Tuesday, June 30, 2020

Booking.com: validity continues to be disconnected from scope of rights

U.S. Patent & Trademark Office v. Booking.com B.V., No. 19–46 (Jun. 30, 2020)

Kind of what I expected, though maybe a little worse in its disregard of scope issues.

Ginsburg writes the majority (Sotomayor concurred and Breyer dissented). 

“Generic.com” is only generic for a class of goods or services “if the term has that meaning to consumers,” and the evidence hree showed that consumers don’t perceive “booking.com” to signify online hotel-reservation services as a class. So, the Court clears the way to register a lot of these domain names, because this is not on its face a holding about how booking.com has secondary meaning, it is a holding about how the PTO didn’t meet the burden of showing that consumers perceive booking.com as the name of a service. 

Along the way, the Court cites the parties’ agreement that “for a compound term, the distinctiveness inquiry trains on the term’s meaning as a whole, not its parts in isolation.” And “the relevant meaning of a term is its meaning to consumers.” Although the Lanham Act provision that “[t]he primary significance of the registered mark to the relevant public . . . shall be the test for determining whether the registered mark has become the generic name of goods or services” speaks directly by its terms only to cancellation, 15 U.S.C. §1064(3), it sets the standard for genericity generally. However, today’s rule “does not depend on whether one meaning among several is ‘primary.’ Sufficient to resolve this case is the undisputed principle that consumer perception demarcates a term’s meaning.” 

Since consumers don’t perceive Travelocity as a “Booking.com,” it is not generic. Goodyear’s India Rubber Glove Mfg. Co. v. Goodyear Rubber Co., 128 U. S. 598 (1888), held that a generic corporate designation added to a generic term does not create a protectable term because it adds no additional meaning. But .com might “also convey to consumers a source-identifying characteristic: an association with a particular website,” because only one entity can have a particular domain name at a time. Since consumers know that, a consumer can infer entity-designating function. (Footnote: this doesn’t constitute utilitarian functionality, which isn’t at issue here.) 

Comment: this reasoning goes even further than the above and suggests what the district court held, which is that individual domain names are likely to be inherently distinctive because of this automatic (de facto) assumption. But the ability of a term to distinguish this entity from other entities or domain names doesn’t seem to come into this reasoning at all, even though it probably should. How easily will consumers remember: was it flower.com or flowers.com? 

Anyway, Goodyear’s rule is more modest than the PTO argued: “A compound of generic elements is generic if the combination yields no additional meaning to consumers capable of distinguishing the goods or services.” 

There is no automatic classification of generic.com as nongeneric, the Court says, but the test doesn’t give any weight to the genericity of what comes before the dot; it’s only whether the full term is perceived as the name of a class. “Evidence informing that inquiry can include not only consumer surveys, but also dictionaries, usage by consumers and competitors, and any other source of evidence bearing on how consumers perceive a term’s meaning.” As to surveys, they “can be helpful evidence of consumer perception but require care in their design and interpretation.” We get a cite to our amicus brief noting that “survey respondents may conflate the fact that domain names are exclusive with a conclusion that a given ‘generic.com’ term has achieved secondary meaning.” But difficult questions aren’t posed here because the PTO didn’t contest the lower courts’ assessment of consumer perception of booking.com specifically. 

The PTO’s real objection wasn’t to Booking’s exclusive use of “Booking.com” as a mark, but to the risk of “undue control” over similar language “that others should remain free to use.” However, the Court thought that trademark law’s existing protections for descriptive uses sufficed: first, only confusing uses are infringing, and weaker/more descriptive marks are less likely to be confused: 

When a mark incorporates generic or highly descriptive components, consumers are less likely to think that other uses of the common element emanate from the mark’s owner. Similarly, “[i]n a ‘crowded’ field of look-alike marks” (e.g., hotel names including the word “grand”), consumers “may have learned to carefully pick out” one mark from another. And even where some consumer confusion exists, the doctrine known as classic fair use, protects from liability anyone who uses a descriptive term, “fairly and in good faith” and “otherwise than as a mark,” merely to describe her own goods. 

[I note that putting a generic term in one’s domain name is likely to be pretty risky for descriptive fair use, given “otherwise than as a mark.”] 

Booking.com concedes that its mark would be “weak” and that “close variations are unlikely to infringe.” [And we will now rely on courts to ensure that it and other users of generic.coms stick to that concession—but since the US doesn’t have a cause of action for unjustified threats relating to IP, the C&D letters can be pretty aggressive regardless of the likely litigation outcome, and it’s really hard to get a fee shift as a successful defendant, especially if (as they often do) courts treat having a valid mark as making any assertion of infringement reasonable.] 

The competitive advantage conferred by seizing on a descriptive domain name doesn’t justify refusing registration. All descriptive terms “are intuitively linked to the product or service and thus might be easy for consumers to find using a search engine or telephone directory,” but they’re still registrable. And the exclusive connection between a domain name and its owner “makes trademark protection more appropriate, not less.” [And here is the issue that European systems will find much easier to resolve than ours: that statement makes perfect sense if we aren’t worried about scope at all. But we should be!] 

Unfair competition law isn’t a good enough substitute for registration because registration confers valuable benefits, such as under ACPA and the UDRP. 

Sotomayor, concurring. There’s no (nearly) per se rule against generic.com registrations. However, consumer-survey evidence “may be an unreliable indicator of genericness.” But the Court’s opinion doesn’t make surveys “the be-all and end-all.” Given the availability of other evidence and the weaknesses of surveys, the PTO might have been right about genericity here and the district court may have been wrong; that’s just not the question before the Court. [Although it is the way to bet about how errors will go, given the parties’ resources. In other cases like Wal-Mart, the false positive issue has led the Court to more restrictive rules—but because the Court generally does want to hold out the prospect of some protection in such cases, the overall result of Sotomayor’s concurrence at least is consistent with Wal-Mart and Qualitex: you should have to show secondary meaning for your specific claimed mark and only then can you get trademark rights.] 

Breyer, dissenting: What booking.com is, is obvious from the name. Genericity “preserves the linguistic commons by preventing one producer from appropriating to its own exclusive use a term needed by others to describe their goods or services.” Adding .com shouldn’t be enough to do so. 

Distinguishing descriptive from generic terms isn’t always easy, especially with a compound term. For a compound, courts have to determine “whether the combination of generic terms conveys some distinctive, source-identifying meaning that each term, individually, lacks. If the meaning of the whole is no greater than the sum of its parts, then the compound is itself generic.” 

Breyer would have held that appending “.com” to a generic term “ordinarily yields no meaning beyond that of its constituent parts.” The combination of “booking” and “.com” “does not serve to ‘identify a particular characteristic or quality of some thing; it connotes the basic nature of that thing’—the hallmark of a generic term.” Any “reasonably well-informed consumer” would understand that trademark.com is the website of the trademark owner, which is why courts generally ignore the TLD when analyzing likely confusion. 

And here we get to the heart of the disagreement: because .com will be ignored in an infringement inquiry, it should also be ignored for validity. (Absent some exception such as where the TLD interacts with the second-level domain in a meaning-changing way; also new gTLDs like “guru” might behave differently.) In other words, Breyer wants to think about scope when assessing validity. 

The uniqueness conferred by the domain name system doesn’t change the Goodyear rule. “Wine, Inc.” likewise “implies the existence of a specific legal entity incorporated under the laws of some State.” [This is an intuitive weakness of the majority’s reasoning: it isn’t explicitly willing to turn its empiricist, consumer-perception-is-all gaze on the Goodyear rule, but Breyer seems likely to be correct that Wine, Inc. is likely to produce different survey results than “Wine.” However, my sense is that Ginsburg would also allow Wine, Inc. to present its consumer survey, perhaps with an eye to the problem of de facto secondary meaning.] 

To Breyer, “functional exclusivity does not negate the principle animating Goodyear: Terms that merely convey the nature of the producer’s business should remain free for all to use.” The majority’s “fact-specific approach” rejects that principle. Although consumers might not call Travelocity a booking.com, “literal use is not dispositive.” [We might call that production versus recognition: consumers may understand a lot of stuff they wouldn’t say.] 

Breyer correctly notes that the facts that supposedly convert some generic.com domain names into descriptive marks are unlikely to vary from case to case: 

There will never be evidence that consumers literally refer to the relevant class of online merchants as “generic.coms.” Nor are “generic.com” terms likely to appear in dictionaries. And the key fact that, in the majority’s view, distinguishes this case from Goodyear—that only one entity can own the rights to a particular domain name at a time—is present in every “generic.com” case. 

So what would vary? Survey evidence. But survey evidence “has limited probative value in this context,” given the phenomenon of de facto secondary meaning. Thus, the TTAB and some courts have concluded that “survey evidence is generally of little value in separating generic from descriptive terms.” Here, while Booking’s survey showed that 74.8% of participants thought that “Booking.com” is a brand name, 33% believed that “Washingmachine.com” was a brand. What’s the difference? Booking.com isn’t inherently more descriptive than “Washingmachine.com” or any other generic.com. Respondents were likely reacting to having heard of Booking.com, suggesting that washingmachine.com could undergo the same transformation by investing heavily in advertising. Association of booking.com with a particular company is simply not inconsistent with genericity.


Thus, quoting McCarthy, the majority rule “[d]iscard[s] the predictable and clear line rule of the [PTO] and the Federal Circuit” in favor of “a nebulous and unpredictable zone of generic name and top level domain combinations that somehow become protectable marks when accompanied by favorable survey results.” 

Also, this rule threatens “serious anticompetitive consequences in the online marketplace” by adding to the non-registration advantages of doing business under a generic name: basically, easier access to consumers’ memories, searches, and trust. Booking says it won’t threaten similar uses, “[b]ut other firms may prove less restrained,” and of what use is trademark registration distinct from domain name registration other than to “extend its area of exclusivity beyond the domain name itself”? 

The majority says that infringement/descriptive fair use will police the scope of protection, but we’ve already seen boundary-pushing. (Citing Advertise.com v. AOL, LLC, 2010 WL 11507594 (CD Cal.) (owner of “Advertising.com” obtained preliminary injunction against competitor’s use of “Advertise.com”), vacated in part, 616 F. 3d 974 (CA9 2010).) “Even if ultimately unsuccessful, the threat of costly litigation will no doubt chill others from using variants on the registered mark and privilege established firms over new entrants to the market.” 

Breyer’s dissent is particularly notable for its repeated use of “anticompetitive” and variants, suggesting the potential for the revival of a mid-twentieth-century analysis of trademarks as having the capacity to confer unwarranted market power.


Anonymous said...

Looks like they did not touch the issue of two (and more) word domains at all! With Example.com as a trademark means businesses operating on CoolExample.com, MegaExample.com, LondonExample.com, BuyExample.com and others could be threatened by that. This is disastrous.

Anonymous said...

Hi, just wanted to clarify your statement that the Lanham Act's primary significance provision, which appears in the Act's cancellation section, "sets the standard for genericity generally." I believe you're referring to footnote 3 of the decision. I read this not as a holding by SCOTUS, but simply as an observation by SCOTUS about how the appeals courts have been operating and what the PTO has in its manual. Do you agree or disagree? I say this because footnote 3 also says: "We need not address today the scope of the primary-significance test’s application." This would seem to allow the possibility that the appeals courts and PTO have it wrong by administering the primary significance test in registration cases. The Chief Justice raised this question at oral argument. In your view, is this still an open question that could be tested in a later case, or not? Thank you!

Rebecca Tushnet said...

Good question; I don't think raising the question is precluded, but I read that more as saying "we aren't opining today on what it means to have primary significance," especially given the PTO's concessions on what the survey showed. I thought of this more as: suppose a different survey showed that 25% thought a .com was generic, 35% thought it was a brand name and recognized it from previous encounters (it had secondary meaning for them), and 40% didn't think it was generic but also had no associations with it (no secondary meaning)--I don't think Booking.com answers whether the "primary significance" test has been met in that kind of case.

Anonymous said...

Thank you very much for your response, which I found helpful and insightful. It would be interesting to see later cases address the issue that you raise. Many thanks.