On Apr 22, 2005, at 9:01 AM, Danny Butt wrote:
Hi Jeremy
I'm pleased to know that you're working on Internet Governance for the AIR conference. Hopefully you'll also be able to contribute comments on the WGIG's papers. More voices are good. We should probably discuss some of these questions in a specific governance forum, but I would like to suggest a couple of things for the casual readers.
1) While the implementation of the Anycast system expands the geopolitical *accessibility* of DNS root servers, it in no way expands the *control* of them. It's kind of like arguing that Starbucks is international because it shows up in a lot of countries. And the 13 "recently original" root servers (10 of which are in the US) still have a critical role which the others "mirror" (if they say strawberry frappuccino's on the menu, *it's on the menu*.
sort of, but there are only contract and consent governing this. it is more because of the fact that it operates this way that it seems to be necessary to operate this way. The necessity though is not there, should a better technical paradigm arise.
2) While I understand it is technically possible for individual root servers to "roll their own" root zone file, with the number of sites coming online every day, in a practical sense they are reliant on the master file from the "A" root server (which, as an irrelevant aside, I understand is not Anycast yet?).
nope, but it could be fairly easily.
3) Any edits to the master root zone file - which, let's remember, includes *all* of the "country code top level domains" that we mistakenly associate with actual countries (.jp, .ie, etc) - are approved by the US Department of Commerce. Not good if, for example, you're at war with the US.
but you see that this does not actually affect anything in real terms. if you are at war in the u.s., and the u.s. refuses to do something, that does nothing to your internal connectivity or external connectivity if you are technically advanced. you can add whatever you want at your own borders, just like companies do now, just like other subnets and supernets do. we rely on dns primarily for commercial services. and it is just naming. naming is important, but it is not as important technically as it might seem.
4) Almost no one - outside the Chinese government ;) - wants to see the ITU running the internet governance show. But still, governance of any sort attains its legitimacy through accountability to stakeholders, and the current US-dominated bodies have a response which is routinely "What's the problem?
other than commerce... the us-domination is not very clear to me other than perhaps as a cultural construct.
We don't *really* run anything?"
well other than ICANN having a say in naming, who does run anything? that is the question that I'm working on answering. the problem is that for the most part, 'we don't really run anything', yet someone runs something. the history of icann shows to some extent who runs things, but so does the history of isoc.
That wears thin after a while, so you might want to watch how your suggestions are received along the lines that, technically, it's a lot more open than it appears :).
true there is a difference between technical possibility and political reality in standards making. in technical possibility it is quite open, but getting the standards approved that is a different matter. however, in terms of ICANN, there is representation of regions and such, likewise in ietf, and almost every internet related system. here is an example of one of the most seniro groups in internet governance, the Internet Architecture Board... http://www.iab.org/about/members.html It is not perfectly representative, but it is not really u.s. dominated as much as dominated by non-governmental interests
Regards,
Danny
-- Danny Butt New Media | Research | Education | Development | Consulting db@dannybutt.net | http://www.dannybutt.net +64 21 456 379 | Aotearoa New Zealand ( in Australia through June 05 - +61 410 524 486 )
On 4/22/05 1:35 AM, "jeremy hunsinger" <jhuns@vt.edu> wrote:
I think that depends on what they mean by 'internet governance' as I noted earlier. If you mean icann, it is representative in a sense. if you mean ietf, etc. it is representative in a sense. however, if you mean the u.s. contract, then of course it is not representative at all. if you mean the powers that be in the non-state and non-technical sense, then those are not representative, and that will be covered in part at my hopefully soon to be accepted ir6.0 talk this year entitled Capital Policy: the transnational subpolitics of internet governance.
not really, because while ICANN can direct, who will enforce? other than economic enforcement... there may not be any real system of enforcement within the system other than exclusion, and that would require a unified effort, one that I doubt is possible.
actually, you can remove routing at any border or any endpoint, just like you can add it at any of those places. true, that doesn't make it disappear for everyone, but you can make whole areas of the world disappear for anyone depending on your place in the chain. That is what is interesting about dns. by the way although icann and the post-arpa roots are the dominant hegemony, they are not the only name system in operation, they can be ignored entirely if your country has the technical capacity to do so, which many countries do.
also, we should note that while the ruling root sits in the u.s. currently, the other root servers are fairly well distributed in various nations http://www.root-servers.org/. what that means to me is that in a realpolitik mode, the u.s. is only governing by consent, because any given one of those could break and become an independent root should they so desire or should the authority in their country demand it.
yes, I should say that I support the dissolution of icann and the movement of naming to a u.n. associated body, but do not currently see that it should be the ITU.
_______________________________________________ The Air-l-aoir.org@listserv.aoir.org mailing list is provided by the Association of Internet Researchers http://aoir.org Subscribe, change options or unsubscribe at: http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
Join the Association of Internet Researchers: http://www.aoir.org/
jeremy hunsinger jhuns@vt.edu www.cddc.vt.edu jeremy.tmttlt.com www.tmttlt.com () ascii ribbon campaign - against html mail /\ - against microsoft attachments