New TLDs

What Does “Demand” Mean to You?


At the ICANN meeting in Seoul, the ICANN board decided to push back the roll out of new gTLDs to at least after the second quarter of this year.  A few weeks later, ICANN opened up a public comment period soliciting feedback on the possibility of creating a type of pre-application program where parties could express their interest in applying for new gTLDs.  In December the Corporation released a draft model for the “Expression of Interest” (EOI), the proposed gTLD pre-registration period, and accepted comments on the draft through the end of January.  As proposed, the EOI requires interested parties to indicate their intentions to apply for new gTLDs and pay a fee of $55,000. Additionally, only those parties who participate in the EOI will be eligible to officially apply for new gTLDs in the first round.
 
According to ICANN, the motivating factor behind developing the EOI program is to measure demand for new gTLDs.  But there’s one big problem – because participation in the EOI is required in order to apply in the first round of gTLDs, the demand measured by the EOI will be artificially inflated by those parties who are unsure of whether or not they really want a new gTLD but don’t want to be locked out of the first round just in case.  (There is a whole host of other problems with the EOI as well, including the fact that it would be introduced before the gTLD Applicant Guidebook is even close to being finished.)
 
Another problem is the “demand” ICANN and others refer to is confusing.  Any EOI results would not signal Internet user demand, but demand among groups that are seeking to possess top level domains either for private use (a company’s .BRAND) or to sell domain names to the public (a gTLD entrepreneur of any kind that wants to sell sunrise and live registrations to businesses and the general public). If an intended purpose of introducing new gTLDs is to offer the public more choices, isn’t the public’s demand for those choices the real “demand” that ICANN should be measuring?
 
I was reminded of this issue when I received a periodic newsletter from one of the corporate-facing domain registrars.  The newsletter mentioned the EOI as a means of measuring “demand for new gTLDs.”  First off, as I just said, it clearly is not an effectual means of measuring meaningful demand. But as I read it, I started thinking about how much it behooves domain registration companies, as well as ICANN, to say that the EOI is measuring demand.  If the so-called “demand” is high enough (and who will determine if it’s high enough?  That’s right, ICANN), then ICANN will be justified in pushing forward with the rollout of new gTLDs without determining if Internet users really want them.

LIMRA Points Out New gTLD Risks


Ken Hittel, Vice President of the Corporate Internet Department at New York Life Insurance Company (a member of CADNA), recently sent me an article written by Todd Silverhart, Corporate Vice President and Director at LIMRA (Life Insurance and Market Research Association, a worldwide association of insurance and financial services companies.) Silverhart writes a column on technology developments in the financial services industry for LIMRA’s MarketFacts, a quarterly publication. Recently, he shifted his focus slightly and addressed the topic of the expansion of global Top-Level Domains (gTLDs) and its effect on brands.
 
The shift in the subject of the column was prompted by a meeting of LIMRA’s E-Business Study Group where a participant mentioned ICANN’s plan to extend the TLD space.  According to Silverhart, the majority of study group members had heard nothing about the plan prior to the meeting.  He found this lack of awareness surprising, given the potentially dramatic impact the change will have on how companies manage their Web sites and overall Internet strategy.  Silverhart goes on to outline the costs companies will incur in defensively registering key domain names in new gTLDs and brings up the controversial question of who will get to control the gTLD for entire industries like .LIFEINSURANCE or .ANNUITIES.  He also mentions the consumer behavior aspect of new gTLDs, and whether Internet users will actually use extensions other than existing ones like .COM. It’s a solid analysis with some great points; I recommend that you check it out.
 
At the end of the column appears a note from the editor that mentions CADNA as a useful resource for those interested in taking action regarding new gTLDs.  CADNA has done a great deal in addressing the issue of new gTLDs, including engaging repeatedly with ICANN, U.S. Department of Commerce, U.S. Congress, and foreign governments to educate them on the risks associated with the expansion, and it was great to see CADNA’s leadership recognized. 

Crying Wolf


It is no surprise that brand owners have registered countless domain names that they don’t need.  Over the past few years, ICANN approved and released TLDs such as .EU, .INFO, and others.  Since there was a lack of real data, brand owners did the only thing they could, which was to register names defensively because of the threat of what might happen.  When .ASIA was released, we saw the number of registrations from brand owners begin to drop – likely a result of being fed up with registrar/registry profit driven domain policies.

The problem is that these prior new TLDs have rightly caused brand owners to be suspect of new launches. After so many sunrise periods that hyped the need for defensive domain name registrations, now it's like the boy who cried wolf- many launches are largely ignored. However, every now and again an important TLD change occurs that does necessitate action by brand owners.  One example of that is .CM.

This landscape of the .CM TLD was just recently formed. When FairWinds published a “Perspectives” on .CM cybersquatting back in 2006, the Cameroonian government was running the .CM registry.  At that time, there were only 200 domain names registered to .CM and all other names ending in .CM resolved to a PPC site.  NETCOM.cm Sari took over control of the registry in 2009 and opened registration of domains in .CM. Trademark owners were allowed to apply for their corresponding domains during a one-month sunrise period between June 14 and July 15 of 2009. Sunrise periods allow those with valid trademarks to register their domain name for a hefty sum before all domains become available to the general public. As of August 1, 2009, .CM domains can be registered on a first come, first serve basis regardless of trademark rights.

Because .CM, the ccTLD for Cameroon, is a very common typo error of .COM, its recent launch of unrestricted domain names opened up a new front on the war against cybersquatters.  As with typosquatting of domain name roots (as in “comcasft.com” for “comcast.com”), this form of typosquatting also involves third parties registering domains containing brand names with the intent of profiting off of users’ typing errors. 

In light of this change, we conducted a study to examine the actual response to .CM’s change.  We drew on the data used for a forthcoming study on general typosquatting to examine this phenomenon.  We began with Quantcast’s list of the most highly trafficked domain names, and from that list, we selected the top 250 that ended in .COM and whose root contained more than six characters.  That initial number expanded to 255 in order to include possible hyphenated versions (i.e., wal-mart.com and walmart.com).  Along with these 255, we included typo variations that receive high volumes of traffic and then checked to see whether each of these domains had been registered in the .CM extension.  A total of 183 had been registered in .CM; 121 domains contained the target root and the remaining 62 were typo variations of those targets. 

Out of the 183 domains, an astounding 97 percent are owned by a third party—only 6 domain names are owned by the target company.

                                         

Of those owned by the target company, only 4 resolve to the target site, while one displays search results and the last does not resolve.  In total, 97 of the domains owned by a third party lead to pay-per-click sites, meaning cybersquatters are directly profiting off 55% of those domains. 

Another interesting tidbit was found in four domains that led to sites for competing brands.  Staples.cm resolves to OfficeDepot.com, Travelocity.cm resolves to Expedia.com, Walgreens.cm resolves to DrugStore.com and Walmart.cm resolves to eBay.com.  None of these four domains are owned by the competitor, but rather by third parties.  While cybersquatters are not profiting directly from these sites, they are still causing damage to the brands in question and potentially depriving them of sales. 

When ICANN approved new TLDs in the past and registries held sunrise periods for brand owners, many brands rushed out to buy domain names in TLDs like .PRO and .INFO only to realize later that they held very little value for their business.  Although .CM has opened to the public, the registration rates among brand owners remain low. This may be a result of brand owners being skeptical of the value of a .CM domain name. Some may be failing to realize how different this new TLD is—after all, it is common for Internet users to accidentally type .CM in place of .COM. Others may be distrustful of a new registry sunrise period given that ICANN’s policy of allowing registries to operate however they want has essentially resulted in brand domain names being held for ransom by the registry in the form of sunrise periods. In the case of .CM though, brand owners must learn that even if they have no intention of doing business in Cameroon, owning key names in that extension is critical because users are typing them into the browser bar accidently.

What I take from this situation is that while we are all growing incredibly frustrated with the name space and ICANN policies, we need to be diligent in our review of each change.  It is usually best to be conservative with launches, but there are cases – like .CM – where a more aggressive and targeted strategy is warranted. 

Game, Set, Domain Name Match


The US Open championship is in full swing in New York City right now, with the finals just around the corner this weekend.  As one of the four events that make up the Grand Slam, one of the highest achievements in tennis, the US Open is always hugely popular.  When I searched for the official Web site of the Open to find out how Roger Federer is doing, I typed “usopen.com” into my browser’s address bar.  Indeed, I arrived at a page for the U.S. Open – but for golf, not tennis.

As it turns out, the official site for the US Open for tennis is usopen.org.  The United States Golf Association (USGA) owns the domain name in the .COM, while the United States Tennis Association (USTA) owns the domain name in the .ORG.  The trademark “US Open” is registered to the USTA, while the trademark “U.S. Open” is registered to the USGA.  The only real difference between the two marks is the use of punctuation in “US,” which does not transfer well into forming distinct domain names.

I did a little digging into what other domains users might search in hopes of accessing the site for either the golf or tennis championship.  First I tried usopengolf.com and usopentennis.com.  The former does not resolve, while the latter leads to a site that is not affiliated with the USTA but hosts links to buy U.S. Open tickets and merchandise.  Then I searched theusopen.com and theusopen.org.  A pay-per-click site and one that doesn’t resolve, but is owned by a Florida ticket broker.  Finally, I typed in usopen.net.  Interestingly, this last domain leads to a pay-per-click site that happens to be for sale.  Argh, do I dare say it?  What if there was .tennis and .golf – could they make life simpler – usopen.golf and usopen.tennis – or am I just asking for even more trouble? According to a recent article I read, .Sports would argue that these TLDs shouldn’t exist, since they belong under a general sports umbrella.

Attempts to set up TLDs like this in the past have kind of gone bust. Last time I checked the major aerospace company names in .aero, most did’t even resolve.  One domain, Airbus.aero, promotes the business of MelbourneIT, a registrar. 

Never a simple answer. 

Getting Out in Front


The Progress and Freedom Foundation recently published a paper titled “New gTLDs: Let the Gaming Begin” that described the numerous efforts of different organizations to preempt the gTLD application process by registering gTLD strings as trademarks with patent offices both here in the U.S. and abroad.  The paper calls on ICANN to slow down and analyze such actions that clearly provide a short-term financial benefit for the applicants at the expense of trademark holders and registrants. 

These subversive actions qualify as gTLD front running.  Some of you may have heard the term “domain name front running” before.  It refers to the act of using some form of insider information about an Internet user's preference for registering a domain name to preemptively register that domain name.  The “insider information” is gathered by monitoring the attempts of an Internet user to check the availability of a domain name. The term first arose in connection with allegations that ICANN-accredited registrars were using insider information to register domain names in advance of the general public and then using them as PPC sites or reselling them at a higher value.  Similarly, gTLD front running refers to the attempt to gain a leg up on the competition by securing the control of a gTLD before others are able to.  In this case, groups are trying to circumvent the ICANN process by registering gTLD strings as trademarks in an effort to establish their “rights” to those gTLDs. 
Because there is so much more at stake with gTLDs than single domain names, both monetarily and operationally, gTLD front running has even greater consequences for the domain name space than domain name front running.  Interestingly enough, last week ICANN released the results of a study it had commissioned to assess how often domain name front running occurs.  The study examined 600 domains over a ten-month period and found no evidence of front running.
I have to wonder why ICANN would devote its resources to commissioning this study when it has already published papers on domain name front running in 2007 and 2008 and a more complex issue with a greater impact on Internet users has emerged.  Back in June, I blogged about how ICANN’s somewhat limited WHOIS study, while well-intentioned, came too late in the game to provide a meaningful answer to the problem in question.  ICANN needs to address the issue of gTLD front running before it too becomes too big to handle.   

Getting the Facts Straight


A recent Slate article titled “www.thosenewdomainnames.areforsuckers,” caught my attention last week.  With all the controversy over ICANN’s planned rollout of a potentially unlimited number of new generic top-level domains (TLDs), I found the frankness of the title refreshing.  As someone who works closely with brand owners, both at FairWinds and CADNA, I have been wary of the prospect of opening up the domain name space in such a radical way.  The article explains ICANN’s plan for launching new TLDs with what I thought was an appropriately critical tone, questioning whether or not there actually exists a need for new TLDs.  But as the article progressed, I realized that the reporter’s reasoning was completely off base.  He actually asserts that cybersquatting is no longer a problem, and that the new TLD program is pointless because “domain names themselves just don't matter that much nowadays.” He claims that most people use search bars to navigate the Internet, and that browsers are now sophisticated enough to be able to “tell” what site the user is looking for, citing Google’s Chrome in particular.
 
FairWinds has conducted studies that prove that all users, at some point or another, use direct navigation to browse the Internet. Domains do matter, because eventually all users type a domain directly into the address bar in hopes of finding the content they are seeking.  Furthermore, nearly all print, radio and TV advertising directs consumers to a Web site, not to Google.  Advertising supports users going directly to a Web site, and both user behavior and the enormous traffic to millions of typo- and combo-squatted domains demonstrate it.  After all, the vast majority of the tens of millions of cybersquatted domains exist to court type-in navigators.
 
Cybersquatters understand advertising and user behavior very well, which explains why cybersquatting has continued to increase every year.  This article failed to recognize that fact, and also failed to realize how likely it is that opening up with domain name space with unlimited TLDs will exacerbate the problem.  Instead, the author claimed that cybersquatters have moved on to social networking sites, insisting that getting one’s identity on Facebook and Twitter is much more important than getting a good domain.  This may be true for individuals, but for businesses, a Twitter account or a Facebook page does not even come close to the legitimacy of a good domain name.  
 
Consumers expect to be able to find a business online via its own web page, not to have to sift through social network pages to interact with the brand. And by making a slew of new TLDs available, it will be even harder for consumers to quickly access the brands they seek.  If it’s so hard for consumers to remember domains now, as the article attests, how will they possibly remember – across all the products and services they seek out online – if the domain is brand.com, city.brand, product.brand, brand.brand, etc…?
 
Even though it began so promisingly, this article completely missed the mark on cybersquatting, how domain names are used, and the expansion of the domain name space.  These are issues of concern not just for businesses, but ones that will ultimately affect the online experience for everyone, particularly for consumers and everyday users.  With so many conflicting interests at stake, it is critical to have accurate information about these issues.  

ICANN’s Meeting Down Under


A quick report of some of the most important developments coming out of ICANN’s meeting Sydney:

Despite concerns over the way in which ICANN is moving forward with the new TLD launch, it appears that ICANN is poised to start accepting applications for new TLDs in February 2010. If an application is accepted, it is predicted that it will take about 7.5 months before the TLD can be used.

At the meeting, there has been talk of applications for new TLDs such as .MUSIC, .BERLIN, .WEB, .LOVE, .RADIO, .NYC, .BASKETBALL, .ECO, and .FOOD. Wolfgang Puck was even flown out here to promote .FOOD, which was covered by National Journal: Celebs Endorse ICANN Domain Plan.  Even Sydney’s mayor announced that he is looking to get .SYDNEY. Seeing that there is some interest in new TLDs, it is discouraging that ICANN is not slowing down to make sure that if TLDs are introduced, they are done so in a way that does not jeopardize the safety and stability of the Internet. The TLDs should also add value to the space.

There are still questions hanging over the TLD launch process that should be addressed before any applications are accepted. For example ICANN’s new TLD policy stated that a TLD must be a least three characters. How will they implement this guideline with IDNs (Internationalized Domain Names)? “Characters” function differently when using non-ASCII characters. Does the guideline need adaptation? Furthermore, while IRT discussions are very active here in Sydney, the board canceled their IRT briefing meeting. Since the IRT report concerned itself with protection mechanisms, questions remain as to how they will adapt the report to recommendations offered by the Internet community.

While new TLDs remain the hot topic at the ICANN meeting, other serious policies are also being discussed. ICANN has publicly shown that it is committed to ending the Joint Project Agreement (JPA) that exists between it and the US Government, which is set to expire September 30, 2009.

The meeting has also touched upon the idea of vertical integration, which would allow registrars to become registries and could have major consequences on Internet users and on market performance as competing registrars may resist offering TLDs sponsored by their competitors.

Overall, the policies being discussed at ICANN all address complex issues that require thoughtful and thorough consideration. However, as Steve DeBianco wrote in his own blog entry, “ICANN often deems that meeting its own deadlines is more important than achieving real consensus on serious issues.”

Mistaken Identity


Anybody who has had a friend sneak onto his or her Facebook profile and change the information knows the heavy reputational damage that can come from being misrepresented on a social media site.

But this harmless prank has a much nastier relative: Userquatting.  Similar to cybersquatting, usersquatting is when Internet users register usernames on popular websites to mislead others as to the true identity behind the username.  With the openness and ease of use of most social media sites, usersquatters can impersonate anyone from your next-door neighbor the CEO of a major company to the President of the United States.  In fact, during the 2008 campaign, a member of the RNC reportedly set up a fake Facebook page in Barack Obama’s name. 

Now that many companies have realized the utility of establishing a presence on social media sites, consumers have become accustomed to seeing different brands sending out tweets on Twitter or setting up Facebook and MySpace pages.  And unless the usersquatter is engaging in blatantly outlandish behavior, there is little to tip off other users that these accounts are being operated by anyone other than a brand representative.  The National Arbitration Forum did an analysis of the top 100 global brands on and concluded that out of those that did have a presence on Twitter, only nine are controlled by the actual company, while 27 are controlled by individuals that likely have no affiliation with the company. 

As of this weekend, Facebook began offering vanity URLs to its users, much in the same way MySpace did a while back.  A New York Times article reported that since the release late last Friday night, close to 6 million users have registered vanity URLs.  But not all of them chose to register their own names.  Some chose names of celebrities, like “facebook.com/snoopdog” and “facebook.com/georgebush,” while others opted for a more tongue-in-cheek approach: one user scooped up “facebook.com/twitterisbetter.”  The article also mentioned that Assetize, a marketplace where members can sell their online accounts, is auctioning such URLs as “facebook.com/iphones” and “facebook.com/hpcomputers.” 

One of the biggest problems behind this issue, in addition to the damage that it can inflict on brand equity, is the fact that enforcement for usersquatting is generally ambiguous and not uniform across different sites. In another New York Times article, Howard H. Weller, trademark lawyer at Mitchell Silberberg & Knupp in New York, commented that “these are all new avenues for abuse, and it’s more resources trademark owners need to devote to policing and enforcement.” And since much of the value of the vanity URL is speculative, brands only time will tell if the scramble is worth it.

 

Bringing in the Big Guns


As the debate over new TLDs rages on, more and more people are joining in the discussion and offering their two cents.  Just the other day, former New York City mayor Ed Koch announced that he is behind the .nyc top-level domain.  He went so far as to call .nyc “the best real estate opportunity since the Dutch bought Manhattan,” and asserted that he could not wait to register edkoch.nyc and mayorkoch.nyc.  Koch even noted that .nyc would create jobs and provide much needed money for city programs, a message that has a special resonance in this economic climate. 

Back in March, I wrote a blog post on a similar topic, about Al Gore’s endorsement of a .eco TLD.  Even the Pope weighed in on this issue, arguing that dot-RELIGION domains like .catholic or .buddhist will lead to bitter disputes over who has the right to each of these TLDs. 

The important thing to remember when it comes to this issue, no matter who comes out for or against the new TLDs or how famous that person is, is that the debate is still fraught with uncertainty and disagreement. 

ARE YOU LISTENING, ICANN?


ICANN claims that the roll out of new TLDs is designed to encourage healthy competition in the domain name space, but the organization has repeatedly failed to provide evidence that substantiates these claims. Furthermore, ICANN has not put forth any evidence to suggest that the new TLDs will be even somewhat beneficial to the wide Internet community. I regularly work on these issues and have read countless documents and letters pertaining to the TLD launch.  There are two documents from a few months back that are of particular interest. The first is a letter from the Department of Commerce to Peter Dengate-Thrush, the chairman of ICANN’s Board of Directors. The second is a letter from the Department of Justice to the National Telecommunications and Information Administration (NTIA). Both letters provide a laundry list of areas where ICANN needs improvement, better foresight, and an overall reality check.
 
The DOC’s letter states that while they acknowledge “the effort and hard work involved in producing the documents currently out for comment, it is unclear that the threshold question of whether the potential consumer benefits outweigh the potential costs has been adequately addressed and determined.” One of their recommendations is that ICANN must “ensure that the introduction of a potentially large number of gTLDs, including internationalized top level domains, will not jeopardize the stability and security of DNS.” Given the rampant cybersquatting that occurs among the current TLDs, and the amount on .com alone, I don’t know how ICANN will justify introducing even more TLDs without first addressing the existing problems. 
 
Then there’s the issue of competition. The letter from the DOJ points out that the overwhelming popularity of .com is far too valuable to ever actually be constrained by new TLDs. Even in the current system, most domain owners register other TLDs in order to augment their Internet presence, not as a substitute for a .com name. Both letters implore ICANN to reevaluate the demand for new gTLDs and to better account for the interests of consumers. The brand owners and individuals who register domain names are, in this case, the consumer. Yet ICANN seems to be moving full steam ahead toward this initiative. If ICANN is not listening to consumers or to the government that acts as its advisor, who is ICANN actually listening to? With the DOC joint project agreement scheduled to terminate this September, what is the likelihood that ICANN will be listening to anyone in the future?