If it's a company's trademark, fine -- let them buy the corresponding TLD. No harm done.
Bzzzt, wrong. Trademarks are industry-specific: that's why Apple Records can exist alongside Apple Computer. If there's no likelihood of confusion, there's no conflict.
And of course, trademarks are not fully international.
Winer is right about ICANN being wrong. There's just so much else that ICANN is wrong about.
As an excellent example of this: Consider "Monster". Both Monster Worldwide (e.g, Monster Jobs) and Monster Inc (e.g, Monster Cable) applied for the .monster TLD, and there's at least one other significant claimant to that brand who didn't apply for the TLD (Hansen Natural / Monster Beverage Company, makers of Monster Energy).
Hilariously, both Monster Worldwide and Monster Inc claimed in their applications that granting them the TLD would reduce user confusion:
Monster Worldwide (application #477):
> The proposed .monster gTLD has the following user experience goals: [...] Reduce the risk of Internet users being misled, believing and⁄or acting on erroneous, information about Monster Worldwide, its business partners and⁄or its products and services presented online by unauthorized 3rd parties
Monster Inc (application #271):
> Therefore, .monster gTLD will: [...] represent authenticity and assurance that the domain names are directly associated with Monster thus promoting user confidence...
In terms of reducing the risk that users would mistakenly use a site where a third party was intentionally looking to pass themselves off as the particular variant of Monster, it is a good argument. When a consumer is aware that the only legitimate Monster sites are those with the .monster suffix, this would reduce the risk of them using fraudulent sites.
If on the other hand a user looking to find the jobs variant stumbling upon .monster domains controlled by the energy variant, there is no risk of user confusion in this regard.
In summary, the confusion they allude to is third parties using the mark or a variant thereof in a deliberate attempt to mislead, as opposed to confusion between established brands.
"Not so sure about Amazon buying .amazon, because it also is the name of a rainforest in South America, and .apple could be a problem for growers and lovers of the fruit."
But you did not cover the critical issue; you have not distinguished between www.sex and www.sex.com:
"the problem. And it extends to words and concepts that weren't created by anyone living today. Sex, love, laughter, babies, books, songs, cars, poetry, etc. These things shouldn't be TLDs, they're too important, too basic to life. Not the kinds of things any company, for crying out loud, should be able to claim to own."
Why is it such a tragedy if poetry is a TLD but poetry.com is no problem? Why is unthinkable that a company could own .poetry but no problem if the same company owns poetry.com?
Okay I'll cover it here (although I did elsewhere in this thread).
Maybe those domains shouldn't have been issued, but they were. And that's why you want to deal with these things before you get an installed base, not after.
You did not cover it...What is the harm of having a .poetry TLD? Or what are the personal+societal costs of the private/exclusive registration of poetry.com?
But you are right we should deal with these things before there is a large user base. So let's sort it out now and come up with a solid framework for what alphanumeric strings are acceptable. What are the rules? Which alpha-numeric strings are too important to let a company own?
My name is Doug, can I register .doogie? Can I register .poetree? Is .oybt acceptable, even if it is rot13(blog)?
It's simple, really. A domain name like poetry.com is as accessible to the average person as it is to a company. A TLD on the other hand is something you need to take out a mortgage to own.
The difference is in the democracy of it all. Acquiring domain names is a very democratic process even if you take into account squatting and other shenanigans. Scrounge up ten bucks and a web connection and it's yours. TLDs aren't even close. It's a total plutocracy and ICANN seems totally unapologetic about it while the companies snatching them up look ridiculous when they put a line in their proposal about how their ownership of some random word as a TLD is in our, the users, best interest.
How can anyone overlook that? It's practically beating us over the head, taunting us.
And how many mortgages do you suppose it would take to buy poetry.com? It's not even close to accessible to the average person. Maybe if the average person saved every penny they earn for a lifetime, they could buy poetry.com.
Yes but domain names that are unclaimed all start at the same price. Now poetry.com is unattainable but before it was bought anyone who wanted it had an equal chance. TLDs are starting their life as unattainable unlike domains which only end up that way later.
You are right to extent, however, even in the absence of a lack of confusion, an action can be brought where a third party in using a similar/identical mark to the original takes unfair advantage or causes detriment to the original mark.
In fact, Apple Inc and Apple Records engaged in a long-standing legal spat over the usage of the Apple mark before finally reaching a settlement in 2007.
While all of your points are correct, I think you missed the point.
That line of argument is expressly what the author was trying to avoid. The author acknowledged there were issues with it but stepped beyond them because there's a larger point that he was trying to make. (One I think has its own issues, but that's for another post.)
Many do not understand this point. Unlike domain names, which are ambiguous, trademarks have a couple of dimensions that allow ambiguity to be greatly reduced: class and geographical scope.
I have actually developed a solution for the domain name ambiguity problem. It respects all trademark owners, not just the ones who pay ICANN the most. (Aside from squatters, registrars and registries, ICANN takes a small percentage of every defensive registration for every trademark in existence. Now ICANN has upped the ante to well over $200K.)
Of course, it is a simple solution and I doubt anyone is interested. The way things are structured, with ICANN encouraging (by ignoring) conflict and promoting a "winner take all" approach to naming, is beneficial to those who are gaming the system. And it is quite easy to exploit ambiguity, "winner take all", and to game this system.
The root really isn't as important as people are led to believe. It's the large registries that are more important, such as .com. If a user wants to access names in a fringe tld registry, whatever it may be, he can tune his DNS to allow that. If the content on sites in that tld is that good, he'll make the effort to adjust his DNS settings. It is not prohibitively difficuly by any means.
ICANN wants to play the registry game, after years of watching domaining and repurposed cctld's making millions for others. The root is all they have. So, like a registry, they want to open the root zone to domainers, which will also force the hand of trademark holders who need to register defensively. ICANN knows exactly how this all works, and this is a calculated move to enrich insiders.
The person at ICANN who was supposedly running the program just resigned. A replacement was just announced.
I'm always surprised when people defend ICANN and their root. The root is not so important. If you have your system tuned to access the .com/.net servers, you can access the majority of the internet. Add .org and the cctld's and you have almost 99%. The root zone is not very dynamic. You can run your own copy. It rarely changes. But the root is all ICANN has with respect to naming (IANA is the real jewel of their crown, though the US Dept of Commerce may take that away if ICANN is not careful). ICANN wants to play the domain name game. If this whole new gtld proposal does not make you see the level of double dealing going on at ICANN, I'm not sure what will.
ICANN's role is overstated. The addresses of the tld servers we all use rarely change. Any user could manage their own copy of the root. The root really just serves to prime DNS caches. It is not something that should get queried very often.
"How does Google get the right to capture all the goodwill generated in the word blog? They are not the exclusive owner of it, as they are with the name Google. However they claim the right to become that owner, by paying $185K to ICANN. Nowhere in their proposal is an offer to pay money to the people who created the idea that they would take over. And what if the creators aren't willing to sell it to them? permalink"
Dave - you own scripting.com. What gives you the right to control that? How is that any different? I can't use me.scripting.com without your approval. You didn't agree to letting anyone do that when you bought the domain for $x per year in fees. You didn't invent the word "scripting" you were just the one to get the domain back in 1995, right?
What gives anyone the right to own any domain name "news.com" or "boat.com" etc?
That said I am not a fan of the new TLD's but that's a separate issue.
It's not entirely different, but those were the rules everyone was playing by then. So in that sense it's very different. We haven't agreed to these new rules yet. I don't think we should. Hopefully that's not too subtle.
The rules are clearly laid out now. The difference is that you need to be able to pay real money to play now. Some of that is greed and some is practical. You need to limit the applicants to organizations with the financial wherewithal to be able to provide the stability needed (that said they do have some kind of program for less financially worth non-profits).
"I don't think we should."
To late for this now. The time to do that was many years ago when comments were invited and there were meetings etc.
By the way there are many parts to the ICANN process that are exclusionary. To get to one of their meetings you have to be able to afford to get in an airplane and fly to an exotic (and sometimes) unsafe location. Nothing fair about that. But that is the way things are (in Washington as well).
Each reason the author gives against the new TLDs also applies to domain names.
For example, he writes:
"Not so sure about Amazon buying .amazon, because it also is the name of a rainforest in South America, and .apple could be a problem for growers and lovers of the fruit."
But the same reasoning applies for domain names:
"Not so sure about Amazon buying amazon.com, because it also is the name of a rainforest in South America, and apple.com could be a problem for growers and lovers of the fruit."
I think part of the point is that its not clear what value these new TLDs add. Google already owns google.com. By creating the possibility of ".google," you aren't doing much else other than forcing them to own it defensively. There isn't really a benefit to mail.google over mail.google.com.
In other words, in a large sense you are right -- all these issues exist for the TLDs we have today, and they've already been hashed out. Why create another round of 90's style litigation around who bought .coke first or bless a new set of owners with "high quality" tld's like .food? This isn't going to realistically expand the "number" of domains available, its only going to force companies to defensively buy even more permutations of theirname.___, and for all the domains that are already taken in ___.com form, it will only add confusion by having a parallel .___ version.
You are not forcing them to own it defensively. For $2k - less than pocket money for Google - they can register their trademark with the ICANN which prevents others from applying for it.
The explanation you gave still requires paying money, again, as a purely defensive move. Fine, you don't care about Google since it has so much "pocket money" laying around. How about startups that can't afford to throw 2K around every time ICANN decides to add TLDs?
"every time ICANN decides to add TLDs"? There are Top-Level Domains. You can't go any higher without knocking everything else down a level. ICANN has already decided that the top-level namespace is up for grabs, and the $2K fee is a one-time payment to protect your mark. Am I missing something? How can they "decide to add TLDs again?"
For normal domains, there's at least a well-defined process for registering, transferring, and disputing ownership. There's no equivalent for top-level domains, and changing the ownership of a TLD would be likely to cause disruption on a much wider scale than for a normally registered domain.
There is. It's the process that's happening right now. Applying for a tld doesn't guarantee getting the tld. Now is the time to object. Paying $185,000+ doesn't guarantee getting the tld. Lastly the intellectual property constituency had a big role in all the rules and discussions of this which has been going on for many years now.
Unfortunately, the objection process is thoroughly stupid. If I recall correctly, there's a fee of several thousand dollars to file an objection to an application!
Plus ICANN was setting the entry barrier so ridiculously high, that generic domain names are almost exclusively landing in the hands of these corporations, who are already strong in a particular market. It is in so far not stimulating competition but strengthening already successful market players.
The next step is most likely that Google returns the favor to ICANN and puts a stronger weight on the domain extension for search result relevance - just another step towards making a more corporate internet.
There are some subtle differences - Apple owns the .com variant of the URL (being a company, which the fruit isn't), but not the .org. It also doesn't own all of the international apple domains. Apple.co.uk for example belongs to some UK-based illustration company.
And these will continue to exist even if Amazon.com the company secures .amazon the TLD.
I think Dave's mistake is that he believes that TLDs represent anything canonical. Originally they were meant to, but only a few do anymore[1]. In almost every case, they're just a few extra characters at the end.
But only .com and .net have no real meaning attached to them. It doesn't really make much sense to register amazon.es or amazon.biz for the Amazon rainforest, now does it?
I haven't seen anyone talk about it, but there is another important implication of all these new TLD's: security.
This will give malware purveyors a whole new plethora of vectors to exploit insofar as social engineering goes.
Imagine regular user 'A' is surfing, looking for a cool new pair of shoes. They know that kewlShoes is their fav shoe company evar. Some entity has paid the huge fee to acquire the .shoes TLD in order to sub-let domains at whatever nominal fee they decide.
User 'A' browses to kewl.shoes instead of kewlshoes.kshoes and unwittingly becomes the latest drive-by-download victim happily handing over their credentials to who-knows-who.
I know this is broad and speculative, but think it is worth consideration.
Has there been other discussion about this out there that I haven't seen?
I think the vector itself is an issue; but I don't think this will make it significantly worse. For any domain, there's already hundreds or thousands of possible similar domains across the existing TLDs - adding a few more TLDs won't make much of a difference.
I also doubt that the future owner of .shoes - whoever that may be - will be stupid enough to devalue his own TLD by selling second-level domains like candy. There's probably more money to be made by banking on its exclusivity.
No, this is ~2000 TLD applications. We have no idea how many of those will ICANN approve.
But many of these new TLD's have been applied for by entities who plan on sub-letting the TLD to others.
My point wasn't that they won't sell domains, but that many or most won't sell them to any random shmoe like .com/.net/etc are sold, but only to businesses in the fields and how are willing to pay a lot more for the privilege, which doesn't describe scammers.
>but only to businesses in the fields and how are willing to pay a lot more for the privilege
Here's to hoping you're right.
Though, I doubt anybody would be surprised to find the bar vastly lower than you expect--especially when it comes time for these purchasers to recover some of the cost and find these fields are smaller than anticipated.
My original thought when hearing about the prospect of new gTLDs was that a central body would look to claim a descriptive term such as 'blog' and make it available to anyone who wanted to host a blog under set criteria, and provided that centralised body committed to providing access to anyone under reasonable terms. This would act as a guarantee of origin of the site and would be helpful to consumers.
I don't therefore see an issue with the registration of new descriptive gTLDs but only if access is provided in line with the above. The issue with Amazon/Google's land grab is that we cannot guarantee the basis on which access will be made available which could in turn lead to segregation, particularly if google were to control a gTLD and also provide greater prominence to it in search results.
At the present moment in time, it is the lack of certainty in the process which is troubling. Of course, if I read ICANN's handbook more fully things might become much clearer.
I don't know how the internet could easily transition out of this mess, but I think it would probably be good to eventually get rid of as many TLDs other than .com as possible. Does it add any real value to users to have to try and remember and differentiate between .com, .org, .net, .info, .me, .ly, .tv, all the others I'm forgetting about, and now .shoes and .pants and whatever else companies come up with?
It just creates more and more opportunities for phishing attacks to have essentially unlimited TLDs created where it makes it harder for anybody to easily figure out what is the actual home of any company.
Also, I'm not worried about Amazon or Google or Facebook or Microsoft, but for smaller businesses, they're going to be faced with costly and unnecessary legal battles and challenges over all kinds of trademarks. It's such a waste of effort. It's bad enough when there were just a handful of TLDs that were relevant, but now that we have essentially unlimited domain names all sorts of additional costly conflicts will emerge.
If these new TLDs have to exist, one potential way to minimize confusion and minimize disputes would be for all .com owners to automatically be assigned that respective TLD, so you always automatically know that "icloud.apple" is owned by Apple.com and that "books.amazon" is owned by Amazon.com and if you went and bought "xyz12345.com" you would also automatically own the xyz12345 TLD so these disputes would mostly be already settled. But that's not going to happen because there would be no extra money in that.
Problem with that (aside from the lack of capitalistic motive) would be management (IIRC, ICANN isn't directly managing the new TLD's, that job goes to the companies who actually end up with them), as well as what happens when someone who illegitimately owns a .com name were to get the associated TLD.
Funny, I've heard all these same complaints about second-level domains.
It's a human-readable addressing scheme. At some point, words will be used. The more memorable, the more usable. All those words that were "too important" to be used as TLDs should be allowed as TLDs precisely because they are important.
TLDs should be open to any string of characters, unicode if possible, and registrars should have universal single price registration fees for any site on any domain.
If in that nightmare world .llama should be bought by a fad in comfortable footwear, depriving some Llama rancher's association of their god-given rights to address space, we will just have to hope Mother Earth will somehow learn to heal.
Instead of imagining future conflicts, we should look at TLD behavior right now. Look at the rise of super short TLDs flooding profits to random countries. It might make you realize that appending the meaningless ".com" at the end of everything is a giant waste of time, since people will pay so handsomely to avoid this, all proceeds to random lottery winners like Tuvalu. (Nothing against Tuvalu, who undoubtedly needs the money for boats to leave their poor drowning island.)
If you look into what consumers want, instead of the complaints of a few random interest groups, it'd be clear we should somehow be able to go to "google" instead of "google.com" or "wiki" instead of "wikipedia.org" or "bit" instead of "bit.ly." Chinese users should be able to spell Baidu in their natural language, without resorting to a half-assed toneless Westernization.
ICANN's biggest mistake is not going far enough. They should chuck the whole TLD system and start from scratch, driven by the revealed preferences of those who actually browse the internet.
One thing I've not seen addressed about these new proposals are how cross-domain cookie wildcards are going to be handled. Right now all the major browsers have large tables they've manually compiled to make sure you can't register a cookie on say, .ac.uk, or .pp.se to track people across domains. Who is going to manage the security of these new TLDs?
The author is blowing the problem out of proportion. If Google uses the .blog TLD abusively or in a way that is irrelevant, users will go to another .TLD. There is no limit to the number of places users can go. The value in the mindshare of a word like "blog" is overblown, and it isn't the same as the monopoly power granted by an exclusive patent. If .blog domains suck, then we won't go there. Nobody will host blogs there. Nobody will care. It won't be a hip place. On the other hand, if someone makes a .TLD called something arbitrary (say .ddffdd) and everyone likes it, then guess what, it won't matter squat what is on .blog, everyone will go to .ddffdd. That's all.
Lets not act like the landing page URL is such a big deal. It is if you built up a trademark and if everyone is used to going to .blog for every single blog. But nobody is used to that. And nobody cares. And even if they are used to that habits can very quickly change. There were lots of so called "monopoly" destinations a decade ago and not even one of them is still standing today. Its a whole new ballgame on the Internet. And every ten years its going to be a whole new ballgame again.
IMHO, the biggest problem of gTLD is the inequality created on the Internet. Google.com's registry cost is less than $8/year, just like any random .com domain. Everyone has the fair chance to get a domain and start a business online.
But when the gTLDs are available for grabs, the big companies will definitely get their trademarks as TLDs, even when they have no intention to open that to the public. gTLD scheme is not new - .COM is managed by Verisign, the same way as the potential .google being managed by Google. What's new is the removal of the requirement of opening for public registration, and much less regulation and evaluation of eligibility.
When the big companies have their special branding schemes, small businesses lose out. Suppose that the new gTLD scheme is a huge success (which IMO not likely), .google, .apple and .amazon are official tags for big companies, and .COM will be the small business paradise. This divide is definitely harmful for internet development.
I've said it on the previous Dave Winer's article about it. I really don't get what's the big deal as long as the all the previous TLDs continue to exist and operate just like they have been operating.
Some enterprises will have the new fancy TLDs, good for them. The rest of us will have to keep going as we have been going (we might even get access some of the new TLDs.) I really doubt that, for example, having my.book will be a significant competitive advantage versus having mybook.com
What am I missing? (Yes, greedy ICANN is evil, but what else?)
The current default is that a company can own its name under all important TLDs. That is feasible because there are only so many generic ones (.com,.net,.org, maybe .info and .name) and a few of countries you want to do business with. So you could cultivate the expectation that you need and once will have your brand name (in its variations) under all of these.
That will no longer be possible. Also, it will not possible to just assume that the TLD is the new domain name, with every middle-sized company snatching up their brand name as TLD.
So when that expectation is no longer there, there will be no problems at all. Some of these TLDs will work similar to the existing generic TLDs, with the possibility for third parties to buy regular domains. Others will operate more like private tropical islands, used as some kind of status symbol.
The whole thing has one likely positive consequence, though. It might weaken the idea that you could somehow own a name or word. Some arbitrary slightly deep-pocketed entity grabs a name that most legitimately interested people cannot afford. That happens often enough, and nobody gives much about that name plaque any longer. Plus, old-fashioned brand operators will have a tough time buying up all "their" domains under the newly formed TLDs until they eventually give up and realize they cannot control "their" name in every conceivable abstract namespace. Everyone wins.
I wonder what the chances are of these new gTLDs being as successful as, say, .biz or .mobi? In other words, not at all successful.
One reason to be optimistic about the failure of gTLDs is recognition. I think people understand that something ending in .com or .net or a ccTDL is an 'internet address', but how do you indicate this with a gTLD like 'book' or 'blog' where there is no visual hint to indicate that it is a domain?
The internet can be made a much better & safer place if the new registries enforce proper regulations. Of course, there will be exceptions but there are many major players in this race who will work towards this objective. If you look at it in another way, it opens up a whole new world of opportunities not just for ICANN or the registry but for us, developers! Lets say I made a super awesome Iphone App and want a website for it, how much is the probability of me finding a relevant .com domain name for it? On the other hand, I have higher chances of getting a very relevant name with a .app extension & if this extension is popular amongst app users (which the .app registry would definitely ensure) then I have just multiplied chances of my app being discovered easily! Now some would argue that apps are to be found in app markets but then this was just a simple example, the possibilities are endless. I believe we are in a historic moment of the evolution of internet, something which would make the internet better, easier & safer for all of us.
Sex, love, laughter, babies, books, songs, cars, poetry, etc..., they're too important, too basic to life. Not the kinds of things any company, for crying out loud, should be able to claim to own.
No one claims that Verisign owns commerce just because they manage the .com domain. Or that Google would own all blogs if they get to sell .blog domains.
These are entirely different things: .com is managed for the benefit of all commercial entities (and Verisign). Google's intended use of .blog would only be to assign domains to Blogger properties.
From Google's application for the .blog TLD (application #527):
> The purpose of the proposed gTLD, .blog, is to provide a dedicated Internet space where Google can continue to innovate on its Blogger offerings. The mission of the proposed gTLD is to provide a dedicated domain space in which users can publish blogs. All registered domains in the .blog gTLD will automatically be delegated to Google DNS servers, which will in turn provide authoritative DNS responses pointing to the Google Blogger service.
In addition to being uninformed about the basic facts here, I think you are overestimating the market for $7 - $50/yr blog hosting and underestimating the brand value of an exclusive Blogger TLD.
Isn't the real issue here that it's just unnatural to use fixed combinations of letters to refer to websites? There is a reason that natural language is so hard to process for computers: It's inherently context-dependent and never has meaning per se. The domain name system on the other hand tries to impose a fixed naming system on us that is absolutely incomprehensible to anyone without technical knowledge.
The only solution to this mess I can see is a non-profit HTTPS server certificate verification authority in combination with a complete revamp of the user experience for browsing web ressources. I should be able to just search for the page I want to visit instead of having to remember its exact name.
With a few exceptions, TLDs are not canonical in any real sense anymore. That ship sailed years ago when .org and .net were opened up to general registrations. And it continues to sail today: is bit.ly really based in Libya? No of course not. So who cares if Amazon.com the company secures .amazon the domain. In practical usage it's not that different from what they've done snapping up amazon domain names for years.
Its all about "getting more money" anyway. Capitalism issue. Things always get corrupted into that. Not that any other system we have had so far was better.
* Domains + sub-domains allows for wildcard SSL cert registrations (e.g. a cert that covers .example.com).
There's also cookie registration.
* How do I differentiate between my_blog_host, my_blog_host_2 and really.my_blog_host? Are they all owned by the same person? Is really.my_blog_host associated with my_blog_host?
Bzzzt, wrong. Trademarks are industry-specific: that's why Apple Records can exist alongside Apple Computer. If there's no likelihood of confusion, there's no conflict.
And of course, trademarks are not fully international.
Winer is right about ICANN being wrong. There's just so much else that ICANN is wrong about.