General manager of public participation at ICANN, Kieren McCarthy, made a post about new generic top-level domains on ICANN’s blog. He asks a good question, namely, what new generic TLDs would be useful and successful as the Internet continues to evolve?

Billions of dollars have been spent on promoting the DotCom brand, so I don’t see how a new gTLD could become as popular as .com anytime in the near future. However, a good gTLD could become the next .net, in my opinion. But what will make a new TLD successful? One gTLD I like is .info, although I must admit I don’t own many .info domain names. It’s a TLD that sort of makes sense to me, because information is what most people are looking for on the Internet. So .info is a very good extension for a Wikipedia-like website. The introduction of .biz, on the other hand, did not make any sense to me, as there has already been a TLD for businesses, which is good old .com. Anyway, none of these newer gTLDs have really been successful so far. DotCom, DotNet and DotOrg are just too big, aren’t they?

Sponsored TLD .mobi saw some good sales shortly after it was introduced, but one hasn’t heard much about it since then. I know, there are lots of domain investors who have spent thousands of dollars on .mobi domains, in the hope of reselling the domains for a big profit. Although it’s true that many people have made money from .mobi by successfully reselling these domains already, I think there is no need for a .mobi TLD, because it’s possible to visit any .com website via modern mobile phones and pocket PCs. Mobile phones such as the iPhone can display normal websites, so special website formats for mobile devices are not required. Therefore, I don’t see any successful future for .mobi. DotCom is just as good a TLD for mobile Internet surfing. The iPhone even has a .com button, which is just another example of the massive branding machine behind the world’s most successful top-level domain!

All of this only shows that new gTLDs have failed in the past, and this doesn’t really answer the question which new top-level domains people would adopt.

In his article, Kieren McCarthy looks at past TLD introductions too:

• .firm: for businesses, or firms
• .store: for businesses offering goods to purchase
• .web: for entities emphasizing activities related to the WWW
• .arts: for entities emphasizing cultural and entertainment activities
• .rec: for entities emphasizing recreation/entertainment activities
• .info: for entities providing information services
• .nom: for those wishing individual or personal nomenclature

This list is interesting for a number of reasons. For one, it is striking that the new gTLDs first approved follow the same pattern as outlined in the report, but without the actual same strings being approved. For example, in the first round, announced November 2000, there were also seven new TLDs.

The .firm and possibly .store TLDs were dealt with through the .biz TLD. The .web TLD has had a long and chequered history and just didn’t happen. The .arts became .museum; .rec was presumably left to dotcom because of the dotcom boom; .info was approved; and .nom (the only TLD that was not patently English) became .name (which patently is). Which left the TLDs that still seem unusual today: .aero, .coop and .pro. It is intriguing that expansion of the Internet was viewed then as an attempt to reflect society in a very sociological, almost paternal, manner: we have business, we have culture, recreation and then personal space.

The next round of approvals in the gTLD space (March 2004) were sponsored, implying an effort to get away from the massive boon in the commercial nature of the Internet, but even so they reflected a commercial edge: .jobs, .mobi, .tel, .travel. There came appreciation of the global nature of the Internet with the approval of .asia and .cat. And the thorny issue of one of the Net’s biggest benefactors – pornography – came in the form of the (rejected) .xxx top-level domain.

Now, what might be the new generic top-level domains of 2008? And which of these TLDs would have the potential to become popular among the majority of Internet users? As with every controversial subject, there are different schools of thought, of course.

One possibility would be the introduction of typo TLDs. Yes, TLDs like .XOM, .CPM and .NT. But why register such TLDs? Well, such domain extensions would be crucial to own for any company that wants to protect its trademarks on the Internet. As of now, domain extension typos are being wrongfully monetized by search engine giants Microsoft and Google. MSFT and GOOG are making millions from these domain name typos! Wouldn’t it be better to let companies register typo TLDs, or why not introduce a global wild card that will automatically forward .XOM and .CPM typos to .COM, .NT to .NET, etc.?

Another possibility is that big Internet companies like Google will launch their own top-level domains. For example, .Google or .GOOG. They could then forward these domains to services they offer. Freemail, web hosting or subdomains, for instance. Google could give away website packages with an email address, web space and a .GOOG domain name for free or at a small annual fee. I’m sure such an approach could come closer to challenging .com than other new gTLDs have come. Google has millions to spend on branding their company’s TLD and they can feature it for free on their existing services. In addition, Google has skilled people and a great infrastructure to offer a .GOOG TLD. Thinking outside the box here, but why not?

A generic TLD I did not like was .TV, but I think this is a promising TLD since Demand Media took it on from VeriSign earlier this year. Demand Media is actively promoting this TLD, and it is offering tools for registrants to easily launch their own online TV channels. That’s a good idea and it might reanimate .TV. The future will tell.

These are the thoughts I had about new generic top-level domains this morning. What do you think? Can new gTLDs still be a success or is .com too big for any other new domain extension to prosper? Which generic TLD would you like to see, or more importantly, which new generic TLD would you register?