November 23, 2010
Wildcarding Subdomains Is OK; Reverse Domain Name Hijacking Isn't--Goforit v. Digimedia
By Eric Goldman
Goforit Entertainment LLC v. Digimedia.com LP, 2010 WL 4602549 (N.D. Tex. Oct. 25, 2010). See the related personal jurisdiction ruling from 2007 featuring a completely different but still ridiculously large and expensive cast of lawyers.
This is a super-interesting dispute involving two not-so-interesting litigants. The plaintiff Goforit runs a type of meta-search engine at goforit.com. After spending 5 minutes at the site, I couldn't identify a single reason why anyone would want to use it. Also inexplicably, Goforit appears to be quite pleased with its trademark rights in "Goforit," a term that seems more like an exhortation than a trademark.
The defendants own or operate many domain names, including "org.com," "com.org," "gov.org," and "org.net." All of these domain names have wildcarded subdomains, meaning that XYZ.com.org will lead to a working web page, no matter what "XYZ" is. (See my descriptive and normative discussion about wildcarding in my Deregulating Relevancy article). The resulting com.org web page presents a mostly useless directory of CPC links. I noticed that the pages now include a disclaimer at the top saying "xyz.com.org was not found on our servers. www.com.org is shown below" and a link at the bottom saying "Information on how you reached this site" which says:
Occasionally we receive inquiries from users who do not understand why they have accessed our site. Please be advised that you are not reaching our site as a result of spyware. We are not exactly sure why you have been directed here, however, we believe it is a result of the autosearch feature of Internet Explorer. If a site entered into the address bar cannot be accessed, Explorer apparently appends ".org" to the name and then tries to access that site. In the case of a search ending in .com, Explorer thus accesses our domain, "com.org."
Taking this statement at face value (which I recognize is questionable), this technical implementation seems like a goof by Microsoft and anyone else who automatically appends a TLD to an address bar entry. The court’s opinion doesn't mention either the disclaimer or the footer link, so they may post-date the lawsuit commencement. FWIW, the latest page in Archive.org from Aug. 2008 lacks the top disclaimer, although it does have the footer link.
Because the defendants are wildcarding their subdomains and various browsers or toolbars may be automatically adding ".com" or ".org" extensions, the net result is that users could enter "goforit" or "goforit.com" into their address bar and end up at the defendants' wildcarded com.org page, where they were presented CPC links. On the one hand, this wildcarding could have contributed to a type of consumer confusion--consumers putatively looking for goforit's navigational directory got the defendants’ navigational directory instead of, say, getting a 404. On the other hand, who cares? It's hard to believe many people were actually looking for goforit.com and yet had their browser redirect them to goforit.com.org, and those few brand-loyal users should have had no problem instantly spotting that they were in the wrong place.
So to boil the dispute down, a low-value no-name search engine is suing a low-value no-name domainer for a few allegedly misdirected users who quickly self-corrected. It's a little hard to work up a lot of sympathy in either direction. In this respect, the lawsuit vaguely reminded me of the unsympathetic marketers battling over dmv.org.
Two more oddities. First, making all favorable inferences for the plaintiff, how many users could possibly have been "diverted" from plaintiff to defendant? Second, after Goforit complained, the defendants "blocked" goforit as a wildcard, causing users who would have gotten to the defendants' site to get a 404 instead. So, adding it all up, exactly how much money could Goforit be legitimately seeking? I'm sure the defendants love the traffic they get from their scheme, but I can't imagine that the slice of their revenue attributable to "goforit" is greater than de minimis. And for this peppercorn, the plaintiff has been suing the defendants since 2006 through multiple courts--with no less than *TEN* lawyers from *THREE* different law firms listed as its counsel on this opinion. (FWIW, I count 4 lawyers from four different law firms working on the defense side, although there were multiple defendants who might have wanted their own lawyers). The entire litigation efforts seems like a nuclear flyswatter.
Fortunately, the parties' exorbitant expenditures do produce one good outcome: this opinion is one of the most conscientious and thoughtful ones I've seen this year. Huzzah for Chief District Judge Sidney A. Fitzwater!
Unfortunately for Goforit, the judge rules decisively in favor of the defendants. The court rejects Goforit’s claims over the wildcarding and sends defendants' counterclaims to the jury. I’m going to get into the minutiae of the opinion, so unless you are a hardcore domain name/trademark nerd, you should probably stop reading here.
The court rejects the cybersquatting claim because Goforit was complaining about third-level subdomains. The court says flatly: "as a matter of law, third level domain names are not covered by the ACPA." The court concludes:
Because third level domains--whether specifically designated or using Wildcard DNS--are not "registered with or assigned by any domain name registrar," a straightforward reading of the text shows that GEL cannot recover under the ACPA for defendants' use of Wildcard DNS in a third level domain.
Goforit's trademark claim fails for lack of use in commerce. Goforit tried two arguments to establish use in commerce: "(1) defendants "used" the mark by registering a TLD website such as "com.org" and enabling Wildcard DNS, intending that third parties would enter addresses containing others' trademarks, such as "goforit.com.org"; and (2) defendants "displayed" the GOFORIT mark in the sale and advertising of their services by programming the address bar to display "goforit.com.org" after the Wildcard DNS directed the user to the "com.org" website."
On the first point, the court says that setting up a wildcarding scheme does not evidence the defendants' intent to use Goforit's trademarks. Citing the 2005 1-800 Contacts v. WhenU case, the court says:
Even if defendants could have foreseen that some third-party web users might type in a trademarked name to trigger the Wildcard DNS, the fact that the TLD websites' Wildcard function processes a trademarked input does not constitute a "use" in "sale or advertising" of services....the Wildcard function in the instant case does not handle the third-party user's trademark input as a trademark, but merely as a web address that internally redirects to the domain name's homepage....Defendants have never tried to leverage others' trademarks into profit by selling specific trademarked keywords to advertisers.
I note that the com.org website now shows the disclaimer--referencing the subdomain and possible trademark--on each page. I wonder if this would change the court's analysis.
On the second point, the court says:
A reasonable jury could only find that, in any instance when "goforit.com.org" appears in the address bar, this is because the user has somehow entered "goforit.com.org" into the address bar at some point, whether intentionally or by inadvertent misuse of shortcut keys, to trigger Wildcard DNS in the first place. It would be different if defendants had configured their website to display "GoForIt.com" on the user's address bar after redirecting users to the "com.org" site and displayed other indications of affiliation with GEL and its GOFORIT mark. But a reasonable jury could not find that there is anything deceptive in itself about calling a website what it actually is: when a user enters "goforit.com.org" into the address bar, the user may be redirected to "com.org," but that does not change the fact that the content the user is seeing is the actual content assigned to the "goforit.com.org" address.
False Designation of Origin
The court also rejects Goforit's claim that the defendants made a false designation of origin by retaining the goforit subdomain throughout a user's visit. The court says that Goforit did not produce any convincing evidence either of consumer confusion or any harm to it.
The court first does a traditional multi-factor trademark confusion test; a few comments about that:
Mark Strength: "Without any evidence of the GOFORIT mark's consumer recognition power, there is no proof that would enable a reasonable jury to find that any of the redirected users was even aware of GEL's GOFORIT mark."
Mark Similarity: the court says the relevant marks are "Goforit" compared with “com.org." The court effectively ignores the subdomain as irrelevant. Even if the Goforit trademark is compared with goforit.com.org, the court says a "reasonable jury could not find that a user would construe an address bar display, especially one that merely retains what was just inputted by the user, as a "mark" when nothing else in the page content indicates GOFORIT sponsorship or affiliation."
Product Similarity: "A reasonable jury could only find, however, that the similarity of product design is attributable to the fact that defendants' TLD Domain Name websites and GEL's GoForIt.com website are both web directories; the functional design choices are common to many web directory websites rather than distinctive to GEL, and GEL has not produced any evidence that defendants have copied any product feature that uniquely identifies GEL's services."
Actual Confusion: "According to Grant's testimony, users on message boards expressed confusion as to how they arrived at websites such as "com.org," but there is no summary judgment evidence that users mistook "com.org" to be affiliated with or approved by the websites they intended to reach. On the contrary, the fact that some users, after experiencing Wildcard DNS, suspected that they had been infected with spyware suggests that it was immediately apparent to users that defendants' websites, which lacked trademark and trade dress similarity with their target sites, were of a different origin." Perhaps this is one of the rare situations where being confused with spyware is a positive!
The court then turns to false advertising proper, saying "a reasonable jury could not find that the display of "goforit.com.org" is a literally false statement." The court considers implied falsity and rejects the elements of that:
GEL has failed to present any evidence that the redirected consumers would have found the display of "goforit" in the web address material in deciding whether to click on the advertisements on "com.org." Nor has GEL adduced evidence that would enable a reasonable jury to find that any of the affected users was even aware of "GoForIt.com" as a competing entity....because a reasonable jury could only find that "go for it" is a common expression in conversational English, and because myriad other websites unaffiliated with GEL incorporate "goforit" into their web addresses, a reasonable jury could not find that any of the users from the log files who typed "goforit" intended to go to GEL's website.
Reverse Domain Name Hijacking
After completely rejecting Goforit's claims, the court turns its attention to the defendants' counterclaims. Defendants sued for reverse domain name hijacking under 15 USC 1114(2)(D)(iv). As part of its scorched earth tactics, Goforit asked the defendants' registrar (Tucows) to impose a registration lock on all of defendants' domain names--including com.org and nearly 300 others.
Tucows did, in fact, impose the registration lock, which allegedly caused all kinds of problems for the defendants. I'm trying to figure out how Goforit convinced Tucows to lock all of the defendants' domain names. On the surface, Tucows isn't looking so great here either. [see update below]
The court says that even if Goforit had a plausible argument against com.org based on goforit.com.org (something the court left open), Goforit had no plausible argument that the other domain names violated its trademark rights. As a result, the court denied Goforit's summary judgment motion, putting the reverse domain name hijacking claim to a jury trial.
I did a little research on the ACPA reverse domain name hijacking provision. I found only 6 other cases in Westlaw citing to the damages provision in subsection (iv) (as opposed to the declaratory relief in subsection (v), which has been successfully invoked only a few times). As far as I can tell, no other (iv) case has resulted in a positive outcome for the domain name registrant. Therefore, I believe the defendants' success so far on the (iv) reverse domain name hijacking claim is the first of its kind.
Tortious Interference With Contract
The court also rejected Goforit's summary judgment on the defendants' claim for tortious interference with the defendants' contract with Tucows. There were some old cases that implied that the ACPA reverse domain name cybersquatting provisions preempted state law equivalents, so this ruling also appears to be novel.
UPDATE: I had an extended email discussion with Tucows about its role in this situation. My contact sent me the following statement:
"Tucows responded in this case as per our policy. When we receive notice of filed litigation, NOT simply a C&D or draft litigation, we lock the affected name(s). We also notify the registrant. Please remember that locking the name does not in any way disable the name, but simply ensures that the name stays in place while the rights are being sorted out. In these circumstances we are also happy to accommodate changes that a registrant may need to a setting such as DNS. Also, in the rare circumstance where this policy creates difficulties for a registrant, Tucows actively works with the registrant to find a solution that balances the registrant's rights with the need to see that an effective legal system sits behind all property rights, including domain names.
While we don't comment on specific matters, Digimedia is, and continues to be, a long time Tucows customer."
TrackBack URL for this entry: