Closed New gTLDs - "Closed Gardens"

[Milton L Mueller]

On Behalf Of David-who-agrees-with-Milton Cake

          This sounds like an extreme formalist argument - ignore the specifics of any given case as long as the procedure is the same.
          What is the point in arguing for the public interest to be considered at all if you don't want to apply it to any specific case, only to abstract procedural concerns?

[Milton L Mueller] I am arguing that the best policy is to have a relatively free and open name space, which means that someone who gets a specific TLD gets to choose what business model they apply, and that we do not impose top-heavy regulations in order to assuage completely hypothetical concerns. Competition issues in specific industries, should they arise, should be handled by actual economic regulators with the authority to do so.

BTW this was in fact the policy we adopted in 2005-6; we went for an applicant-driven process, we rejected taxonomies and top-down structured name spaces, and we opted for a criteria-driven, predictable process of approval as opposed to an arbitrary "public interest" standard defined on a case by case basis. It is not true that no one imagined there would be applications for generic terms at the top level, with controlled name spaces. Thus, another major problem with your argument is that you are asking ICANN to retroactively change its policy - something we oppose when the TM owners do it, and should oppose in this case, too.

          FWIW, I don't think an argument about open vs closed (or restricted) domains has merit in the abstract - but I do think that there is value in considering the merits of individual cases. Apparently, you don't? It is all just strings, and the public interest considerations for .book should be exactly the same as the public interest considerations for .adsfhda or .auieq or .arbitrarystringofletters?

[Milton L Mueller] In other words, you want to say, "we have no defined or predictable rules regarding what is allowed or not for nearly 2000 names. Anyone who doesn't like what someone have proposed can say "this is against the public interest" (an undefined term that can be invoked by anyone, for any reason), so now we are going to have a sui generis debate about whether this specific application is allowable or not."  Accurate description?

To me, that is self-evidently a disastrous policy. Moreover, it is not what the policy currently says. Indeed, one of the key principles of the new TLD policy was that the rules would be predictable and give applicants a good idea IN ADVANCE what was allowed and what wasn't. there was no stipulation that generic terms cannot be applied for. We fought and successfully eliminated the application of a "public interest" standard by the IANA, when the NTIA tried to put it in there.

If you want to oppose Amazon's application you have to show that it conflicts with defined Applicant Guidebook policies. You can't, because it doesn't. Hand-waving about (your specific feelings about) "the public interest" doesn't cut it with me.

There is no commonly agreed definition of that term "public interest." Anyone and everyone can claim that their favored policy is in the public interest. The trademark and business interests think that a more restricted an controlled namespace subject to vetos by brand owners is in the public interest. The Kingdom of Saudi Arabia believes that the public interest would be served by censoring all references to sex or other taboo subjects. I believe the opposite. In fact, there can be no meaningful debate about what is in the "public interest" - all such debates must be reduced to the criteria and methods used to ascertain what is and is not in the public interest, which may be things like competition, diversity, efficiency, security, etc. There is also a question about wtf is "the public" in this context. So get off the p.i. hobbyhorse and start talking about concrete harms and benefits.

Further, if you want to change the policy for the next round and not allow generic terms, start a PDP and change the policy if you can. But don't try to rewrite the rules under which people applied retroactively.

          I believe words have a significant semantic connection to its referent, otherwise language is meaningless.
[Milton L Mueller] semantic, my friend, semantic. Meaning, not intrinsic market power, not control of behavior, etc.

Now, I have made the assertion through naming that Dave-agrees-with-Milton several times, and yet you don't seem to be making any progress. Darn. This is a semantically meaningful assertion. I wonder why it is not happening? Might it be because the semantic connection has no necessary connection to reality?

Yes, .BOOK is a nicer place to put an online bookstore than .REFRIGERATOR. But it does not convey market power, and there is no prima facie case on market competition grounds for making .BOOK restricted to Amazon or open to other registrars. Indeed, refusing to allow it may artificially advantage businesses whose model hinges on making lots of individual registrations for money, and actually undermine a more competitive, well-organized lower-priced market for books. It may also turn out to be economically irrelevant, if the market for traditional books contracts and becomes organized around tablet platforms. It may turn out that the person who registers .TABLET or .APP gets farther. You just don't know...

There is a real argument here to be had about whether we should apply public interest considerations to individual non-community new gTLDs or simply treat them all as commercial investments whose open or closedness is of concern only to the investor

[Milton L Mueller] Sure, it is a real argument; indeed, we are having it. But for this round, the argument is over. We decided to allow the applicant to dictate business models.

- but you seem to not even want to approach that argument, and instead take the position that there is no intrinsic market value to words based on their semantics.
[Milton L Mueller] I am indeed taking that position. You have been unable to refute the examples I gave. And you have not defined "the market" you are talking about, which reveals the dangers of a "little bit of knowledge" about competition policy.

          You want evidence for the claim that people associate books with the word book?
[Milton L Mueller] No, I want evidence for the claim that there is "intrinsic" market dominance or public harm caused by the control of a generic domain name by an applicant who sells products or services that can be described by a generic name. Empirical evidence. One example, even (I gave you two). Still waiting.

But I take it your argument is 'if good names have market value, why aren't all the companies with good names winning in their business category'. To which I can only say, either you misunderstand the idea of competitive advantage, or have an awfully odd idea of how markets work. Businesses with one significant competitive advantage make other poor choices, or fail to overcome other market disadvantages, and fail all the time. Same with domain names. Doesn't mean we should ignore the public interest considerations of any possible competitive advantage just because it doesn't grant guaranteed success.

[Milton L Mueller] Dave-who-agrees-with-Milton, I have been researching and debating competitive advantage issues in the telecom.info space since you were in diapers. You need to understand the relevant concepts and measures of market power. One issue you clearly have not thought about is the importance of defining markets. ICANN can be properly concerned with the market for domain name registrations. I would assert categorically that it is illegitimate for it to try to decide what creates or undermines market power in ANY OTHER MARKET as a basis for assigning names. If you disagree you are, in effect, asking ICANN to be a regulator of the book market as well as the domain name market, which it lacks both the expertise and the mandate to do.

          I said 'significant competitive advantage'. Sometimes, despite a significant competitive advantage, people still drop the ball.
          I specifically 'not a monopoly', in fact. Do you have difficulty understanding the difference between exclusive control and competitive advantage in a marketplace?

[Milton L Mueller] OK, you argue that giving a company a generic term confers a "significant competitive advantage" but at the same time you admit that even with such advantages "people still drop the ball" and fail to dominate the market. Which means, in essence, that there is NO "intrinsic" competitive advantage. You've lost this argument. Sorry.

I am indeed proposing that the meaning of words should have some relevance to whether or not open registration is in the public interest. Further more, I would suggest that the broader social and legal context should have some relevance to whether or not open use is in the public interest, rather than just looking at a dictionary. I'm quite surprised that you find either of these contentions controversial - the relevance of both meaning, and social and legal context, would seem implicit in just about every ICANN policy debate about the desirability of registration of names, I fail to see why the question of open vs restricted or closed registration should be any different.

[Milton L Mueller] In other words, you want ICANN to be a content regulator in "the public interest?" Yikes.

          For example, in most jurisdictions we do, in fact, regulate which financial institutions are allowed to call themselves banks - and while that doesn't detract from the idea that bank is a word with many meanings, it does point to the fact that there are public interest considerations involved in the use of the term bank, and that having some rules regarding registration is likely consistent with both local laws. Furthermore, that financial fraud and phishing does constitute a large enough social problem that it should be considered a factor.

[Milton L Mueller] Here again, you inadvertently reveal how overreaching your argument is. You want ICANN to be a bank regulator, as well as a book market regulator, as well as a regulator of any activity or market that can be NAMED in the DNS. I don't want that.

          Which is an interesting position to take, that public interest is irrelevant, apart from the narrow case of criminal behaviour.
[Milton L Mueller] As stated above, "public interest" is a meaningless term, until and unless you convert it into something else with specific measurable effects, such as competition, freedom, security, etc.

          Though even that difficult to reconcile with the position you have taken - there are indeed criminal laws governing use of the word bank in many jurisdictions, for example
[Milton L Mueller] Two responses. First, those criminal laws are not made by ICANN, they are made by banking regulators. Second, you are confused by semantics again. There are not really criminal laws governing use of the word bank as such. There are laws governing banks, which may attack fraud by regulating the circumstances under which organizations may be identified as banks, or perform the functions of banks.

Going forward, if Dave-does-not-agree-with-Milton I suggest you still must concede that this is not an issue for this round. If you really want to do something more about this, if you are elected to the Council you should test whether there is interest in the SG and among your fellow Councilors in a PDP to develop a NEW policy on this issue. That is, rather than illegitimately attempting to reopen policies that are not in the AG and make arbitrary, de novo "public interest" determinations against specific current applications, you should propose to develop a new policy that can gain consensus.

[Milton L Mueller]
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ncuc.org/pipermail/ncuc-discuss/attachments/20120909/5dff5165/attachment-0001.html>