find out what it thinks, or how to engage it
effectively. That doesn't change the fact that it is the local and global
Internet communities that a top-level domain serves, and they should not be
excluded having a say in who runs it, and how it is run.
Thank you for considering these comments.
Antony Van Couvering
<<<
Chronological Index
>>>
Thread Index
>>>
|