From October 2016 to March 2017 the team is joined by Guest Kats Rosie Burbidge and Eibhlin Vardy, and by InternKats Verónica Rodríguez Arguijo, Tian Lu and Hayleigh Bosher.

Thursday, 5 September 2013

New generic Top Level Domains – who registers them and why?

The IPKat was delighted to receive this post from Emeritus Kat Mark Schweizer.  Thank you very much Mark for this contribution, and the one to come tomorrow!

The new generic Top Level Domains (gTLDS), unpopular with many brand owners, are set to come. The ICANN has received 1,930 applications for new gTLDs (netting the ICANN 1,745 x USD 185,000 = USD 357 million in the process, btw). 1,745 of those passed the initial evaluation, and 19 registry agreements have been signed as of end of August, among them for .clothing, .holdings, .singles, .游戏 (games) and. 企业 (enterprise).

According to the official ICANN “best case” timeline, the sunrise period for the first new gTLD may open as early as 26 October 2013, with name activation beginning on 13 November and open registration from 26 November (there is a tendency for these timelines to be extended).

The list below gives the most prolific applicants, i.e., those who applied for the most new generic TLDs.

Donuts co-founded by Paul Stahura: 307 gTLDs;
Google, Inc.: 103 gTLDs
Top Level Domains Holding: 92 gTLDs;
Amazon, Inc.: 76 new gTLDs;
Famous Four Media: 61 gTLDs;
Uniregistry: 54 gTLDs;
Radix Registry: 31 gTLDs.
United TLD Holdco led by Richard Rosenblatt: 26 gTLDs.

It is interesting to see that leading information technology companies have very different strategies regarding the new gTLDs. While Google and Amazon applied for a host of generic Top Level Domains – among them .home, .book, .how, .baby, .kid, .here, .lol, .fun, .free, .circle, .show and .pay – other large information technology companies such as Microsoft have only applied for a handful of TLDs, most of them trade marks (11 in the case of Microsoft). Apple has only applied for .apple, and Facebook and Twitter for exactly no gTLD. Amazon and Google obviously see a different business case for the generic TLDs, otherwise they would not have spent a combined USD 33 million in application fees alone. This Kat fails to see the business case, but maybe some reader can enlighten me in the comments.

230 strings applied for as new gTLDs are contested, i.e., claimed by more than one applicant. The most popular strings are:

.app (13)
.home (11)
.inc (11)
.art (10)
.blog (9)
.book (9)
.llc (9)
.shop (9)

If the applicants for the same string can’t find an agreement, the ICANN will carry out a “Community Priority Evaluation” if one of the applicants is community based applicant and elects this option. The goal of the Community Priority Evaluation is to determine the most deserving community.

Now, “as a last resort”, if no agreement is reached and no Community Priority Evaluation is carried out, the contested string is auctioned off among the applicants. I am sure the ICANN employee who wrote “as a last resort” was snickering at the time. There are only 67 community based applications out of 1,930 in total. And, as the gTLD Applicant Guidebook (v. 2012-06-04) Module 4 notes, “there is a possibility that significant funding will accrue to ICANN as a result of one or more auctions”. Possibility indeed, given that the string BOOK is contested between Amazon and Google (as is STORE and a number of other generic terms) and HOME between Google and Godaddy, for example.

So somebody is clearly making a lot of money from the introduction of new gTLDs. Brand owners, on the other hand, have to deal with probably a thousand new Top Level Domains in which their brands can be registered as secondary level domains (not all the new gTLDs will be open for registration for anyone and their cybersquatting grandmother, so it is difficult to say how many “open” gTLDs there will be in the end). But surely, the trade mark owners are protected by the Trademark Clearinghouse? Stay tuned for the next post tomorrow.


Clive Bruton said...

I think you're correct in believing that the business case for these TLDs is less than obvious. If we look at the .Apple example, we might get some clues:

Having a TLD means that you can shorten your URL, and become steve@apple and In fact the web server address can be shortened further, to just "apple" (http://apple).

If you can follow the logic there, you can also see that having generic terms like "shop" or "book" (or even "bookshop") could lead to web users just keying-in "book" into the browser address bar, and arriving at an Amazon online bookshop.

So, what were once keywords for search engines now become keywords in browser address requests. I think one can see how this might increase traffic to certain sites.

As to those planning to sell on second level domains within these TLDs, I think we'll see that there will be a few that are successful, and a great many that will fail to gain much traction. Will we see or I doubt it.

Mark Schweizer said...

Note that the "best case" timeline has just been shifted back again. Due to a provision in the registry agreements which provides that the registration must not go live earlier than 120 days after signing, the earliest sunrise periods may be expected to start beginning of 2014.

Anonymous said...


Isn't there a problem inherent in the use of common words - which do not indicate source - and the capture of those common words to a singular source when in an internet environment?

It appears that URLs have more in common with - and must be more tightly controlled than - trademarks.

Whereas a mark may be multiply granted (for different industries, for example), there can be only one location that a browser is directed to on the web.

Mark Schweizer said...

@Clive: Are you sure this is possible? I thought ICANN just recently killed the idea of "dotless domains", see

But I am not the DNS expert, am I.

Anonymous said...

From my own guessing, it seems that the only use of the naked domain (aka dotless) allowed by ICANN is to park it!?

Clive Bruton said...

Anonymous wrote:
>Isn't there a problem inherent in the use of
>common words - which do not indicate source -
>and the capture of those common words to a
>singular source when in an internet environment?

In a word, yes. But what purpose would corporates want to spend millions on TLDs unless it was to gain commercial advantage.

>Whereas a mark may be multiply granted (for
>different industries, for example), there can be
>only one location that a browser is directed to
>on the web.

I understand the point you are trying to make, but URL, even tied to a particular protocol, doesn't necessarily end up in a single point. What I think is probably correct is that that URL/domain name (often) indicates control by a single entity. Where a trade mark might be split between several different vendors.

Although having said that, until fairly recently was controlled by a recruitment company. So, there's some scope for differentiation via TLDs or second level domains (ie,, can all be separate entities).

Clive Bruton said...


Is it technically possible? Well, you've just shown the evidence that proves it is - Google's intention to do just as I describe above. But, the response from ICANN is that "we're going to make a rule that says that you cannot do it".

I hadn't seen that ZDnet article, but you can see the rationale in ICANN's response: browsers divert requests based on user input (that may be incomplete), so if you key "microsoft" most browsers will prepend and append that request with "www." and ".com" respectively.

As well as confusing browsers in this way, they might end up being confused by the DNS system too, since some large corporates use single name ("dotless") URLs for internal servers. ie, in a corporate network the mail server might just be called "mail" and an internal-facing web server "intranet" - so if those generic terms were also available as TLDs then it may be possible to spoof services and cause security problems (though I'm not sure the use of names in that way is really good practice to start with).

My post was based on what one could theoretically do, and my guess about why you'd want a TLD of your own. If Google can't do what it wants here, then I'm not sure I really see the point in this particular marketing exercise.

Clive Bruton said...

It may be worth considering who is registering these new TLDs, which are technology companies, and which of these are in a position to modify the (confusing) behaviour of their browsers.

ie, Microsoft, Google and Apple are registering the TLDs, they all develop browsers, so they are in a position to update the browsers as they see fit. If these companies want dotless URLs, they are going to be fairly persuasive to that end.

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':