Structure: New Top Level Domains, Part 1
After years of wrangling, flame wars, and lawsuits, new top level domains were finally implemented in 2001. Here are my comments about them. I judge them by how much of a useful addition they make to the namespace from the standpoint of Internet developers and users (both commercial and noncommercial, individual and organizational), not by how much money they make (or lose) for registries, registrars, speculators, cybersquatters, marketing types, and lawyers. See more discussion at: ICANNWatch, ICANN.Blog, CircleID, Robert Shaw's Weblog, GrepLaw, DNForum, Slashdot, icann.org (new TLD agreements) [Closed], icann.org (new TLD evaluation), icann.org (.info country names) [Closed], and new.net.
As of 2006, the sponsored TLDs are up for renewal of their ICANN contracts; see here for more information and a chance to comment.
Somebody is selling a 4-page PDF document on new TLD registration policies for the outrageously gouging price of $195 (US). And this isn't the most outrageous ripoff in domain-name-related electronic publishing; somebody else has a 19-page document on "Domain Names as a Foundation for Internet-Based Identitites and Branding" for $995. Since those things are available on Amazon.com, and I've signed up with their Associates Program, I'd get a pretty nice commission if you'd follow those links and buy that crap, but I like to think my readers aren't that stupid... you can find out all you need to know about the new TLDs and their policies for free by reading this page and the other sites I link to from it, and other sections of this site will give you some tips about using domain names as a foundation for whatever Internet-based stuff you might be doing, whether or not it involves "Identities" or "Branding" or whatever other buzzwords are popular with the marketroids. So save your money! (I'll never make it as a Marketing Type...) (None of this is intended to disparage Amazon.com or its fine Associates Program... they've got lots of great books at competitive prices, and if you go there from links in my site and buy stuff you're helping to support my site. But still... $195 for a 4-page PDF file with the same info you can find for free on the Web???)
I have both positive and negative things to say about the new TLDs; I've got lots of opinions, but I try to be evenhanded rather than just adding to the ranks of either the marketing-hype promoters or the bellyaching bashers. Those in charge of launching new TLDs have a rather thankless job; even if they did it with Solomonic wisdom and the ethics of an angel, they'd still get bashed for it, as there are a whole array of different groups out there with conflicting interests and large stakes (both financial and emotional) in the outcome of the process. Whatever pleases the trademark lobby will displease the domain speculators and vice versa, and both may be detrimental to those who just want to use a new domain name rather than speculate in it or protect a brand name with it. Thus, no matter what policy is made and how well or poorly it's administered, somebody is bound to complain, and perhaps even file a lawsuit about it. Criticisms in various forums (including mine!) need to be read in this light; pay careful attention to who is making the criticisms, what vested interests and prejudices they may have, and how much of their criticism is based on verified facts rather than exaggerations and rumors. Some will find something to attack no matter what -- this was seen in the days following the September 11 attacks, when .info and .biz adjusted their launch schedules to accommodate New Yorkers facing hardship meeting the original schedule. Forums were full of messages attacking these registries for their insensitivity in not delaying their schedules at first -- and once they made the change, there were new bursts of messages attacking them for making such a last-minute change. Damned if you do, damned if you don't... That said, I think there are some facts about the launch process that deserve criticism -- and others that deserve praise. So here I go with my commentary...
Here they are... the new TLDs!
.info -- for information-based services
.info is for information-based services, whether commercial or noncommercial. It is run by Afilias, a consortium of registrars chartered in Ireland but having its offices in Pennsylvania. My first impression was that it was too vague and meaningless to be a useful namespace addition, but it's grown on me since then; it may be ideal for informational sites where the type of organization behind them -- commercial, noncommercial, governmental, etc. -- is less important than the fact that it's a site with info about something. It's also useful as a TLD that's chartered explicitly from the outset as a free-for-all to be used by anybody for any purpose -- just what .com is (ab)used as, but it's more open and honest to have a domain that's set up that way explicitly. I just wonder whether abortion.info will end up going to a pro-choice or a pro-life group? (The answer so far is "neither"... whoever's holding it doesn't seem to be doing anything in particular with it.)
.info started live registration on Oct. 1, 2001 -- and then temporarily stopped it for "maintenance" on Oct. 2. Confusion reigned, as the landrush names were supposed to go live on Sept. 27, but were delayed a few days, and some pre-registered names from some registrars never did turn up at all. Registrars took their time at informing the successful landrush registrants that their names were registered, and giving them an interface to add name servers to them so they can be used. Some reserved names got mistakenly registered to people, and then taken away. Sometimes multiple registrars took money for the same name during the so-called "real time" registrations, since the WHOIS server wasn't updating very rapidly. Messages appeared on forums from people accusing Afilias or registrars of "stealing" their domain names due to various cases of their telling people that their names were successfully registered, sometimes even showing them in the WHOIS record, then later awarding the name to somebody else or showing it as a reserved name unavailable to anybody -- however, the old saw that you shouldn't attribute to malice what can be adequately explained by incompetence probably applies here. Afilias certainly didn't give a very good impression of technical competence during this launch, despite their self-congratulatory press releases. After the initial glitches, real-time registrations eventually functioned, and the .info WHOIS server went live.
A big event for .info speculators, and those interested in actually using good .info names alike, occurred in July, 2002 -- following the conclusion of the sunrise challenges, including the last-resort challenges placed by Afilias themselves, a large number of desirable generic names were freed up for a "Landrush II", and many people hoped that at least some of them would end up in the hands of registrants who would actually use them. I managed to get dan.info in this landrush, and I'm now using it as the primary address for my informational sites.
.info eventually reached the milestone of its millionth registration. Unfortunately, this "key" name is the godawfully-messy los-angeles-real-estate-search.info.
See my comments on the sunrise rules, and other aspects of the launch and challenge process, below.
Perhaps the most notable registrant is New York City's MTA, at mta.info, which reports a great increase in Web traffic since adopting and promoting this address in place of its old mta.nyc.ny.us location. One site that actually switched from a dot-com address to .info is racing-reference.info, formerly racing-reference.com.
Two years after the .info launch, the total number of registered .info domains started to decline, due to the expiration of many initial two-year registrations owned by registrants who saw no need to renew them. However, at last check, the total is still over 1 million currently-registered .info names.
I've noticed that the official site for .info shows a "sm" (service mark) next to the ".info" name. This seems to indicate that they're claiming intellectual property rights to the suffix, which goes against the U.S. Patent and Trademark Office's policy that TLDs have no source-identifying significance for trade and service mark registration purposes, as well as ICANN's policy that the domains in the root are coordinated by ICANN for the common good, are not "owned" by anybody, and are merely being assigned to registries for management via contract. When the contract runs out, ICANN could turn the management of .info over to a different registry company, but it appears that the present one would threaten to sue because it claims to own property rights over it. ICANN raised some objections to the government of Singapore attempting to gain intellectual property rights over its country code domain (the government in question backed down and dropped that attempt), but hasn't raised any objections to new TLD operators apparently doing a similar thing.
Some .info names in use:
.biz -- for businesses
.biz is Run by Neulevel, a commercial company. This originally struck me as rather useless; just another, less-known and sillier-sounding alternative to .com for commercial sites, rather than a clearly-differentiated namespace for a specific function not well served by the existing TLDs. However, the rules of .biz actually require that the intended use of the domain be commercial in nature, and that whoever is registering it actually intends on using it instead of hoarding and reselling it; if they actually can be made to stick, this new TLD won't be subject to as many of the silly abuses that .com is, and can hence actually gain some meaning.
Their official site, however, is laden with heaps of marketroid gobbledygook; they talk of "the convergence of business communications and commerce" (a way to say, with big fancy words, that you can use the Internet both to find information and to buy stuff), with nothing concrete about just how a new TLD is going to help you take advantage of all of this "convergence". Then they say that .biz will help "identify where your brochureware ends and your business begins" -- a dubious statement given that the proponent of this TLD is using its own nic.biz site to disseminate its own brochureware. And they promise "a near-military level of security and reliability", whatever that's supposed to mean. (Their nic.biz site has been unreachable at times. And, anyway, the military's treatment of domain name security might not actually be the best standard to go by, given reports of vulnerabilities in the .mil registration system!) And they suggest that "[y]our business customers and sales team, for example, can become www.businesspartners.biz and www.salesteam.biz", a silly idea given that these names don't even give a clue what company they're the business partners or sales team of -- it would make much more sense to use subdomains of a domain (whether .biz, .com, or something else) based on your company name, like salesteam.yourcompany.biz. But this new TLD may prove useful despite the silly hype of its proponent.
In May, 2003, it was noticed that going to a nonexistent address in .biz triggered a really obnoxious browser redirect to random and often unsavory sites. This was apparently done by a "wildcard DNS record" in .biz. .museum uses a wildcard record, too, but that goes to a reasonably useful index of .museum sites; the one for .biz imitated the worst of the pesky typosquatters. It was apparently part of a registry experiment which has since ended. On the other hand, a few months later in September, 2003, Verisign set up a similar wildcard record in the more popular .com and .net roots, leading to a search page that probably brings them per-click revenue as well as statistics that let their insiders know which unregistered domains get the most traffic (potential cybersquatters would love that info!), sparking waves of controversy.
An alternate registry claims to have created a .biz domain before this one, but yet another alternate registry claims to have created one even earlier, and is yelling about how their domain was "stolen" by the other registry who in turn had it stolen by ICANN. Nonsense... it never belonged to anybody to steal... .biz didn't exist as a part of the official Internet namespace until ICANN authorized it, and the fact that somebody somewhere set up an unofficial server just for kicks that had those three letters in it didn't give him any rights to it.
A licensee of the country-code domain for Belize, .bz, which intended to market it for business use, did a lot of whining about how .biz supposedly was an infringement of their rights. Courts didn't put much stock in this, though.
Some .biz names in use:
.name -- for individuals
.name is for for individuals and their personal web sites. It is run by Global Name Registry, a UK-based commercial company. This one serves a useful purpose, since the current domain structure offers no particular place for personal sites -- thus, of all the new TLDs, this one has the strongest justification on the basis of a strictly logical namespace. The structure of .name originally let people register only third-level domains so that nobody could monopolize an entire surname; you have to get john.smith.name. I liked the concept, but wished they had come up with a better name for it than "name"; it just doesn't sound right. For any domain suffix that's a complete English word, try actually reading a domain name under it out loud with the dots treated as spaces, and see if it makes a reasonable phrase. The current .net works pretty well; the domain ibm.net refers to the "IBM Net". The .info and .museum domains in this new plan also work; many likely names under them can be read out loud reasonably, like west-podunk.art.museum. On the other hand, "John Smith Name" sounds funny. Anyway, all domain names are names. Normally, the top-level domain gives some indication of what type of entity the name is for, like .com for commercial operations and .org for organizations. By that standard, the earlier-proposed .per for personal sites would make more sense. An abbreviated three-letter name that's not an actual English word has some advantages in that you don't have to worry about whether it produces an awkward phrase.
This new domain still has great potential if it can be marketed well to the millions of Internet users who don't yet have personal domain names; there's a definite advantage to having one which you can use for your e-mail and Web address and keep from having it tied to your ISP, employment, geographical location, or other potentially changeable things.
Somebody in a message board said, in criticism of .name, that "from a speculator's point of view, it doesn't provide many opportunities for investment." To me, that's one of the good things about this TLD. Finally, they're starting up a TLD intended to be used, to provide a logical address for something on the net, rather than hoarded, speculated on, and used as the object of get-rich-quick schemes. It's unfortunate that a mindset has developed whereby the Florida-swampland-salesman aspects of domain names has taken center stage and any TLD without that aspect is considered unworthy of attention, when in fact domain names were always intended as a logical namespace, not something to make money fast on.
The official site has some talk about how .name domains will somehow be used to let people make their personal information available to merchants and others needing it, but no details are given of how this is to be accomplished, and how security will be maintained for this.
The first batch of names were supposed to go live on December 13, 2001, with applications accepted until November 22, but the launch date was quietly pushed to January 15, 2002 instead. It went live on schedule, and you can now look up the names in their WHOIS. Unfortunately, they were only processing updates of domain records (name servers, contacts, etc.) every two weeks along with new batches of names, which means that my own registration of dan.tobias.name stayed in an unusable state for a while -- I owned it, but hadn't yet been able to get the nameservers changed to ones that respond usefully. I finally managed to get this changed, and my site is now live. A few months later, it went dead for a day or so due to a screwup by the registrar, GoDaddy -- somehow, they programmed their site so that changing the forwarding address for the associated email@example.com email address resets the name servers for the dan.tobias.name domain. But now most of the glitches are ironed out and it is possible to make real-time updates to .name domains.
About that firstname.lastname@example.org address -- although domain registrations in .name were (originally) done at the third level (dan.tobias.name), the registry (in a slightly controversial move) also made available email forwarding services at the second level. Technically this is a separate registration from the domain name, but some registrars provide this at no extra charge along with registration of a .name domain while others charge extra.
bill.clinton.name was registered to somebody in Barcelona, Spain. monica.lewinsky.name remained available at first, but seems to be taken now. george.bush.name has a defensive registration by the Republican National Committee (do they really own trademark rights on that name?). britney.spears.name was registered by somebody named Bimal Shah in England.
The early marketing push for .name fell considerably flat, not attracting that much interest among either the general public or even the registrars, few of which made any attempt to promote .name (many didn't even make .name registrations available at all). The only registrar participating in the .name affiliate program chose to stop offering .name registrations for a "temporary" period that's still going on now. However, in early 2003, a new promotional push was supposed to be starting, this time focusing more on the use of .name in email addresses than for Web and other uses, as e-mail is really' the "killer app" for .name. With various ISPs forcing customers to change their address multiple times due to mergers and marketing changes (AT&T Broadband has been through a couple of such changes), the concept of a stable address based on your own name is bound to have some attraction.
A registrar, Personal Names, was founded as a "sister company" of Global Name Registry; this raised some hackles on message boards about conflicts of interest among registrars, registries, and ICANN. However, at least this created a registrar that actually cares about this TLD, and might promote it better than the others do.
In early 2004, direct second-level name registrations began to be offered in .name due to a renegotiation of the contract with ICANN. All second-level names already in use as part of third-level registrations are reserved, but anything not so used can be registered, first-come, first-served (with no landrush or sunrise). This means that you can get a .name based on your single-word nickname or generic word if you wish; perhaps this will increase interest in this domain. Some on the message boards are critical of ICANN for allowing such changes to the structures by which new TLDs were accepted originally.
At about the same time as second-level names began to be registered, a threat emerged to the original third-level names; yet another stupid patent was issued (as is plaguing software and the Internet over the last few years; the patent office has shown no intention to enforce the law barring patenting of obvious things or things with prior art), claiming exclusive rights to the concept of issuing both a third-level domain (e.g., john.smith.name) and a corresponding e-mail address under the second-level domain (e.g., email@example.com). No matter that this exact thing has been done for years by many ISPs and webmail providers; the patent owner is suing. Hopefully, it will be thrown out of court, but you never know; Microsoft, with lots of money and lawyers, has thus far been unsuccessful in fighting off a patent-infringement lawsuit regarding the use of embedded applets in Web browsers, though this too has prior art.
Some .name names in use:
.museum -- for museums
.museum is for museums, including online "virtual museums". It is run by Musedoma, a non-profit organization. It has a relatively small constituency, but it could be useful within that field. See my discusion about how "niche" domains like this might actually be more successful than broad-based ones. The desirability of .museum may have been reduced, however, by the complex naming rules used at the launch, where museums must register in at least the third level of the domain space (and maybe even higher) instead of just "yourname.museum". Names may end up more logically organized than in most other TLDs, but longer and harder to type and remember. Also, museums that don't fit easily into a single category might have trouble placing themselves in this system. However, in early 2004, the sponsoring organization announced that direct second-level registrations would soon be accepted (making them the last of the seven new TLDs to announce such an opening of the second-level namespace). This was done in conjunction with opening up internationalized registrations, on the grounds that non-English names wouldn't fit well in the English-based generic subdomains that presently exist. At any rate, .museum will be checking on the legitimacy of a museum (which must be a not-for-profit institution with exhibits open to the public) before allowing them to register, whether at the second or third level.
The .museum startup procedure is refreshingly based on discussion and consensus within the targeted community rather than by clashes of lawyers, speculators, and marketing types as are most other domain policies these days. It seems kind of a throwback to the way the original TLDs were introduced in the mid-1980s when computer geeks and academics still ruled the net. Let's hope it can work for limited-community TLDs like this. However, another section of the agreement with ICANN imposes a challenge procedure for those who claim a registrant is unqualified, which would drag in external agencies to judge this (as with the UDRP for trademark infringement cases). Anyway, the agreement between MuseDoma and ICANN was signed on October 18, 2001, and the domain is now live (though still officially in its testing phase with no registrations final and no charges yet made to the registrants). Name requests in the first phase of the startup process were taken through November 1, and the first batch of names was announced on Nov. 8. A list of the first name batch is here. There is also a WHOIS server and a hierarchical directory, and "wildcard" DNS records have been set up so that you can try any .museum name in your browser and will get the directory site if no other site exists at that address.
During the experimental phase, the sponsoring organization hosted the DNS entries for all .museum domains rather than delegating them to the servers designated by the registrants, which limited the utility and flexibility of the names until a more permanent registration and delegation system was put into place. They announced, however, that Phase 1 of the experimental registry system would conclude at the end of February, 2002, to be followed by a Phase 2 where a naming structure is pretty well set and registrants are encouraged to follow it (in the first phase no structure was announced in advance and registrants had to suggest second-level domains to put their third-level registrations within). After the end of this phase, true registrations and delegations would begin. In April, 2002, the official eligibility verification service was inaugurated, in which registrants pay a fee and submit documentation to prove their museum status.
Some .museum names in use:
.coop -- for co-operative business organizations
.coop is run by the National Cooperative Business Association, a nonprofit industry association. Another niche interest, but useful within its market (and hoped by its proponents to encourage the expansion of the market for co-ops). It's likely to be misread as "coop" as in an enclosure for chickens, though (however, chicken.coop seems to be restricted from registration, even to a legitimate chicken cooperative, since it shows up as available in the WHOIS but is listed as unavailable if you try to register it), and anyone who hears it pronounced correctly may try to type it in as co-op (with a dash -- this form was actually proposed for the suffix at an earlier stage). It also may resemble .com too much to succeed in getting people to recognize it as a distinct ending.
Co-ops are defined by the sponsor of this domain as being a special type of business owned by its members rather than existing to make a profit -- as such, they fall somewhere between the present .com and .org categories, so a new suffix just for them makes some sense.
Negotiations with ICANN were reportedly stalled due to a request on the part of the sponsoring organization that the terms be changed in a major way from their original application -- they supposedly wished to use a monopoly registrar rather than competitive registrars as originally proposed. However, negotiations resumed, and now the agreement has been signed and the domain was officially launched on January 30, 2002 and a WHOIS server is up.
One interesting thing: in attachment 23 of their agreement, the sponsoring organization promises to set up a site with information for potential registrants at www.communicate.coop. However, this name is not on the list of names the sponsor has reserved for itself, which they're required to disclose in the ICANN agreement. When .coop was first added to the root, communicate.coop didn't even work (while nic.coop did), but it was added a few days later as a redirect to a page within the site.
Their Web site was redesigned soon before the launch, and for a while it had a "fancy" but useless Shockwave animation on its splash page... I saw no sign of any graceful fallback for anybody without the right plug-in, such as a plain-jane text link to get into the real site. That's an unnecessary accessibility barrier, and a sign of clueless Web development. It seems to be gone now, though.
A Names Strategy document formerly in the site (it doesn't seem to be up any more) seems to be trying to get co-ops to follow the same manaical strategy of registering boatloads of stupid unnecessary domains in .coop that regular businesses are in .com -- suggesting that they register every conceivable variant of their name, their product and service names, misspellings of them, marketing gimmick slogans, and any other name you might possibly think of using some day or want to prevent your competitors from getting first -- and if you've already got the .com, .net, and .org versions, you should "complete the set" with .coop. So much for using new TLDs as more meaningful and controlled namespaces.
.coop seems to have been one of the quietest and most uneventful new TLD launches; I haven't noticed any controversy or discussion relating to it, or any speculative activity, or much in the way of registration and use of the domain by legitimate co-ops for that matter. Trying their WHOIS shows that just about any name you can think of is still available (contrast this with .com, where it's hard to think of an even marginally meaningful name or acronym that isn't long since taken).
Some .coop names in use:
.aero -- for the air travel industry
.aero is for services and companies dealing with air travel and the aerospace business. It is run by SITA, an international air transport industry association. This is a possibly useful concept, though in a rather small niche. The original proposal of .air might have been more familiar to Americans from words like "airplane" and "airport" (when ".aero" is read out loud, they won't be sure how to spell it -- "Arrow"? "Air-O"?), but in other English-speaking countries such as England "aeroplane" is the word in common use and hence this spelling will come naturally. A few other languages have words like "aeropuerto", so they might find .aero natural too.
One useful feature from the original proposal is to assign addresses to all airports in a uniform style for functions such as checking arrivals and departures -- like arrivals.jfk.aero for the JFK airport in New York. If this is done consistently and widely used, it will be very useful to travelers who can figure out what URL to go to in order to check this information. A set of consistent URLs to check flight information worldwide would have been useful, for instance, during the tragic week of September 11, 2001, when the air transport system in the U.S. and around the world was thrown into chaos and nobody could be sure what flights would or wouldn't depart or arrive.
My early worries were that this, along with the other new TLDs for narrow and (presumably) easily policeable categories would be forced by ICANN to adopt the same sorts of ridiculous pro-trademark policies as the broader-based domains, forcing the registry to go through the hassle of processing "protective registrations" from thousands of trademark owners not even in the appropriate field, and possibly blocking some reasonable uses of names -- like, any three-letter airport abbreviation that happens to be identical to some non-air-travel-related trademark would get blocked, marring the consistency of the naming system. However, the announced startup procedure had no such provisions, thank goodness. (I discuss startup policies more below.
The ICANN negotiations seemed stalled for a while, but were finally complete; the agreement was signed, and .aero entered the root. Edward Hasbrouck had some critical comments about the agreement, to which ICANN made a response. No "insiders" at ICANN objected, though, and Hasbrouck's request to delay the signing was turned down. While I'm not sure Hasbrouck's criticisms are entirely valid (this is a limited-purpose TLD, so a tight focus is appropriate), I agree that narrow TLDs can be set up in a way allowing more participation by the affected community than .aero is -- for instance, .museum has an open public forum to discuss issues regarding its launch, and you don't have to run a museum to participate in the discussion there. In contrast, only industry insiders seem to have any ability to be in on the .aero launch in any way.
In a major "win", .aero gained permission from the country-code maintenance agency to use 2-letter codes at the second level -- in this case, they wish to use them to reflect airline codes, rather than country codes, but ICANN rules require permission of the country code agency to prevent confusion. Three-letter airport codes are also being used (though there would be a problem if an airport had a code like COM or ORG, since those are restricted by ICANN rules to avoid confusion with other TLDs). According to a 2004 announcement, all of the airport and airline code domains have now been activated even if the airline or airport in question hasn't actually registered it, to provide predictability for the public; however, when I tried a few of them in my browser now they didn't seem to work.
.aero has a newly created site in nic.aero, not the stagnant "official site" that was within the sita.int site for months (which has now gone away, and the old address gives a 404 Not Found error rather than redirecting to the new one). Registrations are through accredited registrars, rather confusingly called ".aero shops" (which sounds more like where one might purchase licensed merchandise like a T-shirt with the .aero logo).
The launch doesn't seem to be attracting much interest so far; it's hard to find any names in their WHOIS that are actually taken. Some Irish airports actually have their sites in .aero domains now, as do various minor aviation-related companies, but I haven't noticed any use by major industry players. Nevertheless, a news article touts .aero as a success, most likely based on reports from the marketing department of the sponsoring organization.
Some .aero names in use:
.pro -- for professionals
.pro is for professions such as law, medicine, accounting, etc., and is run by RegistryPro, a for-profit company. This has some potential if the professionals can be convinced to use it, and if there can be sufficient policing to prevent con artists who are not actually accredited as professionals from abusing it. (Let's hope the professional credentials are checked better than trademarks were checked during the .info rollout!) Its structure originally had registrations take place at the third level, within domains like med.pro and law.pro; however, as with other new TLDs that had a multilevel structure at first, it was eventually modified to allow registrations directly at the second level in addition to the third-level registrations.
.pro finally concluded its negotiation process with ICANN, last of all the new TLDs. Progress was stalled for a long time, and even seemed to be going backwards, because the registry announced that it wished to make significant changes in the agreement, specifically to reduce the amount of money it agreed to spend in promoting the new domain. This led some commentators to suggest that ICANN approval should be revoked and given to a different applicant instead, rather than letting them change the terms they agreed to in their original application. This TLD was in last place in the race to launch, but finally limped onto the track. It made it into the root as of May, 2002, but took until 2004 to accept registrations.
Their Web site used to have a fancy-schmancy scrolling applet that made it take a while to load and chew up lots of system resources (even crashing my browser on occasion). Can't anybody Keep It Simple, Stupid? Then, in late 2002, it was redesigned with a different set of fancy-schmancy doodads. It took months after that before it started having actual information updates, though.
The sunrise period for trademark owners started April 21, 2003, and continued long after live registrations of real domains were supposed to start in July.
Some controversy erupted in early 2005 when it appeared that the credentials of .pro registrants weren't being checked, making this TLD join the list of those intended for a specific purpose but turned into a free-for-all. Actually, the truth is that one registrar, Encirca, is letting anybody register .pro domains, using the registrar's credentials rather than the registrant's (technically the domain is registered in the name of the registrar); once a domain is registered, there seems to be no limitation on whom it can be used by, and effectively owned by, qualifying or not. Now, Encirca is actually threatening to sue anybody who publicizes the truth about them.
The State of the New Domains
As you can see, the new TLDs are currently "live", but there hasn't been much "buzz" about them outside the domain speculator and intellectual property communities. The new domains are desperately in need of something to get them into the public eye so people start getting interested in using them. Perhaps some sort of joint marketing campaign by all of the "alternative" domains -- the seven new TLDs, perhaps joined by .edu (spun off a while back to a nonprofit sponsoring organization) and .org (now under the control of a nonprofit registry connected to The Internet Society) -- ought to be conducted, with some clever slogan like "Not just another Dot-Com!". It could highlight useful and interesting sites in non-dot-com domains (even including the likes of .gov and .mil) emphasizing how you're more likely to find real content as opposed to marketing hype and make-money-fast schemes, and how the domain endings have real meaning that clues you in as to what type of site you're going to.
Important issues in the launch of new domains are the procedures to allocate registrations in their opening period, when lots of people are going to be trying to get "good" names, and also the procedures for resolving disputes over name conflicts. One part of this is the Uniform Dispute Resolution Procedure (UDRP), imposed by ICANN on all global TLDs (possibly with minor modifications to suit the nature of specialized new domains, but not differing in its essentials). Read more about this in the conflicts section of this site. This procedure allows somebody who claims their trademark is being violated by somebody's name registration to challenge it. However, on top of this, special new rules and procedures, different for each new TLD, are being introduced to handle who gets first shot at getting the new domains as they become available, as well as dealing with challenges regarding eligibility for domains that have special conditions (like .biz being just for commercial use, .name just for individuals, and .coop for co-ops). Some details of these procedures have already been announced:
.info rules: Sunrise Period
The .info domain is completely open, with no eligibility rules such as what type of entity can register a name (commercial, noncommercial, etc.). However, to satisfy the trademark lobby, they created a "sunrise period" preceding open registrations, during which only owners of matching registered trademarks can register .info domains. As a "cybersquatting solution", this strikes me as overkill which bends the rules too far in the direction of trademark owners. Practically every word of the English language is somebody's trademark in some field, so it was clear, even before the whole process started, that a lot of very generic names would get locked up in the sunrise period before anyone who wishes to use them in their nonproprietary sense would ever get a shot at them. For instance, the Caterpillar tractor company made a big deal about how essential it was for them to protect their "Caterpillar" and "Cat" trademarks, so I knew they'd grab caterpillar.info and cat.info, keeping them out of the hands of insect and feline enthusiasts who might want to put up informational sites on those subjects that don't in any way infringe on the rights of the tractor company. Other companies would similarly make overblown and excessive pre-registrations in that vein, and some domain hoarders and speculators managed to get trademarks registered in some country or other to allow them to jump the queue for some other generic words. As a result, the deck was stacked against people without big bucks who merely want to get a good web address for an informational site they wish to set up, even though that's precisely the sort of people this new TLD was intended for in the first place. See my newsgroup message on this subject.
One phenomenon I didn't really expect, though, was the extent to which blatantly fraudulent registrations would be done in the sunrise phase. Since the WHOIS server has been available for .info since an early point in the sunset period, it's possible to check on the status of names, and there turns out to be a large number of them, including many highly desirable generic names, that have been registered based on completely nonexistent trademarks, with the registrants hoping to bluff their way through the registration process without being rejected or challenged.
A study of sunrise registrations shows that an estimated 15 to 25 percent of registrations are bogus, with no real trademark, and the percentage is higher where the "best" names are concerned. DomeBase has some interesting comments, and a proposal to solve the fraud problem by giving the "winning" landrush registrant to a name already taken in the sunrise period a chance to file a challenge and win the domain. (Afilias has ignored this proposal so far.)
A problem with an overly pro-trademark-owner policy is the risk that a large number of the names in a new TLD will end up being owned by companies that have no intention of using them in Web site, e-mail, or other Internet addresses, but are just trying to (possibly overzealously) protect their trademark. The problem with this is that, to gain public acceptance of a new TLD, it's important that it be "out there" getting used -- if people see lots of .info addresses on TV, in magazines, in the "From" field of e-mail messages, etc., they'll get used to it, while if those addresses are never seen, they won't. For instance, few people know that the .int domain (for international treaty organizations) exists, since so few organizations use it, and those that do aren't heavily advertised and promoted. If .info gets monopolized by companies who never actually use it, it will suffer the same fate.
Afilias announced that they would file challenges themselves against sunrise registrations with questionable trademarks cited as their basis, if nobody else did. This increased the chance that a bogus sunrise registration would be overturned, since in cases of generic words that nobody owns trademark rights to, any successful challenge would simply put the name back in the general pool for anybody to try to register, making the challenge not really worth it since, though a successful challenger gets most of the challenge fee back, $75 is non-refundable. They've created an e-mail address to report questionable registrations. Independently of Afilias, somebody has created a Web site to report on and list bogus .info sunrise registrations that have been uncovered, so they can be reported and so that the status of any challenges against them can ultimately be tracked. Afilias has announced a new landrush to distribute names released as a result of these challenges, and it's now underway (though not all registrars are participating yet).
The challenge period is over; to see the challenges and results, go to the .info page of the WIPO site. A number of challenges are listed on this site, along with a very sketchy list of the outcomes (which doesn't even say who the complainants and respondents are, let alone give the wording of the opinion as is done for UDRP cases). Some critics say that, even after all the "last-resort" challenges by Afilias (which aren't publicly listed), lots of really dubious landrush names have been allowed to stand.
A last-minute rule imposed by ICANN in between the sunrise and the landrush period was to restrict registration of country names (e.g., "southafrica.info") until early 2002, when ICANN will decide whether to hand them over to the governments of the countries involved or open them up to the public as the original rules would have done. Considering that the WIPO report on expanding protection of names beyond trademarks explicitly declined to request immediate action on the protection of geographical names (while calling for protections on generic drug names and international treaty organization names and acronyms, which ICANN has not done anything about), there doesn't seem to be much justification for this ICANN rule, especially for it to be imposed in the middle of the launch already in progress. ICANN has created this forum area to discuss the disposition of these names.
Here are a few names I checked on during and immediately following the sunrise period, and my comments on them. Note that due to subsequent challenges and transfers, the current owner might no longer be the one I mention here.
.biz rules: IP Claims
The rules for .biz were more fairly balanced, I think. Registrations were open to all at the same time, whether they had a trademark or not. Rather than getting first shot at all names as in .info, trademark owners merely had the ability to file an "Intellectual Property Claim" on names to which they assert ownership. This didn't give them any advantage at securing the registration of that name, but it did put anyone else attempting to register the name on notice that there was a claim on it, and that anyone else registering it was likely to be challenged or sued. This might dissuade bad-faith cybersquatters, but somebody with a legitimate use of a word that's generic in a different sense from that used by the trademark owner could still go forward with a registration and (hopefully) win any challenge that results. (However, the arbiters have been uneven in adjudicating the challenges; some recognize valid uses of generic words, while others seem to think that any trademark, in any industry, in any country, outranks all generic uses even of a very common word.) Ironically, the .biz domain, intended for businesses, actually would have more logic behind a pro-trademark stance such as the .info domain enacted, and less need to protect the alternative name uses of non-monied people, such as noncommercial website operators who have more bizness in .info than .biz, but may not get the chance if the trademark people pick that domain clean first.
However, the fact that people were allowed to make multiple registration attempts at the same .biz name, and must pay a non-refundable fee for each attempt, and the more "entries" they have the greater their chance of getting the name they want, caused the .biz registrars to be charged with running an illegal lottery, in a lawsuit filed in California. (And some other me-too attorney filed another similar lawsuit afterward.) This case led to a preliminary injunction stopping the .biz lottery, and though this was ultimately lifted, Neulevel decided to award only uncontested names at first, keeping the most desirable ones on hold for a few months. In December, 2001, they finally cancelled the lottery and announced a .info-style landrush for these names in March, 2002, with only one registration attempt per name per registrar.
In other legal action, Amazon.com wrote a threatening letter, hinting that they might sue not only over the illegal-lottery aspects, but also because of the lack of a .info-style sunrise period giving absolute precedence to trademark owners. Meanwhile, Sun Microsystems jumped in themselves with a threatening lawyer letter aimed at registrars, demanding that they refuse to let anybody but Sun register various names they claim to be their exclusive properties, which includes some common words like "enterprise" which are in fact used by many other businesses other than them (e.g., Enterprise Rent-A-Car, not to mention the Star Trek starship). (New York Times Article -- requires pain-in-the-butt registration to access.)
In the "dim bulb department", somebody sent a threatening letter, kind of in a "me-too" style from the Amazon and Sun letters, claiming to have rights to business.biz based on alleged use of that name for a comedy act. First of all, a check of the USPTO database shows that the claimant only has an application for a service mark which has not yet been approved. But most importantly, business.biz is on the Neulevel reserve list, meaning that registrars couldn't successfully register this name even if they wanted to defy this claimant's alleged "rights".
It would be a big shame if, as a result of these suits and threats of more, all future TLDs are forced to grant complete blockage of any word that is trademarked from its generic and fair-use meanings. The .biz registry, NeuLevel, has pre-emptively sued Amazon to head them off. See details (2 megabyte PDF file of complaint).
In addition to trademark challenges, there is a challenge policy whereby any .biz registrant can be challenged by anybody else on the grounds that the domain name is not being used for bona-fide commercial purposes in accordance with the charter of this domain. If the domain is merely being used for personal or noncommercial purposes, or was registered merely for attempted resale, it can be transferred or cancelled if challenged. The challenger need not have a trademark on the name, but if they don't have one, the name will merely be cancelled instead of being transferred to the challenger.
.name rules: Defensive Registrations
The .name domain is intended for individuals, not corporations. In an ideal world, then, there shouldn't even have to be any conflict rules regarding corporate trademarks, as corporations have no business in that TLD anyway. However, such is the power of the trademark lobby that they had to make such rules anyway. The .name agreement calls for "defensive registrations" to be allowed by companies that fear abuse of their trademarks. Since .name domains are registered at the third level as firstname.lastname.name, defensive registrations can be for a full third-level name like ronald.mcdonald.name or for a generic set of names under a particular first or last name, like ronald.*.name (which would pre-empt anybody registering "ronald" under any surname) or *.mcdonald.name (which would pre-empt anybody registering any name under "McDonald"). As you can see, pre-emptive registrations can be highly potent; as the .name registry says themselves, they "block an extensive part of the .name real estate." This seemed like a scary amount of power for corporate types to have over a namespace intended for individuals, but it's really not so bad, because of the challenge procedure.
Any individual whose actual given name falls within the "fenced off" area of a defensive registration is allowed to file a challenge. These challenges are handled in a "loser-pays" way -- both parties must put up an escrow amount to cover cost of the arbitration, and the winning party gets his/her/their money back. Somebody who demonstrates that "Ronald McDonald" really is his name would be allowed to register it despite the defensive registration. And, in the case of defensive registrations that cover an entire first or last name, there's a "three strikes" rule where if the owner of the defensive registration loses three different challenges to names within this space, the defensive registration is cancelled altogether. This should discourage companies from being overzealous in claiming common personal names defensively even if they do have a trademark on them in some context. (Several articles about the .name registration procedures have instead claimed that there would have to be three successful challengers to a defensive registration before any of them would be allowed to breach this registration, but based on my own reading of the rules, I believe that this is a mistaken interpretation.)
Owners of defensive registrations are allowed to grant consent to individuals to register within their defensive perimeter, and if this is done voluntarily it doesn't count against the three strikes. There's a rule against companies asking for money to allow such registrations, to prevent defensive registrations to be used as a moneymaker by companies wanting to corner the market on common names. So it's possible that the owners of some trademarks that are also common personal names will file defensive registrations then grant permission to legitimate individuals to register under those names anyway, and hope they don't get three people going around them and challenging formally. If my name fell under somebody's trademark like that, I'm not sure whether I'd want to go meekly and ask for permission (and possibly be forced by some corporation's legal department to sign some demeaning contract regarding what sort of use I can make of my own name), or go the more difficult but less demeaning challenge route whereby I can get a formal declaration that I indeed own my own name and can't be bullied around regarding it.
To date no challenges of defensive registrations have shown up on the dispute resolution provider sites, but a handful of challenges of normal .name registrations have been filed under the Eligibility Requirements Dispute Resolution Policy, resulting in the cancellation of instant.messenger.name and aim5.instantmessenger.name at AOL's request (they didn't have to prove they owned trademark rights to them, only that these names didn't represent the personal name of the registrant), and the transfer of donald.trump.name from a registrant who is not named that to the famous guy who is.
.museum, .aero, & .coop Rules: Charter Eligibility Dispute Resolution Policy
These three are restricted, sponsored domains, with a sponsoring organization which enforces a charter that requires that only legitimate entities in the appropriate category are eligible to register a domain. However, ICANN apparently doesn't trust these organizations to police their own eligibility, so it insisted on a dispute resolution policy as a "check-and-balance" on them, as noted in Attachment 12 of the sponsored domain agreement. Somebody who feels that a .museum registrant isn't a true museum, a .aero registrant not a true air-travel-related entity, or a .coop registrant not a true co-op, can file a challenge, which would be judged by the same sorts of folks who handle UDRP cases. Hopefully more fairly, though.
Regarding how names are to be obtained in these domains, .museum has been taking preliminary requests for some time now, subject to their acceptance based on consistency with their naming conventions and eligibility of the applicant. .coop has begun to allow members of sponsoring organizations (various associations of co-ops) to request names, first-come, first-served. After a few months, the general community of unaffiliated co-ops will get their chance. The .aero domain is being launched in stages to various segments of the aviation community, with some segments (like individual crew members) only being able to register third-level domains within categories to be established, while others (like airlines) able to register second-level domains. Some names, like two-letter airline codes and three-letter airport codes, are reserved for the appropriate entities.
Regrettably, .coop has joined .name and .pro in introducing protective registrations for overzealous trademark owners who don't qualify for regular registrations in that TLD but still want to prevent others from getting particular names. They call it the Brand Safe Program, and it helps make money for the registry and for trademark lawyers and consulting firms that file such protective registrations on behalf of clients, while constricting the available namespace for the group that the TLD was actually intended for.
.pro Rules: Qualification Challenge Policy
The agreement for the .pro domain gives some information about the challenge process. In addition to the normal UDRP challenges, there will be a special challenge process to object to a registrant not having proper credentials to be eligible for a .pro domain (e.g., not being a doctor but registering in .med.pro). There will also be a sunrise challenge policy to object to a registrant's alleged trademark rights used to get first shot at a registration. Some of these concerns apply to the three sponsored domains above as well, but even more to .pro because it's a commercially-run domain with a broader user base than the niche markets of the sponsored domains.
The .pro domain requires registrants to indicate their professional license number and what state or country issued it. In the original agreement, the sunrise period was to be open only to trademark owners who also had professional credentials as required for the .pro domain -- so trademarked pros (a minority given that this domain is aimed at individuals rather than firms) would get first shot, but not trademark owners in general. The .pro registry originally somehow managed to get away with not giving the trademark lobby some sort of pre-emptive rights of unrelated trademark owners to stop the use of names that have the appearance of infringement, like apple.med.pro for a doctor whose last name is Apple. However, the UDRP still applies, so such registrations can be challenged as cybersquatting.
Update: An amendment proposed in November 2002 and approved by ICANN in December adds a "protective registration" provision similar to that in .name, where trademark owners, even in fields unrelated to the professional fields .pro is intended for, can block registrations of "their" name at a cost. This seems to me like a scheme by the registry to extort money from trademark owners in a "protection racket", probably proposed to make up for lower expectations in income from normal registrations given the not-so-stellar performance of limited TLDs so far. It's a rather useless service, since cybersquatting and domain speculation isn't likely to be much of a problem given the registration restrictions and the lack of speculator interest in domains like .pro that only permit registration at the third level.
An unusual requirement for .pro is that all registrants must obtain a secure certificate, which is to be used to certify that they are who they say they are.
It will be interesting to see if any attempt is made to challenge the large number of current .pro registrants who appear to be entirely unqualified under the domain's rules.
An Also-Ran Name: .geo
One of the more interesting proposals (but not accepted in the first batch) was for the .geo top level domain for geographical indexing. (They used to have a Web site at dotgeo.org, but, for some reason, they stopped using it and now instead have their site at dotgeo.net and dgeo.org.) That wouldn't be used for "human-understandable" names like other top level domains; .geo domains would be long things referring to ranges of latitude and longitude (and hence wouldn't have the trademark and cybersquatting issues of other domains). Their intent is not to have them be used as Web addresses to type into browsers, but to be used to identify a decentralized network of servers that would oversee a global Internet index by geographical location in which your site that you wish to be identified by location (e.g., of a restaurant) would be registered with the appropriate server so that users can use software that shows them a map with all such sites indicated by location, and perhaps be directed straight to the place through a GPS-enabled mobile Internet terminal. I haven't studied it enough to have an opinion on its practicality and desirability, but at least it's "thinking outside the box" and coming up with a use for the domain name system other than as names for marketing types to hype and lawyers to fight over; and it explicitly makes use of the distributed, hierarchical nature of the system which everyone else seems to be trying to disregard or flatten. (But does it need a new TLD? Not really, except for the marketing purpose of attracting attention by being one of the few entities to win one. At a technical level, however, the service could be implemented just as well using second, third, or higher level domains. Perhaps the most logical place for it would be under .arpa, where network infrastructure functions are now being placed.)
I will point out, however, that although the proposed structure of .geo avoids the sorts of disputes and abuses that are rampant in the rest of the domain name system, humans are ingenious at figuring out new ways to create disputes and abuses, and will do so with .geo if and when it goes into common use. A possible abuse is "spamming" of the geographical system so that your silly porn and make-money-fast sites come up as geographically tagged to various high-traffic locations with which the site has absolutely no connection -- a tourist in Times Square who uses his mobile terminal to bring up local points of interest will get inundated with crap that's been tagged to that spot because it's so popular. Another possible abuse, if the geographical database has sufficiently fine resolution to let sites be tagged precisely to an exact building, is for people to put up sites tagged to the location of a business which advertise a competitor or which criticize that business (e.g. a union's site accusing them of unfair labor practices) -- is this abuse or free speech? Or, more sinister, a site attacking a person that is tagged directly to that person's house -- is that an invasion of privacy? There's a famous case of a few years ago where an anti-abortion group published the home addresses of abortion doctors and implied that killing them would be a good idea. Future versions of that might well be .geo-enabled and allow fanatics with wireless GPS-enabled access to be directed straight to the nearest potential victim.
Somebody else has an unrelated GeoTags initiative to get people to insert meta tags associating their Web sites with a geographical position. Since no "geo-servers" are created in this system, unlike the .geo plan, the usefulness of this would depend on there being search engines that pick up on these tags and present Web sites arranged geographically. Another site along those lines is GeoURL, which has its own set of meta tags but recognizes those of GeoTags as well.
This page was first created 14 Mar 2001, and was last modified 21 Sep 2013.