CircleID

Let me respond to three of the points cited in the main article.

First. ICANN's chairman says that his position has been misunderstood and that his position is that ICANN should have "a solid rationale for adding new TLDs."

Why in the world should ICANN have the authority to substitute its own judgment of what is proper for the internet in lieu of those who want to use the internet and innovate upon what has been done so far?

Indeed, who has given ICANN the legal authority to inject its judgment to control who may and who may not attempt to enjoy the rewards of being a participant in the domain name marketplace?

If we were to apply the position of ICANN's chairman to the timeframe of the 1970's it is questionable whether the internet would have ever occurred.  The 1970's were a period in which the incumbent telecos could have said, exactly as ICANN's chairman now says, that packet switching should be deferred until it can demonstrate a solid rational why it should be permitted to be deployed.

ICANN role is supposed to be merely that of limited technical coordination of a very small number of items.  Yet that is a job that ICANN does not do.

Instead, according to ICANN's chairman, ICANN sits in judgment of the business plans of every entrepreneur who wants to try his or her hand, at the risk of his or her own time, money, and reputation, in the domain name business.  If that's not the heavy hand of excessive regulation and imposition of arbitrary choices about business and social policy then there is nothing worthy of that description.

Second - There is that odd comment about "the average Joe".

We should not forget that ICANN has denied the average Joe any role, except that of distant observer, in the making of ICANN policy.  ICANN's crocodile tears for the average Joe are a sham.

Second, when, and by whose authority, has ICANN become a consumer protection body?

And if ICANN is such a body, how come it rejects and prevents the participation of the consumers it purports to protect?

And third, ICANN's chairman doesn't want the ITU to take over ICANN's job.  That's fine.  I would therefore presume that ICANN's chairman would have no objection to the ITU or other body from doing those jobs that ICANN does not do.

Considering that ICANN has decided to define its job as regarding only the business and economic aspects of DNS and doing nothing related to actual technical coordination of DNS (such as establishing operational requirements for root server operaters or requiring TLD operators to refrain from data mining or discriminatory levels of service) it is quite clear that there is a vacuum that ICANN does not fill and does not intend to fill.  Certain then ICANN, if we are to believe ICANN's chairmen, would not object should the ITU or some other body step forward and offer to perform those jobs of real technical coordination and oversight that ICANN has chosen to exclude from its job description.

With regards to adding new TLDs, are you suggesting that anyone that meets minimum requirements should be able to start a new TLD?  Please clarify.

Not related to your comment, I'd like to stress that Vint Cerf's comments about "some asshole" were said in a joking manner—it's hard to completely express that in writing :)

You ask: With regards to adding new TLDs, are you suggesting that anyone that meets minimum requirements should be able to start a new TLD?

Yes indeed.  I have felt that way for years and years - A half dozen years ago I proposed a policy.  See "Domain Name Policy - Top Level Domain Policy"

The only questions that ICANN should ask an applicant is whether the applicant will abide by published and broadly accepted internet technical standards and technical practices.  ICANN should not inquire into the applicant's business plan, business practices or policies, or require applicants to agree to contractual terms unrelated to the applicant's promise to abide by those technical standards and practices.

Nor should ICANN charge the outrageous application fees.  I figure a fee of $250 dollars ought to cover ICANN's actual costs of asking the appropriate questions and, if adequately answered (a simple "yes" is sufficient), for the actual cost of updating the root zone file.

(ICANN could charge a service fee for NS updates to the root zone file - $200 for each update should cover that.)

ICANN is attempting to be a gatekeeper that controls who may and who may not enter the domain name marketplace - the more blunt term for this is "restraint of trade".  Now, whether ICANN's restraint of trade is legal in some or all jurisdictions (or not) is a question I'm not answering here.  However, things that restrain trade are generally disfavored in free markets and are usually allowed to exist only when there is some clear and compelling reason.  Were ICANN to be making its choices on the basis of some objective technical criteria then ICANN's role in restraining trade might be accepted as being rational and good.  But ICANN has instead made its decisions on the basis of subjective and non-technical criteria and thus ICANN's position as overlord of the DNS marketplace is on a weak foundation.

Now there is the question of what to do if everybody and his brother were to come along and apply.  We know that even though from a technical point of view the DNS root can hold tens of millions of names without problem that such numbers raise the risk of administrative and human failures, not to mention the increase in time needed to disseminate root zone file images.  So we'd probably want to cap the number of top level domains at some rather smaller number.

And since that smaller number, whether it be ten million or ten thousand, is smaller than the number of people who might want a TLD, we need to adopt a fair allocation scheme.  And troublesome as they may be, it seems from considerable discussion, that the best method is probably some sort of auction coupled to a lottery - many slots would go via auction, the remainder by lottery.  By the way, notice that I used the word "slot" rather than "name" - it seems to me that the selection process should be blind to the character sequence being proposed and that the way to do this is to consider "slots" (right to operate a TLD) rather than "names".

I digress a bit from Karl's comments on TLDs to mention that Vint's participation in this event was something that I received a great amount of gratitude for in the hallways during the show, and in emails from participants as followup afterwards.

Understanding the point of view from either side of discussions often lends itself to better direction in the marketplace, and Vint's participation at the Domain Roundtable allowed many in the audience to hear about some of the 'why' and backstory of the evolutions of the whole process that we now have for things like new TLDs and root changes.

I am of the opinion that there is no perfect approach, or rather perfection in some blanket solution, but that we are making some progress in growing the namespace. 

Technologists, entrepreneurs, IP/policy advocates, governments, security experts, and general internet users all have opinions on the direction that things migrate, and there are places where these opinions collide.  This makes for a challenging exercise to get progress, but there is progress.  It is by no means perfect, but the processes are certainly further along than they were 6-7 years ago.

I hope that attendees came away from the keynote with an appreciation of the challenges that are in play towards making progress, and that Vint brings some reason and wisdom into the process.

On a separate note, my comment to Vint about sharing the 'Father of the Internet' title with Al Gore prompted a great backstory from Vint's personal recollection on the evolution of the interconnection of the supercomputing centers to become the internet, and about Al Gore's new movie on realities of Global Warming's impact.

As for the 'Asshole' comment, everyone present saw that Vint made that reference playfully, and there was no harm taken from my side on it at all.  It was memorable, and we'll ask him back for future Domain Roundtable conferences.

I would like to add my concern over the so-called "search vs. domain-name-guessing" controversy. As I mentioned in response to Mr. Cerf's argument (which has also been explicated in his previous interview of March this year here on CircleID), it is actually impossible not to select a domain before performing a search. Turning on a computer selects a machine-readable domain (and filters out paper documents that have been discarded to be picked up by the trash collector). Going online / using the Internet also selects a domain. Google.COM, Yahoo.COM and MSN.COM are all domains. So are hotels.COM and/or hotels.NET and/or hotels.US and/or hotels.DE and/or hotels.TV (and so on).

Whether the domain selection is being guessed or willfully chosen—or whether there are manipulative attempts to influence the choice via some "outside force" may be rather difficult to assess.

I hope and expect that those who are working in educational fields to increase online literacy (and/or reduce the so-called "digital divide") will be able to teach future generations of users to become experts at maximizing the potential of the Internet via an enlightened approach to selecting domains wisely.

I also hope that Mr. Cerf will clarify what he refers to as "search"—and I expect he will agree that that choice of domain is not only the starting point but also quite probably one of the most crucial elements in the search process.