ICANN policy blegging

by Maria on July 3, 2006

It’s not often (well, ever) that I blog about anything directly work related. That’s because I work for an organisation that gets sued every 5 minutes, has a unique (and uniquely exposed) institutional model, deals with complex and controversial issues, and has endless stakeholders from dozens of bloggers to international organisations. That’s all by way of encouraging you to join in the fun!

ICANN is “an internationally organized, non-profit corporation that has responsibility for Internet Protocol (IP) address space allocation, protocol identifier assignment, generic (gTLD) and country code (ccTLD) Top-Level Domain name system management, and root server system management functions.” (try saying that in one breath.) Every year, our Nominating Committee meets to nominate individuals to decision-making positions in various parts of the organisation; the Board of Directors, the councils of the Generic Names Supporting Organisation, Country Code Names Supporting Organisation, and the Interim At Large Advisory Committee.

So, gentle CT readers, I suggest you take a look and see if being a NomCom participant at ICANN might interest you. You don’t get paid for attending our marathon meetings three times a year (though you do get your costs covered), you have to participate in conference calls outside your time zone, you have to get briefed and decisive on a wide range of issues, and you will take some flak for your decisions. This is not a commitment for the unenergetic. One current Board member reckons ICANN-related work takes up half of his working time.

So, it’s quite a challenge, all in all. On the plus side, NomCom appointees act as individuals and on behalf of the global Internet community. Which means you get to make a real public service contribution to policy regarding the domain name system of the Internet at a really crucial moment in its history. If this appeals to you, then please visit the NomCom website and check out the FAQ. Applications close on 16 July, 2006, and the successful candidates will take up their positions at our meeting in Sao Paulo this December. (Questions/comments; email nomcom hyphen comments at icann dot org.) Good luck!



James Wimberley 07.03.06 at 1:37 pm

One reason ICANN is “uniquely exposed” is that a lot of people who should know better think it runs the Internet. It doesn’t, only the name system – which is important enough. Who’s in charge of the physical infrastructure and technology? Telcos, the research community, and the ITU. Who’s in charge of the protocols? The IETF and W3C. Who’s in charge of security? Beats me. Who’s in charge of cybercrime? National cops and conventional intergovernmental organisations like the Council of Europe and OECD. Who’s in charge of strategic development and the funding model? Nobody, which has been a blessing up to now – but may be a growing problem in the future, as the Internet is one the core public infrastructures of our time. Sample non-ICANN problem: if the USA abandons Net neutrality in pricing, what should other countries do?


Antoni Jaume 07.03.06 at 3:07 pm

” Sample non-ICANN problem: if the USA abandons Net neutrality in pricing, what should other countries do?”

Well the attack has begun in the UK, as per a Guardian article published yesterday. I quite expected it.



Antoni Jaume 07.03.06 at 3:20 pm

I don’t know if thats linkrot proof, but tha’s the url:



Jane Shevtsov 07.03.06 at 10:47 pm

Why do most websites outside the US have a country suffix in their names? I can see how this would be useful for some businesses, but otherwise it seems to defy one of the coolest things about the Internet — that it makes nationality irrelevant. Plus, it would seem to say that US affiliation is assumed and anything else is an aberration.


Maria 07.04.06 at 2:20 am

Thanks, James. Not to be too pedantic or anything, but cybercrime isn’t in the purview of the OECD at all (though there are guidelines on privacy and security that OECD members are supposed to honour). The Council of Europe passed a Convention on Cybercrime a few years ago, which included broad provisions for mutual assistance between states. As far as I know, it hasn’t been ratified by so many member states yet (last I heard, the US hadn’t, but that may have changed). Essentially, computer related crime is handled bi-laterally between states and doesn’t have an organisation overseeing it. Though Interpol has some capabilities in this area.

But yes, one of the problems we have is that some countries think ICANN does a lot more than it actually does.

Jane, there are two types of top level domain; generic (e.g. .net) and country code (e.g. .uk). The US has its own country code, .us; it’s just not used as much in the US as generic TLDs are. There are a lot of good reasons for using cc’s. Some of them are legal; amazon in Europe only uses .uk, .de and .fr because it only wants to be captured by consumer protection legislation in those countries. Some are cultural/path dependent; .de in Germany and .uk in the UK are used by more businesses there than any generic TLDs are.

Nationality is almost never irrelevant, especially when you’re running a business with a defined market. I’d argue that the really cool thing about the DNS is universial resolvability. Wherever you type in name.com, or name.co.uk, you’ll get to the same place.


SusanC 07.04.06 at 10:41 am

I’d argue that the really cool thing about the DNS is universial resolvability. Wherever you type in name.com, or name.co.uk, you’ll get to the same place.

Well, it’s supposed to work that way. And many Internet applications don’t work properly when it isn’t true. For example, when you click on a link in a web page, you’ld like to get the page that the author of the link intended you to get, not a completely unrelated page or an error message. For this to work, https://crookedtimber.org (for example) ought to resolve to the same thing both for the author of the page that contains it and for the viewer of the page.

To obtain this highly desirable property, you need some kind of committee to agree on what the top-level domains are and who they’ve been delegated to. (Second level domains like .uk can have a regional body to decide what goes in them etc). This is where ICANN comes in.

An important technical point: any end-user can reconfigure their computer to ignore ICANN’s top level domains and use an alternative set provided by whoever they feel like. Of course, because you want to see the same TLDs as everyone else, there’s little point in doing this on your own. More to the point, an ISP can reconfigure their DNS so that all their customers (by default) see a different set of top level domains.

So ICANN’s real source of authority is that sticking with them is less painful than switching to an alternative (for customers, and ISP who have to deal with customer complaints that “the Internet is broken”), and most national goverments have not required the ISPs in their country to use non-ICANN TLDs.

Unfortunately, ICANN’s decisions on which TLDs to add have upset sufficiently many major players that the consensus is starting to fall apart (or fail to come together in the first place), and major ISPs are adding new TLD’s without ICANN’s permission. Technically, they can do this, and ICANN has no ability to stop them. If enough of them do this, ICANN is toast.

For example: customers of Chinese ISPs get to see non-ICANN TLDs like .公司. Advantage of this: Chinese-speaking Internet users in China can enter URLs in their own language. Downside: links to these URLs are broken for Chinese-speaking users browsing the web from anywhere else. If ICANN had managed to agree on some Chinese-character TLDs we could have had Chinese-character URLs that work everywhere.

The Chinese TLD issue isn’t very visible for English speaking users in the US. US users would see the impact more if (for example) their ISPs started adding a .xxx domain for porn sites without ICANN approval.

Steve Murdoch has a good explanation of what the Chinese ISPs have done:

New Chinese TLDs
BBC article on new Chinese TLDs


SusanC 07.04.06 at 11:05 am

It occurs to me that the dysfunctional nature of ICANN might make an interesting case study in public choice economics.(Maybe one of the economists here would like to do it?)

e.g. one possible effect of adding .公司 as a TLD for Chinese commercial sites is that whoever is providing .公司 will gain a lot of registration fees from companies who want a domain name ending in .公司, and whoever is providing .com will loose a lot of registration fees from companies who don’t need a .com now that they have a .公司. So someone has a strong financial incentive to lobby ICANN against adding 公司, even if it’s in the “public interest”…


Tom Hudson 07.04.06 at 4:32 pm

Given the apparently dysfunctional ICANN governance process that’s been publicly visible (e.g. Karl Auerbach), can you say anything to reassure those of us who thought to avoid ICANN altogether? I don’t see ICANN doing much of anything “on behalf of the global Internet community” after those events.


Maria 07.05.06 at 2:25 am

Tom, I don’t know which events you’re referring to, so all I would say is ‘do your homework and make up your own mind’. There are lots of blogs and sources about ICANN maintaned by lots of different people – not just the noisiest ones. You could start with the list of blogs provided by ICANNwiki that I linked to in the post.

Susan, ICANN is actually working on internationalised domain names and new TLDs right now (the issues have some overlap but aren’t quite the same). People have always been able to create different TLDs themselves at different levels (their own discrete network, or locally through an ISP and at the browser level) – nothing new there. But these applications have limited use because they lack universal resolvability. FWIW I tend to think of universal resolvability as a public good, with all that entails.

As to a public choice analysis of ICANN, Milton Mueller started with a 2004 book I linked to in the post above, using a fairly light-touch institutional economics approach. More recently, the part of ICANN I work for – the GNSO – has had a review done by the Government Dept. at LSE. I expect the report will bring at least some public choice analysis to bear on it. Otherwise, the field’s wide open…

Comments on this entry are closed.