February 7, 2011

No Early Birds Here

In business the early bird catches the worm.  It seems, however, that in the domain name business that there are going to be a lot of uncaught worms.

If you were starting a new business would you sit on your hands waiting for an approval that you do not need, pay fees that you do not need to pay, publish your customer list to the public, risk having hour business name given to someone else, and be required to sell your product through a distribution chain that you do not control?

Of course you would not.

But I see that there is conference this week of people who are doing exactly that - .nxt A Conference About New Internet Extensions.

Don't these people realize that they can go into business today? That by so doing they can establish an early priority date for the start of use of their name in commerce thus putting any competition for the name of their business into a defensive posture?

This conference is for people who are sitting on their hands hoping to apply to ICANN for a new top level domain (TLD).

But ICANN is not the only game in town - Any one can set up an operating TLD business this afternoon.

That sounds like magic, but it is not.

ICANN (in conjunction with the US Dep't of Commerce and Verisign) constructs what is a called a "root zone file".  That is merely a textual computer file that lists all of the TLDs that that trio wants to recognize along with pointers to the domain name servers run by each TLD provider.  (There is other stuff in there pertaining to DNS Security, DNSSEC - I'll get to that in a bit.)

That zone file is picked up by an independent group of people who run what are collectively called "root servers".

What most people do not understand is that anyone can construct a root zone file and that anyone can run root and TLD servers.  There is neither a technical nor legal obstruction to doing so.  In fact such competing root zones, root servers, and TLDs have been out there for years.  But they have gained practically no market share because for the most part they were operated poorly, never promoted, and had to overcome the Fear, Uncertainty, and Doubt (FUD) spread by ICANN.

So let's put ourselves into the shoes of someone going to the .nxt conference who wants to run a top level domain call .cabbage.

If they follow the ICANN path they will construct a thick pile of paper demonstrating that they are willing to structure their business according to the ICANN approved business models, that they are willing to distribute their product through ICANN approved distributors over which they have no control, and to frost that paper cake with a check for $185,000 and a promise to pay ICANN a piece of the gross revenue from .cabbage.

And, of course, ICANN might sit on these applications, as they have done in the past, for more than a decade - there are people who applied in year 2000, paid the money, and who are still being strung along by ICANN.

OK, so what's the alternative?

First of all, the folks at .cabbage can set up operational servers for their TLD.  With today's ability to lease physical and virtual servers in "the cloud" the .cabbage folks could have a worldwide array of name servers for .cabbage up and running within a few hours.

They could begin accepting customers immediately - the registration system at first could be as crude as a stack of 3x5 index cards - automation can come later as the customer base expands.

Given that it is only right and proper to inform customers that .cabbage is not going to be seen by the current majority of network users the question is how to overcome that obstacle and get people to sign up.

The answer is the same one that has been used for years - give early customers a good deal.  Make the cost nil or very low, in perpetuity.  Or cut them in for part of the revenue or profit should .cabbage ever go big-time.  (Make sure that you check with you lawyer for details how to structure those sorts of things, you don't want to trip over a securities law.)

If you or your customers don't like publishing your customer list to the public via that thing called "whois" then don't - there is no legal obligation to do so.  That obligation only arises when you sign a contract with ICANN.

And how would you ever get ISP's to include your TLD (or root) into the constellation of TLD's that they export to their users? Again, it is easy, offer 'em a cut of the revenue or profit from .cabbage.

These things ought to sound familiar - they are - they are akin to the Google Adsense and Adwords ideas moved over to the domain name space.

Once you begin getting customers make sure you get 'em in all 50 states and in countries other than the US.  And in addition you may want to file fictitious name statements or set up corporate entities.  By so doing you will have an actual operational business-in-being that may be recognized as establishing priority rights to the name of your business in the state or country.

If ICANN then approves someone else under the same TLD name the promoter of that other TLD will have to overcome the possibility of being excluded from all of the states and countries that recognize your priority.

The largest and hardest issue is how to make money.

Nobody is going to get the windfall that was .com for Verisign.

So, take a cue from Google - there is gold in DNS queries.  Those queries represent an instant by instant view of the interests of users.  That data can be mined and sold.  (Play it safe - make sure that your sales contracts clearly disclose this to customers and ISPs, particularly if the resulting data stream could be tied to individual users.)

I said that I'd mention DNSSEC.  If you want to have a DNSSEC protected root zone you will have to generate the appropriate cryptographic keys and undertake the management tasks.  Those are not huge tasks, but they are obscure and known only to a few.

So, do you want to wait for an ICANN Godot who never shows up or do you want to get into business now and try to make some money?

Posted by karl at February 7, 2011 3:55 PM