November 19, 2003

Domain Names and the FTAA: A Bad Mix

[HTMLized version of document presented to today’s conference on Negotiating Intellectual Property Provisions in Free Trade Agreements]

FTAA

Intellectual Property Rights Chapter, Part II, Section 1
Article 13 — Domain names on the Internet

[13.1. Each Party shall participate in the Government Advisory Committee (GAC) of the Internet Corporation for Assigned Names and Numbers (ICANN) to promote appropriate country code Top Level Domain (ccTLD) administration and delegation practices and appropriate contractual relationships for the administration of the ccTLDs in the Hemisphere. Each Party shall have its domestic Network Information Centers (NICs) participate in the ICANN Uniform Dispute Resolution Procedure (UDRP) to address the problem of cyber-piracy of trademarks.]

[13.1. Each Party shall make efforts, to the extent possible, to promote an adequate administration of domain names.]

Summary

The first paragraph is highly objectionable. The second is acceptable, although it is vague.

Nations should not be required to impose the ICANN Uniform Dispute Resolution Procedure (UDRP) by law on their citizens because:

  • Governments should not be subordinate to the decisions of a private corporation, ICANN;
  • Nor should the legal rights of citizens be decided by a private corporation;
  • The UDRP is subject to change, and governments should remain free to opt-out if its provisions become unacceptable;
  • As it stands, the UDRP fails to guarantee basic due process to consumers;
  • Serious questions have been raised about the even-handedness of some of the arbitration service providers who supply the arbitrators for the UDRP.

Governments that have thought most carefully about what rules should be imposed on their ccTLDs and their registrants have imposed their own arbitration systems which, while modeled after the UDRP, use home-grown rules that more closely track local law, customs, and commercial expectations. Canada is the leading example of this. This choice should not be over-ridden by the FTAA.

The term “Network Information Centers (NICs)” used in the first version of Art. 13.1 is poorly chosen because it refers to an entity concerned with the allocation of Internet Protocol numbers, rather than bodies such as registrars and registries that actually allocate Domain Names.

Requiring nations to participate in the ICANN Government Advisory Committee commits them to use resources attending far-flung meetings in expensive venues all over the globe, meetings which may not bring much in return. The GAC is in any event a secretive and undemocratic organization, so it is not obvious that governments would necessarily see their participation as warranted. The appropriate level of participation in the GAC should be left up to individual governments to decide for themselves.

In contrast, the second paragraph in the proposed text of Article 13 is acceptable. Although unobjectionable as written, its very lack of clarity — what is “an adequate administration of domain names”? — is slightly troubling, and perhaps invites future mischief.

Analysis & Explanation

Background

Domain names are the alphanumeric text strings to the right of an “@” symbol in an e-mail address, or immediately following the two slashes in a World Wide Web address. Thanks to a massively distributed hierarchical system for resolving domain names to IP numbers, the DNS, Internet software can rapidly and invisibly convert a domain name to its unique Internet Protocol (“IP”) number, the number used to route data correctly.

There are many “top level domain names” (TLDs), including .com and also 244 “country-code top-level domains” (ccTLDs), all of which are two-letter codes, and most of which use the two letters associated by ISO Standard 3166 to refer to a country. Thus, Canada’s ccTLD is .ca, and Columbia’s is .co. These ccTLDs are managed either by national governments, or by private citizens domiciled in the relevant nation, ensuring that the government has regulatory authority over the ccTLD.

Traditionally, second level domain names, such as “example” in example.ca, have been allocated on a first come, first serve basis. Every ccTLD has its own rules; some impose limits on who can register what, but others do not. This sometimes results in unhappy trademark and service mark owners, late to the Internet, discovering that “their” name is already registered by another. In some cases the earlier user is a legitimate business from a different sector, or is a non-commercial user who cannot be considered an infringer. But in other cases, the first registrant is either a standard trademark infringer, or a so-called “cybersquatter” — a person who in the business of registering domain names in the hope of reselling them to owners of identical marks, and who counts on the high cost of litigation, or its slow pace, to negotiate a windfall.

The Internet Corporation for Assigned Names and Numbers (ICANN) is a California non-profit corporation that manages the Domain Name System under contract to the U.S. Department of Commerce. It has an international board of directors, and an unusual structure that includes a “Government Advisory Committee” (GAC) which, despite its name, participates in the selection of ICANN directors and has a right of consultation on key policy issues.

ICANN imposes a mandatory arbitration clause on registrants to .com and other “global” TLDs (gTLDs) that are not linked to anparticular country. The clause can be invoked by anyone, anywhere, who thinks that his trade or service mark is infringed by a second-level domain name registered in a gTLD. The UDRP is popular with mark holders, who find it quick and relatively cheap. It has been criticized by the majority of academics who have studied it for lacking procedural due process, having structural biases, and for failing to give due consideration to the expressive rights of non-commercial users seeking to criticize corporations by hosting web sites with derogatory domain names (e.g. “companysucks.com”).

Analysis of First Paragraph of Art. 13

The major problem with the first paragraph of proposed Art. 13 is that it forces signatories to impose ICANN’s UDRP on the users of their ccTLDs, who will primarily be their own citizens (ICANN already imposes the UDRP on all registrants in gTLDs such as .com, and no FTAA action is required to maintain the status quo in this regard). Participation in ICANN’s UDRP for every ccTLD should not be required by treaty because ICANN is a private corporation. International law should not make governments subordinate to a private corporation. Nor, if the national ccTLD is private, should a government be required by international law to subordinate its nationals to decisions made by a foreign corporation.

Indeed, some of the governments that have thought most carefully about this question, such as Canada for example, have adopted their own domain name arbitration systems for their ccTLDs. While modeled after the UDRP, these use home-grown rules that more closely track local law, customs, and commercial expectations. Canada’s system is notably fairer procedurally than ICANN’s.

ICANN can change the UDRP at any time. Indeed, discussions are currently under way to expand the reach of the UDRP. Even if one is satisfied with the UDRP at present, states should not allow themselves to be locked into a system that might change in ways they find objectionable. And, there are reasons to think the UDRP is not ideal:

As it stands, the UDRP has a large number of obscure but significant procedural defects. Because of these, the system fails to guarantee basic due process to consumers who register domain names.

In addition, serious questions have been raised about the even-handedness of some of the arbitration service providers who supply the arbitrators for the UDRP; as currently written, the UDRP creates economic incentives for arbitration providers to be “plaintiff-friendly,” and to discriminate subtly against consumers. Service providers’ are nor required to disclose their methods of recruiting and assigning arbitrators, and the system permits provider manipulation of panelist selection to achieve a desired result in a given type of case.

Currently, the UDRP rules do not [corrected] require actual notice to the defendant, nor do they require sufficient efforts reasonably calculated to achieve actual notice, especially in countries with inferior postal systems. The time limits to respond are very short, and put ordinary people who do not have an internet-savvy trademark lawyer on permanent retainer at a substantial disadvantage. Although the UDRP purports to have word limits, these are routinely circumvented by the use of “attachments” further adding to the burden faced by parties needing to prepare their responses in a short time.

A final note on the first paragraph: The term “Network Information Centers (NICs)” used in the first version of Art. 13.1 is poorly chosen because this term usually refers to an entity concerned with the allocation of Internet Protocol numbers, rather than to the bodies such as registrars and registries that actually allocate Domain Names. The confusion could be damaging as to date the two regimes have intentionally been kept separate, and indeed are subject to very different governance mechanisms. Internet Protocol numbers are controlled outside the ordinary ICANN process via regional bodies (1) and by the Internet Assigned Numbers Authority (IANA), which is technically a separate function from ICANN’s management of domain names.

Analysis of Second Paragraph of Art. 13

As noted above, the primary problem with the second paragraph in the proposed text of Article 13 is that it is vague. It is not clear what constitutes “an adequate administration of domain names”, which perhaps invites future mischief. Since the duty to act is qualified by “to the extent possible” this provision does not appear too troublesome.



For further Information

     On ICANN

On the UDRP

1. These are APNIC (Asia Pacific Network Information Centre) - Asia/Pacific Region, ARIN (American Registry for Internet Numbers) - North America and Sub-Sahara Africa, LACNIC (Regional Latin-American and Caribbean IP Address Registry) - Latin America and some Caribbean Islands, RIPE NCC (Réseaux IP Européens) - Europe, the Middle East, Central Asia, and African countries located north of the equator, and soon to be joined by AfriNIC (African RIR, proposed).

Posted by Michael at 06:05 PM | Law: Trademark Law | Permanent Link | Comments (2)

How School Vouchers (Don't) Work In Florida

As Congressional Republicans plan to violate Washington DC’s home rule, and force a school voucher system on the city, I think it’s useful to review how vouchers are working here in Florida.

I’ll support almost anything reasonably calculated to improve the access of poor people — and rich people too! — to quality schooling for kids. Abstractly, vouchers look like they might do that, subject to a small host of caveats about the effect on kids with serious problems who might end up being treated like the ‘lemons’ of a primarily for-profit system, or dumped into the vestigial remains of an underfunded public system.

Given the Florida experience, however, one has to ask whether vouchers in practice are in fact reasonably calculated to be helpful on balance. As usual, the Florida Blog is on top of this one, with links to:

Talk about waste, fraud and abuse! Not to mention special interest lobbying, hogs at the trough, and every other cliche too.

Posted by Michael at 09:55 AM | Florida | Permanent Link | Comments (0)

Compare and Contrast

In one corner, William Safire, being tactical. In the other corner Joshua Marshall being reasonable. (And, in the background, the Defence Dept. disowning the report Safire is relying on.) You gotta wonder about prundits when they rely on stuff that’s already been disowned and refuted. Or just turn to your Daily Howler for a very jaundiced view of the pundit class.

The Safire article is a list of those who he thinks should confess error. Among them,

Former spooks who convinced reporters that there was never any connection between Saddam’s Iraqi regime and Osama bin Laden’s terror network would forthrightly assert they were uninformed about the decade-long links that were revealed in the classified memo the Senate Intelligence Committee requested from Under Secretary of Defense Douglas Feith. (The secret memo detailing 50 instances has gone relatively uncovered by major media because it surfaced in the current Weekly Standard, but is the subject of an automatic leak investigation — yet another time-wasting mistake.)

It’s an especially strange comment to make since the Dept. of Defense — no stranger to the neo-con world view, basically disowned the memo’s conclusions, saying “News reports that the Defense Department recently confirmed new information with respect to contacts between al Qaeda and Iraq in a letter to the Senate Intelligence Committee are inaccurate.”

Joshua Marshall has already been there, weighed it, and found it wanting

…is it “case closed”? Not quite. More like, case restated.

What do we already know about the intelligence wars over the Iraq-al Qaida link?

We know that most of the Intelligence Community didn’t think there was much there. Some contacts, but nothing substantial. We also know that Doug Feith — along with other administration appointees — didn’t agree. And Feith set up his own intelligence shop at the Pentagon to review all the raw data and find what the CIA and others had missed, misinterpreted or buried.

They came up with a raft of purported connections between Saddam and al Qaida. But when they presented their findings to professional analysts in the rest of the Intelligence Community, most notably at the CIA, the consensus was that those findings didn’t pass the laugh-test.

And who put together this new memo, the one the Standard article is based on? “The U.S. Government,” as the headline of the article says?

Not exactly. …. This memo is what Doug Feith sent them representing their side of the story. With the exception of some tidbits from interviews with Iraqis now in custody, this is, to all appearances, the same bill of particulars that Feith’s shop put together in 2002 and which was panned by the analysts in the rest of the Intel community.

So, the first point to make is that there seems to be little if anything here that the folks in the rest of the Intel Community — outside of Special Plans — did not see before concluding that there were no significant links between Iraq and al Qaida.

Point two is that Feith’s shop, the Office of Special Plans, the original source of this memo, gained an apparently richly-deserved reputation for what intel analysts call cherry-picking. That is, culling raw intel data to find all the information that supports the conclusion you want to find and then ignoring all the rest.

Now, of course, Feith’s advocates say that everyone else was just doing their own sort of cherry-picking, picking the evidence that supported their preconceived notions, etc. But this is simply another example of a pattern which we see widely in this administration: the inability to recognize that there is such a thing as expertise which is anything more than a cover for ideological predilection (for more on this, see this article.)

More to the point, there’s now a record. These are the folks, remember, who had the most outlandish reads on the extent of Iraq’s WMD capacities and the most roseate predictions about the ease of the post-war reconstruction. So their record of interpreting raw intelligence is, shall we say, objectively poor.

Spot the dinosaur.

Posted by Michael at 01:06 AM | Readings | Permanent Link | Comments (0)

Alcee Hastings Picks a Presidential Candidate: Lieberman

I didn’t see this one coming. Lieberman to pick up endorsement of Florida Rep. Hastings

I have no idea how much of this is tactical, and how much is something else like personal friendship. Alcee Hastings is running for Senate in Florida. In a crowded field he’d have a shot at the nomination, but how endorsing Lieberman helps him get that nomination I fail to see. Peter Deutsch will probably get most of the Jewish vote. And the odds of anyone being fooled into thinking Rep. Hastings is a conservative Democrat are pretty low, although he’s less predictable and more interesting than most Representatives. I suppose there’s some reason to think Lieberman will run strong in Florida, if his campaign has any juice left by the time we have our primary.

It could well be that Hastings just likes Lieberman for some reason. But it’s hard to see what that reason could be other than something personal. Lieberman is my least favorite of the major Democratic candiates because of his disgraceful reaction to the courageous decision by Illinois Governor George Ryan to commute the sentences of 167 death row prisoners

“Governor Ryan’s action was shockingly wrong,” Mr. Lieberman said … “It did terrible damage to the credibility of our system of justice, and particularly for the victims. It was obviously not a case-by-case review, and that’s what our system is all about.”

Sorry, that decision of Governor Ryan’s was shockingly right.

Here’s Alcee Hastings’s political mapping (vintage 2000):



Here’s Joe Lieberman’s position on the same map:



Hmmm. Less distance than I’d have expected.

Posted by Michael at 12:21 AM | Florida | Permanent Link | Comments (1)
Powered by
Movable Type 2.64