The power of the Internet lies in its limitless reach. And that is where some of these Geo TLDs may lose their luster. Yes, you can capture the “flavor” of a city by branding yourself with a dotNYC or dotParis, but most business owners, including myself, wouldn’t consider branding (limiting?) ourselves with anything but dotCom because all other TLDs, including ccTLDs, brand somewhat contrary to the power of the Internet. DotCom is the definitive brand for the Internet because it has no boundries. They don’t call it the dotCom revolution for nothing. Tim Berners-Lee (the father of the WWW) makes a strong case against new TLDs here: http://www.w3.org/DesignIssues/TLD

And as I’ve written on this blog before, we had a powerful experience with Rate.com. Rate.net spent a small fortune on advertising and Rate.com’s traffic immediately spiked nearly 1200%. You and I may be too deep in this business to fully understand all the reasons. The bottom line is that the herd thinks differently, they react differently and they will always have the final say about our business.

While you are correct about a successful brand being memorable, there are (at least) two aspects to memory that are useful for this discussion — recognition and reproduction. Recognition means that someone can identify the brand if it’s put in front of them. Reproduction means that they can recreate (by speaking, typing, whatever) the brand entirely from memory. Reproduction is much much harder than recognition, as anyone who has taken a language test in high school or college can attest — it’s a lot easier to read French than to speak it.

A user looking at a search engine result page or a link only has to recognize a brand in order to click — here the advantage of the “default” .com is negligible, especially because the search result page or the website containing the link will provide lots of context and other clues about what the “correct” choice is.

A user reproducing a brand by typing it in to a browser has a much harder task. They are much more likely to get it wrong and end up with a misspelling or a misremembering — which can easily include the TLD extension.

In either case, a very high percentage of people get it “right.” In the case of recognizing a brand and clicking it, the percentage is very high; in the case of type-in navigation, user error is higher, but certainly not even close to a majority, especially given that so much of type-in traffic is either to a frequently-used site where a mistake is immediately recognized and corrected (which I would not count as a navigation) or is from a bookmark.

So, if 15% of all traffic is type-in and the rest from search, and mistaken navigation is rare in search and not a majority from type-ins, then the actual % of mistaken navigation is much less than 15%, and will decline over time as the default position of .com declines.

This is clearly already the case in Europe, Asia, and Canada, where multiple TLDs are the norm; there is no reason to suppose it will be any different in the United States. That’s still a lot of mistaken traffic, and owners of .com names will still benefit, but as percentage it’s small, and certainly not enough to outweigh the benefits of a great name to the left of the dot, compared to a mediocre name with a .com at the end.

Truthfully, if you were registering a new name, would you rather own hotels.nyc or gothamlodging.com?

Also, my statement about Sean Miller was not based upon intuitive traffic. People get hung up on search engines as the proof positive about successful branding and the truth is that search engines have absolutely nothing to do with branding. Successful branding is based 100% upon the viewer’s or listener’s ability to remember (memory) and the number of people who will see or hear an advertisement for a site that ends in dotNYC and then accidentally default to dotCom when they look it up will be much, much higher than 15%.

The figures I’ve seen suggest that about 15% of Internet traffic is type-in/bookmark traffic. That’s a very respectable number, but not nearly a majority. A much larger proportion comes from search-engine result pages. So if Google determines that .NYC is meaningful for searches where “new york” is a search term, then that will drive traffic to .NYC sites.

I don’t disagree that in the early days at least, new TLDs are going to have very little typed-in traffic. So, if someone were to suggest that a new TLD needs to live or die on typed-in traffic, I would agree that it’s not a very good idea. But the vast majority of Internet traffic comes from other sources, and although it’s painful to lose up to 15% of traffic to someone else, that number will come down over time. Also, although I may make a mistake from time to time (e.g. craigslist.com instead of .org), I usually know that I have, and re-navigate to find the correct site. So a site owner using a new TLD won’t “lose” all of the type-in traffic.

New TLDs are going to re-introduce semantic meaning to extensions. .COM, .INFO, and .BIZ mean nothing. .NET used to mean something, but doesn’t any more. .ORG still retains meaning as a place for not-for-profits. .NYC will mean New York City, and that will aid user comprehension and reduce confusion.

There is a panel on this at the GeoDomainExpo from April 23-25 in San Diego, where I have the distinct privilege of participating on a panel with David Costello and Phil Corwin.

Candidly, after spending time following all of this, there’s a lot of opportunity on all sides without speaking ill of the new or established.

We’ll have a good session, and I am sure that a lot of the things on this thread will be discussed and more.

See you at GeoDomainExpo.

Antony, here’s some “empirical evidence” for you.

We own Rate.com. In 2003 Rate.com suddenly went through the roof in traffic – an immediate jump of over 1200%. My brother and I were stumped until we saw an advertisement for a huge new financial site called Rate.net.

In 2004 LagunaBeach.com jumped overnight from 2,500 uniques a day to 11,000. We discovered that the night before MTV launched the show Laguna Beach (and LagunaBeach.com was # 3 on the search engines – the sites above us did not show an increase in traffic). MTV was strongly promoting their own web site for the show, but a lot of them simply assumed it was LagunaBeach.com.

And let’s not forget Dick Cheney during the VP debates when he told viewers to go to FactCheck.com – instead of FactCheck.org.

There are countless other times I’ve watched newscasters give the wrong address for a site because they assumed it was dotCom. I could go on and on, but it would be (sorry, can’t resist) silly.

I never say never, but I assure you that Sean Miller of NYC.com will be loving the massive accidental traffic he’ll receive from dotNYC.

That said, however…

“People talk about how users automatically default to using .com, and how that will never change, but frankly that’s just silly. No-one has ever produced any empirical evidence to support that idea.”

Well…other than the fact that it’s actually been shown for many years now (sorry I don’t have the studies in front of me) that that is exactly what people do: Take the product/service type/category they’re looking for (or the name of the specific company/entity if they know it) and add a “.com” to the end to see what shows up.

“…everyone used a radio, and they still do, but it didn’t take long to switch to television. Everyone used 8-track tapes, until cassettes came along, and they were displaced by compact disks.”

These types of comparisons are misplaced. A new TLD is more akin to a new language, or in the least a new “form” or variation of an existing language than to a radio or cassette.

Far, far more difficult (and expensive) to get people to speak (or act) differently than it is to get them to change from what they’re used to, comfortable with, and trust.

“If there’s a benefit, people will switch.”

These kinds of “if it’s better, they will come” statements have been loudly trumpeted by all or virtually all the leaders of the various “not-com” extensions.

Yet .com remains the world (and to an even higher degree here in the largest economy in the world) leader.

Furthermore, exactly what makes .nyc; or any new TLD for that matter, “better” (for any purpose or use) than .com?

Saying so doesn’t make it so.

“I’m completely convinced that this is an inflection point in the Internet, and that there are green fields, and great rewards, for people with imaginations and a few dollars to spend.”

For no less than the reasons I’ve given (and there are others), these new TLDs (including .nyc and .eco) will prove to be…not green fields with great rewards…but scorched earth financial and emotional wastelands for any but the largest of companies…willing to suffer the huge monetary losses these TLDs will slice from their income & profits…for as long as they own them.

Is Icann restricting the minimum number of characters allowed in a gTLD?

I’m going to guess they have to be 3 characters or longer.

Are there any restrictions on not using certain characters or numbers?

thanks.

Josh

Will be interesting a few years from now if we will see gtlds sold at domainer auctions like we see individual names now…

That being said, I see this as being a boon to dotCom. As we all know, Corporate America is not known for taking chances and the moment one of them trips badly on their own TLD the rest will embrace dotCom as the “safe” Gold Standard for Internet Marketing.