VeriSign Granted a Patent Covering SiteFinder

An anonymous reader writes "Remember VeriSign's SiteFinder? Turns out that a couple of months back VeriSign was granted a patent on resolving unregistered domains. This came about thanks to its acquisition of eNic, operator of the .CC Domain. How long before Verizon, Earthlink, and OpenDNS are hit up for licensing fees?"

Comments

Oh the Humanity

[DECS]

Maybe we should patent REALLY BAD IDEAS to prevent them from spreading. Of course, it's hard to imagine in advance that ISPs and a company like VeriSign would make a business from poisoning and subverting DNS.

Flash Wars: Adobe in the History and Future of Flash

That might be a good thing...

[Whatanut]

If it stops DNS providers from using this practice... I'm all for it.

Many Reasons this is Appalling

[billstewart]

There are so many things wrong with this. The first one is that it doesn't actually work as indicated in Claim 1, because it's operating at the wrong levels of the protocol stacks. DNS maps between names and IP addresses, and is used for many different kinds of Layer 4, 5, and 7 applications, but URLs are a Layer 7 function typically supported by browsers, and the identification of what kind of service the client is interested in is not known at name resolution time, or even what Layer 4 transport protocol or Layer 7 application protocol, and in fact the methods used in the patent have the DNS operator's web server decide what kind of response web page to provide in response to a URL included in a HTTP request, even though the client's DNS request might not have been intended to be used for HTTP. When Verisign implemented their annoying breakage of DNS functionality, they supported HTTP on port 80, and had a stub email server that did a sloppy approach to rejecting connections, and AFAICT didn't provide other services, such as correct rejections on SSL's TCP Port 443 or SSH's TCP port 22. It's not clear that they even did the right thing at Layer 3 - if you were trying to "ping misspellllled-example.com", they not only should have answered the DNS request with a "No Such Domain" error message, but if you sent it a ping, it shouldn't respond (I forget if they responded to pings or not; many systems don't do that for self-defense.)

Another reason this patent shouldn't have been accepted is that wildcard domains were a standard capability, and having a web server try to provide useful information in a 404 page was probably a known capability, or at least obvious to someone skilled in the trade. Responding to a DNS request with the IP address of a web server that isn't the one the customer was looking for might not count as "obvious to someone skilled in the trade" because it's obviously wrong.