Slashdot Mirror
ICANN Investigates Insider Domain Name Snatching
Tech.Luver sends us word that, hot on the heels of reports that Verisign may be planning to sell DNS root server lookup data, ICANN has opened an investigation into a suspected practice by registrars it calls "domain name front running." The suspicion is that insiders at some registrars are using information from whois searches to snatch up desirable domain names before interested customers can register them. Here is ICANN's announcement of the investigation (PDF). ICANN asks that anyone who suspects they have been victimized by domain name front running to email them with details.
When a domain is snatched, usually it doesn't matter if the original owner gets it back or not. That's not the point, in most cases. Thieves will use the domain to drive traffic to their astroturfing/spam network and drive their PR up in the process. That stays in memory indefinitely and has a beneficial impact on any site like that.
If the owner gets their network back, they still have the stigma of the bad activity associated with the domain.
Preventing domain theft is going to only get increasingly more difficult as technology becomes more complicated.
The dangers of knowledge trigger emotional distress in human beings.
I don't think it'd work. It'd be very easy to load them into a table, filter them against dictionary words, and sort them by # of hits.
Human eyeballs could pull the top 1000, do a quick spot check on the list, remove garbage names, and register the rest. Once setup, it'd take about 10-15 minutes of human intervention a day.
Camping on quad since 1996.
why not make a domain named www.ICANNOT.org and just make it a listing/cache of domain names already taken so users looking for a domain can see if a name is already taken...
Oops, too late, already taken...
Politics is Treachery, Religion is Brainwashing
Why have domain name service at all?
.af.mil,...) containing a nice orderly heirarchy (.mit.edu -> .ee.mit.edu -> .rle.ee.mit.edu -> myhost.rle.ee.mit.edu). The real world doesn't look like that at all, which is why most people these days couldn't even tell you why there are dots in the URL and what purpose they were supposed to serve.
That is, why do we have this superelaborate expensive annoying structure, the only purpose of which is to translate one string (the hostname) into another (the IP address)? Sure, a nice 32 bit number (0x4a7d1368) is easier for programs to work with than a variable-length alphanumeric string ("www.l.google.com").
But so what? The only legitimate purpose of technology is to make our lives easier, not to serve as a temple in which we practice the complicated correct forms of worship. My 2007 Odyssey is way more complicated under the hood than my 1968 Volkswagen was (and of course that means car designers and car mechanics have a much more complicated and demanding job these days), but the 2007 car is much easier for the user to drive and take care of than the 1968 car. That's as it should be. Technology should be designed and evolve so that the ease and convenience of the user is the first priority. How easy or cool it is to implement should be a distant secondary goal only. (But programmers should not complain, because the more complicated and difficult a scheme is to implement, the better-paid the job of implementer is.)
The alphanumeric string that human beings find easy to remember and use should be the "real" address of an Internet host, and it should be up to the robots and programs behind the scenes to cope with the complexity of correcting routing packets to the destination using only this string.
More fundamentally, the idea of making one giant and (literally) global hash in which each host is mapped to a unique ID tag is violently contradictory to the way people naturally think. We naturally think in terms of local variables and namespaces. It perfectly possible for a bookstore in Liverpool to have the same name as a bookstore in Atlanta, because human beings consider the bookstore name a local variable and use the context ("Am I in England or Georgia?") to figure out the correct global meaning. Internet hostnames should work in a similar way; it should be possible for the Liverpool and Atlanta bookstores to have the same name on the Internet, too, with some method of choosing context to resolve ambiguity. Yes, I realize the dotted aspects of hostnames was supposed to do something like that ("foo.bar.com" versus "foo.baz.com"), but it clearly didn't work out that way. Perhaps because it was designed by people for whom the world was broken up into a few very large organizations (.mit.edu,
I also know lots of schemes that rely on the present madness would be broken. Tant pis. Can't make an omelet without breaking eggs.
A good friend of mine had a very successful website with 300,000+ users that made him over $100,000/year. The domain had been registered using some free email account that he stopped using. Eventually the email address was reclaimed and made available again and some guy registered it and hijacked his domain. It took him over a year and a half plus thousands of dollars in legal fees to finally get his domain back. By that time the domain was worthless because all of his customers had gotten fed up with the service outage and left. About the only thing going for it now is a Google pagerank of 7. He's also looking for a job.
The moral of the story is to keep tabs on your email addresses.
Of course it does. Any IP communications which uses a name rather than an IP number is using some type of name resolution. Since the real question posed by this situation is "has this domain name been registered", you can't answer it without consulting with the domain name resolution system. And that is either a WHOIS query at a registrar or a name resolution check through a DNS, either incidental (ping my.foobar.foobaz.org) or intentional (dig my.foobar.foobaz.org).
And I have doubts about using DNS to verify it anyways. Domains aren't hosts; the domain "foobar.foobaz.org" might have many host names within it (such as "my", mentioned above), but you can't guarantee that you can guess them. Yah, www.foobar.foobaz.org seems like a likely place, but if I'm front-squatting the foobar.foobaz.org domain, I may not host a site at that address. (Of course, I'd be an idiot not to, since hits on that site make measuring interest in the domain easy, and I can aways linkfarm or upload drive-by malware for a bounty.)
Welcome to the Panopticon. Used to be a prison, now it's your home.