Rebecca Tushnet

copying without infringement, falsity without materiality in adhesive case

Posted on March 14, 2019 by rtushnet

J-B Weld Co., LLC v. Gorilla Glue Co., 2018 WL 6356768, No. 17-CV-03946-LMM
(N.D. Ga. Oct. 17, 2018)

Among other things, this case is a reminder that when you sue a competitor for false advertising you can often expect counterclaims–here the only thing that (partially and provisionally) survives summary judgment.

J-B Weld sells a two-part epoxy adhesive: when epoxy resin
and paste hardener are mixed together, a reaction produces a permanent adhesive
bond. Both J-B Weld Original and its newer KwikWeld are sold in major retail
stores and online sites. The current
Original trade dress dates from 2012, and J-B Weld defines it as: (1) two
squeezable tubes in a clear blister package with the two tubes arranged in a
“V-shape”; (2) red coloring on one tube and black coloring on the other tube;
(3) a clear plastic blister package holding the tubes curved downward; (4) a
drop-down technical information box, located between the two tubes, that has
four lines of information separated by white lines; (5) a colored banner across
the bottom end of the card; (6) a colored banner bar that includes “Weld”
within the banner; (7) the phrase “Steel Reinforced Epoxy” and “World’s
Strongest Bond” on the card; (8) a list of uses near the bottom right corner of
the card; and (9) a card that is five inches.

J-B Weld’s packaging has described J-B Weld Original as a
“Steel Reinforced Epoxy” and the resin tube was labeled with the word “STEEL”
since 2009. That was also when Gorilla
Glue allegedly became aware that J-B Weld intended the word “steel” on its
packaging to indicate the presence of an iron-containing additive. Gorilla Glue
neither contacted J-B Weld about nor challenged its use of the word “steel” on
its packaging until this case. J-B Weld has also featured “World’s Strongest
Bond” on the packaging for all its adhesive products since 2011, including its
two-part epoxy adhesive sold in a plunger syringe launched in 2012; Gorilla
Glue learned of the use of the phrase when the packaging debuted. Gorilla Glue likewise
didn’t object until this case.

In 2012, Gorilla Glue introduced a new Gorilla Epoxy
formula, which improved performance and provided a clear, rather than
yellowish, formula. It then began working on a two-part clear syringe product
with methacrylate chemistry (MMA) that it referred to as “Plastic Plus” or
“Heavy Duty Epoxy” during development and consumer testing. It ultimately
became the two-tube adhesive product known as “GorillaWeld.” During the design
process, Gorilla Glue’s Brand Manager instructed its graphic designer to
develop three blister card designs, including one that was “Close to JB Weld
brand” and that “should follow closely to going head to head with JB Weld and
play with such elements as black and red color palate and bringing in some
visual reference to steel reinforcement.” The graphic designer stated: “The
objective of this project was to go straight up against the top competitor (JB
Weld) and create packaging that mimics the competitor’s architecture. I was
able to pull subtle elements into our package, but still keep our package
looking tough and geared towards the Gorilla brand.”

The ultimate trade dress had: (1) a 5-inch-wide orange
blister package card; (2) an image of a gorilla at the top next to the phrase
“Gorilla Incredibly Strong”; (3) two squeezable tubes in a clear blister
package in a V-shape; (4) a gray tube with red coloring on the bottom and a
gorilla image next to the phrase “Gorilla Incredibly Strong” towards the top;
(5) an orange tube with black coloring on the bottom and a gorilla image next
to the phrase “Gorilla Incredibly Strong” towards the top; (6) a clear plastic
blister package holding the tubes curved downward; (7) a drop-down technical
information box located between the V-shaped tubes, that has four lines of
information separated by white lines; (8) a gray banner across the bottom end
of the card; (9) a gray banner across the top of the card that includes the
name “GorillaWeld”; (10) the phrase “Steel Bond Epoxy” on the card; and (11) a
list of materials to which GorillaWeld bonds on the card’s bottom right side.

[Note: the descriptions here are really not that helpful,
though clearly carefully crafted to make J-B Weld’s case—the overwhelming fact
is that the dominant colors create a completely different visual impression.
Yes, they look like the same product, the way that different manufacturers’
light bulbs look like the same product, but they do not in any way look like
they have the same source.]

Although whether GorillaWeld is an “epoxy” was disputed, the
back of GorillaWeld’s packaging, the Safety Data Sheet (SDS), and Gorilla
Glue’s website all state that GorillaWeld uses MMA chemistry.

J-B Weld sued for trademark infringement and false
advertising of GorillaWeld as an epoxy adhesive containing steel. Gorilla Glue counterclaimed
against (1) “World’s Strongest Bond”; and (2) “Steel Reinforced Epoxy.”

Trade dress first: the overall impression wasn’t
similar. Much explanation of why
omitted, focusing on colors and their arrangement. J-B Weld urged the court to find similarity
“despite the brand names, logos, and color schemes.” But “it is precisely the
prominent display of distinctive brand names and logos coupled with the use of
radically different color schemes that negates any possibility of consumer
confusion in this case—thus precluding a finding of similarity.” This wasn’t a
swap of one name and logo for another, but a “radically different color scheme”
and a prominent use of a distinctive logo “to cure the effect of any visual
similarity.”

J-B Weld offered as actual confusion evidence three
employees’ testimony. One indicated that
a Loctite sales rep asked if J-B Weld made Gorilla Glue’s “twin tube product
for [Gorilla Glue]”; another had an encounter with a Loctite rep who asked if
J-B Weld was “private labeling for GorillaWeld” after seeing the packaging; and
another indicated a buyer from O’Reilly’s Auto Parts asked if J-B Weld had
“anything to do with” Gorilla Glue’s product. These were admissible hearsay, going
to state of mind and not the truth of the matter asserted. But “[i]nquiries indicating that ‘consumers
perceive a difference between the designations and are skeptical of the
existence of a connection between users’ may not establish the existence of actual
confusion.” Those inquiries indicate that the speakers actually know there’s a
difference. They were inquiring into whether there was a possible
collaboration, but they weren’t confused about source nor did they believe that
there was a collaboration (if they did, they wouldn’t have needed to ask). And there was no other evidence of confusion.

Intent: the Eleventh Circuit has held that: “If it can be
shown that a defendant adopted a plaintiff’s mark with the intention of
deriving benefit from the plaintiff’s business reputation, this fact alone may
be enough to justify the inference that there is confusing similarity.” But the
court here noted that “there is a difference between ‘intentional copying’ and
adopting a design ‘with the intent of deriving benefit from’ another person’s
design.” Intent to copy aspects of a
trade dress doesn’t show intent to derive benefit from J-B Weld’s reputation
through confusion. [Not unrelatedly,
Gorilla Glue didn’t adopt J-B Weld’s mark,
even if it copied aspects of that alleged mark.] And bad intent is neither necessary nor
sufficient for infringement. The instructions to Gorilla Glue’s graphic designer
didn’t reveal an infringing/confusing intent. As she herself stated, “I was
able to pull subtle elements [of J-B Weld’s Dress] into our package, but still
keep our package looking tough and geared towards the Gorilla brand.” “The
Eleventh Circuit has noted that public policy favors permitting companies to
imitate the products of competitors—so long as there is no intent to deceive
consumers as to the maker or origin of the product.” Gorilla Glue intended to compete, for sure,
but its use of its own well-known color scheme and distinct logo “clearly
indicates that Gorilla Glue did not intend to confuse consumers as to
GorillaWeld’s origin.” In addition,
certain design features may have had non-trademark functions, such as the
V-shape showing consumers that the components need to be mixed together and preventing
leakage. “Where, as here, ‘there may have been many other motivations for
Defendant’s actions,’ intentional copying does not necessarily indicate a
desire to capitalize on another’s goodwill.”

The remaining factors changed nothing. Gorilla Glue was
entitled to summary judgment on the infringement claim.

False advertising based on Gorilla Glue’s use of the phrase
“steel bond epoxy”: The court started with “epoxy.” Literal falsity requires an
unambiguous message and is a “demanding” standard. J-B Weld argued that Gorilla Glue’s use of “epoxy”
is literally false because “epoxy refers to a polymer containing one or more
epoxy groups.” Even if the court accepted that definition, the claim failed for
lack of materiality. “[T]he presence of a false statement alone is not
sufficient to prove materiality.” J-B
Weld’s consumer survey didn’t show materiality; it failed to show that consumers
even understood “epoxy” as chemically defined. J-B Weld’s expert didn’t
question survey respondents about whether an “epoxy resin has a specific type
of chemistry to it”: as he said, “Do you really think [consumers] care at the
end of the day if it takes two things, you put it together, and it stays there?
They’re happy.” J-B Weld’s speculation that Gorilla Glue’s prior attempts to
launch an MMA product under the name “Plastic Plus” and alleged repositioning
of the product as an epoxy indicated that “epoxy” was material were
insufficient.

“Steel bond”: J-B Weld argued misleadingness: a false
suggestion that GorillaWeld contains steel. Again, assuming that its survey
showed misleadingness, it still didn’t show materiality. The survey simply showed
respondents both J-B Weld Original and GorillaWeld, and then asked which of the
two products contained steel. (Amazon also originally categorized GorillaWeld
as a “metal-filled epoxy” on its website, but a retailer’s misclassification wasn’t
evidence of consumer deception.) Nor
would the court presume deception from Gorilla Glue’s deliberate conduct
because the Eleventh Circuit hasn’t adopted that rule and in any case Gorilla
Glue’s acts weren’t of an “egregious nature.”

J-B Weld pointed to some 2009 evidence from Gorilla Glue
employees noting that steel or a similar filler would provide “some advantage
from a marketing standpoint.” But that old evidence wasn’t a direct indication
of what actual consumers would do today.
[Quite notable how differently courts treat evidence of advertisers’
beliefs about consumers in advertising cases versus trademark cases. There’s no
question that a substantial number of courts would treat similar evidence about
the appeal of, e.g., a particular trade dress as indicating both an intent to
deceive and likely success in doing so. I think those courts are wrong, but it’s
also a bit bizarre to presume that active competitors in a market have no idea
what might appeal to consumers.] Summary judgment for Gorilla Glue.

Counterclaims: J-B Weld argued that claims based on “steel
reinforced” should be dismissed based on laches. In Georgia, the borrowed period
for assessing laches is four years. Gorilla Glue argued that its awareness of
the phrase did not ripen into a provable claim until the present litigation. The
court agreed. In 2009, its reverse engineering attempts led to a report from an
adhesive manufacturer in 2009 stating that the “one puzzle in this adhesive is
presence of iron powder in the formula. It does not appear to add anything to
the properties … one clue to iron powder … can be found in the name ‘steel’
that labels the resin.” But that didn’t mean that Gorilla Glue knew or should
have known that it had a provable claim for false advertising simply based on a
comment from a third-party that the presence of iron in JB Weld Original was
“puzzling.” Even if it did delay, Gorilla Glue raised a material issue on
excusability, because the parties weren’t directly competing on two-tube epoxy
products until GorillaWeld was launched.
The Eleventh Circuit has held that delay is reasonable where a plaintiff
waits to sue until coming into direct competition.

And there was a genuine dispute of material fact on literal
falsity. Gorilla Glue had evidence that
the “presence of ‘steel’ in JB Weld glues has no significant overall
reinforcing effect,” while J-B Weld’s expert said that the addition of steel
“strengthens the adhesive, supports the adhesive, allows those steel-containing
products to provide enhanced performance and improved mechanical behavior.” [Is that the question? I thought it was whether there was “steel” in
there in the first place.] Anyway, could be literally false or misleading, although
the court had “reservations” about ultimate materiality and would allow
additional briefing on the matter. J-B Weld’s motion for summary judgment
denied.

Claims based on “World’s Strongest Bond” were not only
barred by laches, but also failed as a matter of law. Gorilla Glue learned of
J-B Weld’s use of the phrase in 2011, discussed challenging it in 2013, and did
nothing. Unlike with “steel reinforced,” J-B Weld was using the phrase on all its
products, including those in direct competition with Gorilla Glue. Gorilla Glue argued that it didn’t sue because
a comparison between its Gorilla Epoxy product and J-B Weld’s competing
ClearWeld showed comparable bond strengths. “Yet if the two products had
comparable bond strengths, … it stands to reason that J-B Weld’s product could
not literally create the ‘World’s Strongest Bond,’ because another product
could create that very same bond.”

J-B Weld also showed prejudice because it “extensively used
the J-B Weld Packaging Statements and built its brand around them,” spending
millions on advertising and marketing its products between 2012-2017—all of
which prominently feature the phrase “World’s Strongest Bond.” Gorilla Glue
argued that there was no prejudice because J-B Weld knew from the beginning
that it was a wrongdoer and had no testing supporting its claims. But J-B Weld
argued that the phrase was puffery; it could hardly be said that J-B Weld would
have spent the same amount of money and effort “notwithstanding any threat of
litigation from Gorilla Glue.” [Really?
They might have changed that one phrase; the advertising/marketing is
for the products as a whole, not a single phrase.]

Gorilla Glue argued that laches didn’t apply because it
sought only injunctive relief, but that’s only the rule where the public interest
in preventing consumer deception outweighs the effect of a plaintiff’s delay in
bringing suit. Without strong evidence of likely or actual confusion, laches
applied.

Anyway, “World’s Strongest Bond” was nonactionable puffery—exaggerated
and general. Gorilla Glue argued that the strength of an adhesive bond is
measurable and thus falsifiable. But it was too general, not a “detailed
factual claim.”

from Blogger https://ift.tt/2O6vAuu

Omission of side effects in lash “cosmetic” ads was plausibly false & misleading

Posted on March 4, 2019 by rtushnet

Lewis v. Rodan & Fields, LLC, 2019 WL 978768, No.
18-cv-02248-PJH (N.D. Cal. Feb. 28, 2019)

Nine plaintiffs brought a putative class action alleging
that defendant Rodan failed to disclose that its Enhancement Lash Boost eye
serum, advertised as a cosmetic designed to make eye lashes longer and more
beautiful, “had harmful side effects linked to an ingredient in” the product, a
synthetic prostaglandin analog. One plaintiff’s eyes changed color, another
“developed a grey spot in her vision and had central serious retinopathy,” another’s
eye lashes fell out and not all of them have grown back, and another “developed
a rash on her eyelid[,] [ ] her eyelid became discolored and darkened, … and
lashes no longer grow where [a] bump” developed. “Many of these side effects
match those associated with all prostaglandin analogs.”

Indeed, for these reasons, the FDA previously warned another
manufacturer of “cosmetic” lash-enhancement products that used the ingredient
that the products violated the FDCA because they were unapproved new and
misbranded drugs and failed to reveal important side effects. Rodan, too,
didn’t disclose the serious side effects associated with the ingredient. The warning states, as relevant here, “For
external use only. Avoid getting in the eye; in the event of direct contact
rinse with cold water. If you develop irritation or swelling discontinue
product usage.” Rodan’s website and marketing materials did no better and, in
some instances, affirmatively distinguished “drugs” that cause those side
effects from Lash Boost’s side effects, e.g.: “The only serious side effects we
have heard about are those associated with drug products, not cosmetics.” Rodan
claimed the product was “a cosmetic.”

Plaintiffs alleged that, had Rodan included an adequate and
“full[ ] disclos[ure about the] adverse side effects of Lash Boost, plaintiffs
would have decided not to purchase Lash Boost.” And plaintiffs alleged that “they
did not receive what they paid for when purchasing Lash Boost,” because they
paid for a product with, at most, side effects limited to irritation but
instead received a product that had serious and sometimes permanent side
effects. They asserted claims under
various states’ common laws and false advertising laws, as well as a RICO claim
that was dismissed because it was a RICO claim.

The false advertising claims were all based on an omission
theory. For omission, Rule 9(b) “requires that the complaint adequately allege
why the omitted fact is true, as well as being material to consumer
decision-making.” The complaint did
so. Plaintiffs plausibly alleged that
the product could cause serious side effects, and that this is material in that
reasonable consumers would have been likely to act differently because of this
fact. Given the materiality of the
omission and the other allegations, it was plausible that, had Rodan reasonably
and adequately disclosed the serious side effects, the plaintiffs would have
been aware of the disclosure and acted differently. And plaintiffs adequately
alleged injury: they didn’t receive the benefit of the bargain, a
minimal-side-effect product. Finally, as
to the states that require some kind of intent, plaintiffs adequately alleged “that
defendants were aware of or had [ ] reason to know of” the allegedly omitted
information.

from Blogger https://ift.tt/2NIEq1i

Posted in Uncategorized | Tagged consumer protection, false advertising, Omission of side effects in lash "cosmetic" ads was plausibly false & misleading california | Leave a comment

Belmora doesn’t replace 43(a)’s usual requirement of protectable subject matter in standard trade dress case

Posted on February 28, 2019 by rtushnet

Secret of the Islands, Inc. v. Hymans Seafood Company, Inc.,
2019 WL 917209, No. 2:17-cv-342-BHH (D.S.C. Feb. 25, 2019)

SOTI sells salt scrubs and other body products; salt scrubs can
be used as hand soap, but also exfoliate and moisturize. Hymans sells salt
scrub and skin care products as Holy City Skin Care. Hymans initially displayed
SOTI restroom samples in its Charleston, South Carolina restaurant and sold
SOTI salt scrubs in its attached gift shop. After two years generating $100,000
in retail revenue from SOTI products, Hymans allegedly started relabeling SOTI
products in its gift shop, and SOTI terminated the relationship. After that, Hymans allegedly “misappropriated
SOTI’s brand-value and goodwill by employing restroom-sample displays materially
indistinguishable from SOTI displays, employing the same distinctive slogans
that SOTI created to market its products to the hospitality industry, and
duplicating SOTI’s distinctive packaging.” In July 2012, a SOTI manager
informed Hyman’s marketing partner that Holy City’s products infringed on its
rights.

The court first found that SOTI pled itself out of court on
laches/statute of limitations issues.

The complaint clearly indicated that SOTI knew of the
alleged infringement in 2011, when it terminated its relationship with Hymans,
and then again in 2012. But SOTI didn’t file suit until 2017, well outside the
three-year limitations period applicable to the South Carolina claims. The
statutory unfair trade practices claim failed also because it didn’t allege
sufficient facts to show an adverse impact on the public interest, as required.
Claims of public confusion and deception weren’t enough to transform an
“essentially private” business dispute into a matter that could cause
“substantial injury to consumers.” There was no allegation that the Holy City
scrubs were dangerous, or financially more costly to consumers. Diversion of consumers and revenue, along
with misappropriation of goodwill, were mere private wrongs.

The Lanham Act borrows state law statutes of limitation for
measuring laches in the first instance.
Laches requires unreasonable delay by the plaintiff plus harm to the
defendant from the delay. Here, there
was a nearly five-year delay between the latest point at which SOTI knew of
Holy City’s allegedly infringing products and marketing practices, and the
filing of suit. Not only was there a
presumption of unreasonable delay given the timing, but the complaint also pled
events that established unjustifiable delay, given the prior relationship
between the parties. Even if it was
reasonable not to sue when the relationship terminated, “it must have set off
alarm bells for SOTI when, in July 2012, it subsequently discovered that a
major corporate entity in the hospitality industry, U.S. Foods, was
distributing Hymans’ now fully-branded competing product line—Holy City Skin
Care—in notably similar mason-jar packaging.” Instead, SOTI waited to sue until
“Holy City’s product line and business were successfully established. The Court
finds that this delay was unreasonable, and would have the perverse effect of
dramatically multiplying the damages to which SOTI might be entitled if the
Lanham Act claims were permitted to proceed, damages which could have been
easily mitigated if SOTI brought its claims when it knew they were ripe.”

Prejudice to defendants was thus also shown by the
allegations of the complaint: “Plaintiff … avers that by building a business
model dependent upon SOTI’s trade dress and trademarks, Defendants have
successfully supplanted hundreds of sales accounts, and diverted millions of
dollars in revenue. Thus, the amended complaint demonstrates that Defendants,
relying upon SOTI’s inaction, built a valuable business over the course of
approximately six years, a venture that axiomatically required the commitment
of substantial economic resources.” This was the rare case where no factfinding
beyond that alleged in the complaint was required for laches.

Regardless, the Lanham Act claims were also substantively deficient. “The closest that SOTI comes to stating a plausible claim to relief
based upon misappropriation of its intellectual property is its reverse passing
off theory, where it alleges that Hymans, in 2011, took some amount of SOTI’s
sample salt scrub product and placed it in jars with a different label for sale
in the Hymans General Store.” But that claim was time-barred.

Instead, the core of SOTI’s
Lanham Act claims was that its marketing system of “providing salt scrub
samples in hospitality business restrooms with signs encouraging users to
purchase retail product in attached shops, its marketing slogans such as ‘Turn
your restroom into a profit center,’ and its packaging the product in mason
jars with an attached wooden spoon affixed by an elastic tie around the neck of
the jar” was protectable. But beyond
conclusory allegations, SOTI failed to allege that its marketing system,
marketing slogans, and mason-jar packaging “were anything other than generic,
functional sales modalities.”

SOTI argued that Belmora
didn’t require it to possess a protectable mark to proceed under §43(a), but Belmora “does not support the kind of
open-ended unfair competition claims that SOTI suggests are permissible under
Section 43(a).” Instead, Belmora
allowed the owner of a foreign mark to proceed against a US user; it didn’t “open
the door to ‘boundless application [of Section 43(a) ] as a remedy for unfair
trade practices.’”

SOTI’s false advertising claim was premised on the fact that
Holy City labels its salt scrub jars with a gross weight of 780 grams, whereas
SOTI labels its packing using the net weight of the salt scrub itself, which is
455 grams. SOTI alleged that South Carolina’s Uniform Weights and Measures Law and
implementing regulations to the federal Food, Drug, and Cosmetic Act require
that consumer packing must state the net quantity of contents within a
packaging container. But there was no showing that 780 grams was not the gross
weight, and there was no private right of action to enforce the laws SOTI
cited.

from Blogger https://ift.tt/2tF3Tzx

Posted in Uncategorized | Tagged Belmora doesn't replace 43(a)'s usual requirement of protectable subject matter in standard trade dress case false advertising, trademark | Leave a comment

ICANN and the New Top-Level Domains, panel 2

Posted on February 26, 2019 by rtushnet

ICANN and the New Top-Level Domains

“Walled Gardens:” Should gTLDs Become Private Platforms?

Becky Burr, ICANN Board & Neustar: We used to talk about
.kids as a walled garden/moderated content for kids, a safe space. [Based on
what we know about who abuses kids, it’s not surprising that it hasn’t worked
all that well.] Others are more general—open to anyone to register, but with
rules for registrants. Whether that’s good or bad has to be more granular.

Sarah Deutsch, ICANN Board: To me, depends on what the garden
is: if it doesn’t allow other people in and there’s bad activity, such as
anticompetitive activity, that wouldn’t be good. Other spaces might just be
regulated. That could be a walled garden.
Worries: where someone gets exclusive rights to run a generic term as
one competitor in the market. [Example
from Aufderheide: L’Oreal owning .beauty]

Kathy Kleiman, Center for Information Technology, Princeton
University: When we started, gTLDs were to be managed in the public interest. We
looked for abuse of the structure (malware, botnets) and not bad content (where
laws differ worldwide and where it wasn’t our job to judge content). Country
rejected SOPA/PIPA domain name blocking. We protected due process. But on the way
to new gTLDs, ICANN decided to open up registry agreements to voluntary
commitments. Registry applicants slid a
lot of other stuff into those agreements, which ICANN allowed. Donuts, w/hundreds of TLDs, created a policy
allowing it to block registrations based on agreements w/TM owners, a policy
that had been rejected at the ICANN level.
Judge © and TM claims. Minds +
Machines promises to “constructively work with law enforcement to address reported
cases of abuse”—law enforcement asks for a lot of things based on mere
allegations, and no due process is mentioned.
Yet ICANN has a limited mandate.

Nonetheless, ICANN apparently embraced voluntary content
regulation in its new bylaws. Question
for the panel: what’s left of the multistakeholder model and what is ICANN’s
continuing role in protecting the open structure of the internet?

Jeff Neuman, Com Laude/Valideus: The ultimate end users are also
of concern. (1) registries w/restrictions—only allow certain entities in. (2)
closed generics: taking a generic word and using it w/in own organization/its
affiliates. (3) ability to take down names that violate your policies. All 3 of these can be and are good things.
Organizations are facing increasing scrutiny for the content delivered through
their platforms—FB, domain name registry [those actually have very different
levels of control of individual posts/incentives to take users’ interests into
account/levels of public exposure]. Have
to think of it from their perspective. Wish it was as easy as saying we don’t
regulate content. But there’s child porn, imminent threats of harm,
counterfeiting and infringing movies. When things are that obvious, and you
know that you will be criticized for allowing that activity through your
platform, then the choice isn’t as simple as not looking at content. He’s seen actual threats of car bombings [so
he would have gotten rid of FB if he ran .com?] and he’d rather take that down
than not do so.

Mitch Stoltz, EFF: Not super concerned with defining “walled
garden.” As a concept it’s useful to
talk about gatekeepers of speech on the internet, of which there are many
varieties. FB has an incredible amount of power over who gets to speak and who
will hear it. It’s scary that registries can make your entire website
disappear; registry is a private company w/its own business motivations
exercising arbitrary power over speech.
It is important that different registries can do it differently–.com
historically doesn’t have policies like those the new ones are adopting. Cloudflare has immense power. Content industry wants to leverage that to
make it police for copyright if it polices for child porn/foreign pharmacies.
But it’s easy to see where it goes from there.
If we built a smart highway would we want it to scan to make sure you
were driving in an ok way? [Note that China’s social credit system is doing exactly
that—you won’t be able to travel easily unless you have good social credit.]

Burr: There are enormous numbers of issues here. ICANN shouldn’t have taken on the burden of
trying to figure out compliance w/all these public interest commitments, which
applicants made in order to get a competitive advantage. Now that people used those to get a competitive
advantage, holding them unenforceable is unattractive too. We grandfathered the
existing commitments into the new bylaws in order to kick the can down the road
(outsourcing compliance which isn’t great either).

If there are lots of different registries then it might not
be a big problem if they have varying rules.

Example of problem: something weird is going on in a set of
domains. Sites selling every dog breed in the world to US consumers, shipping
the puppies sight unseen. She didn’t buy a dog so she can’t say she was ripped
off, but there was no question that this was a scam & we took them down. It’s
critical that people be able to exercise that kind of judgment where consumer
harms are involved. There’s no state action & scammers can go register on
some other site; we aren’t going to protect it. She doesn’t think that’s a
matter of free speech.

Deutsch: all sorts of third party sites have provisions
about copyright & TM etc; DMCA gives you obligations for takedown. It’s not that surprising for stuff to be in a
contract. There is also more pressure on platforms, ISPs, even possibly
registrars & registries to take more fiduciary responsibilities for what’s
on their sites. But shadow regulation also tends to grow.

Neuman: .biz had a problem with being a known source of
spam; made it hard for legit clients to get emails through. Created an anti-abuse policy for acting on
malware, phishing, spam: first policy in the big registries. Registrars would
have 12 hours to take the domain down or we’d do it. That made us pariahs in
the community. But there wasn’t a slippery slope. In 2005, took down 32,000
domains with zero complaints from the registrants. We’ve taken down 100,000s of domains. Took
.biz from most abusive to one of the safest TLDs. Not all slippery slopes have to
be slid down.

Stoltz: There are issues of norms & due process. Should
the power company combat fraud by accepting complaints about fraud & cut
off a business’s electricity if it finds the complaint persuasive? That sounds
bizarre. Power is used in the commission of all sorts of crimes, but it’s still
not their role to police that. There is
space for competition among policies, but that only works if that’s evident to
consumers—if people know what they’re buying, and if there is actual competition. Donuts owns 1 of 6 of the new domains, and
they use policy rejected by ICANN.

Kleiman: the key move here is ICANN’s abandonment of the
principle that it was about regulating infrastructure and move into regulating
content. The attempt here is not just to
get the regulations into the agreements but to get to ICANN to enforce content
regulations.

Q: if an applicant made promises & got the domain b/c of
the promises, it’s sensible for ICANN to be one of the entities that can hold
them to its promise. If Goodyear got
.tires and biased the search within that gTLDs—that would be ok if it’s
disclosed, but maybe ICANN should stop it from happening. FTC said it would
watch and if there’s deceptive trade practices the US would enforce against it.
That’s how it happened at ICANN; there’s no blanket prohibition on a competitor
running a closed generic and we deal with problems after the fact; would it be
better to have it otherwise?

Karanicolas: Infrastructure layer makes a difference: there’s
a difference b/t Twitter kicking Milo off and Comcast installing a filter to
prevent anyone on their network from seeing Milo’s content. But where do
registries fall on that spectrum? [Right,
one of the distinctions to be made here is the puppy mill v. Craigslist, both
of which might be offering scams. Do we
have any way of distinguishing that in order to protect Craigslist against a
significant chilling effect?]

Q: We need to know what the rules are about people being
kicked off. If there are pre-agreements w/certain TM owners, is the
decisionmaker biased? Is there any way
to appeal? If you’re making government-like decisions then there should be
government-like protective structures for decisions.

Deutsch: .bible—applicant wanted to exclude nonChristians
and anybody of which it didn’t approve, and to make UDRP panelists sign a
statement of faith. This is a real world example of what happens in a closed
generic space and we need to be mindful of that.

Neuman: .disaster could easily be exploited for fraud. But we need to look out for end users more
than a right to register any name that you want. A closed generic could be a
lot more useful for end users than what’s in the market now. If Amazon had
.book and could give an official page to any author, is that better or worse
than letting anyone register and maybe never getting .book into widespread use.

Burr: doesn’t like the power company metaphor b/c power is a
utility; I don’t have a choice about where I get power from. Registries/registrars
doesn’t fit into that paradigm. DMCA
issues: registries/registrars can’t simply take down one piece of allegedly
infringing content: that makes them fit poorly into the DMCA framework. Issues w/r/t copyright & TM are not the
same as issues about what’s not legal in Thailand (criticizing the royal
family) or consumer fraud. [Which is
very telling insofar as Donuts and similar agreements are about copyright and
TM, and side agreements w/© and TM owners are what have brought us here. Donuts didn’t make any agreements with the
gov’t of Thailand, or with the FTC or even the BBB.]

from Blogger https://ift.tt/2IxDq1c

Posted in Uncategorized | Tagged copyright, ICANN and the New Top-Level Domains, panel 2 conferences, trademark | Leave a comment

ICANN and the New Top-Level Domains part 1

Posted on February 25, 2019 by rtushnet

American University Washington College of Law

YouTube
link

Welcome, Christine Haight Farley, American University
Washington College of Law

We are in the midst of an historic expansion of internet
domain names with more than 1200 new generic top-level domains (“gTLDs”) now
competing with <.com>. This 5000% increase in gTLDs is the biggest change
to the internet’s domain naming system in thirty years (and more are coming
soon!). Accompanying these new gTLDs, are new and innovative–but little
known–IP rights protection mechanisms. These developments could have a
profound impact on the rights of IP owners, domain name registrants, and the
public, and on the architecture of the internet.

Trademark Protections in the New gTLDs

Mary Wong, ICANN

What is the TMCH? Ten years of history. One of the
mechanisms developed at ICANN (not by ICANN necessarily) for the latest round
of expansion of gTLDs. ICANN has staff
but can be described as a world wide community, always talking about
participants—gov’ts, businesses, lawyers, civil society participants,
academics, other individuals. Participants
develop policy, not ICANN staff (which would be easier and faster!). These mechanisms were developed/agreed upon
by a wide variety of interests.

TMCH is not a rights protection mechanism (RPM) as such, but
supports other RPMs in that TM owners submit marks into the TMCH. Those marks
(their existence in a national register somewhere, or b/c they’re protected by
treaty) would be validated by Deloitte, the operator of the TMCH, and you end
up with a global repository supporting Sunrise and Claims (of which more soon).
Over 80% of marks submitted are validated; marks from over 100 jurisdictions;
10,000s have been submitted since establishment 5-6 years ago. Allows three categories of marks: (1) protected
by a national registration, w/no distinction across countries; (2) marks
validated by a court of law; marks protected by a statute or treaty.

Brian King, MarkMonitor: What did TM owners want from the
TMCH? Assume you’re in-house at Delta Airlines.
Responsible for brand protection: monitoring for infringement, TM
registries around the world. You may have 100s or 1000s of domain names,
including country code domain names (delta.co.uk), and have to monitor for similar/infringing
names. Might hire MarkMonitor to help w/that. Privacy protections can help
people hide identity of registration.
ICANN decides to multiply existing gTLDs by a lot. Now I have to worry about .air, .flight, .sucks,
.porn: concepts Delta doesn’t want the brand associated w/ and may have to
register domain names in merely to preempt other people from doing so. This is scary and new RPMs help address the
fear. Sunrise allows Delta to register
before non-TM owners. TMCH also provides
a notification when someone else registers [an identical domain name] in a new gTLD.

Michael Karanicolas, University of Toronto (delayed due to
weather; I was able to rely on his notes)

Kathy Kleiman introduced Sunrise in his absence. Mandatory
30-day period (some registries expand it to 60 days) that all new top level
domains are required to provide to owners of trademarks, whereby any owner of a
mark which is recorded in the TMCH may pay a special fee to the registry to
register a domain matching their mark before they are opened for general sale.

Why 30 days? And not forever? A balance designed to give
weight to the interests of current trademark owners, but also to the interests
of other potential registrants — the innovators, speakers, entrepreneurs who want
to register in a new gTLD as well. There
are multiple entities with rights in Delta, for example: dental, educational, fraternity,
etc. Both Sunrise’s duration and its
limits—requiring an exact match with what the trademark owner has
registered—reflect the desire to balance these competing interests.

Griffin Barnett, Winterfeldt IP Group: Q to address: Purpose
of TMCH for TM owners as it has turned out?
Keep in mind function of TMs, which is to protect consumers from
cybersquatting/fraud etc. Benefits to
brand community: operationally, TMCH generally works well. Sunrise allows defense against phishing. Two aspects to claims service: one is notice
to brand owner who’s recorded the mark if a domain name matching the
registration is registered, but also during the first 90 days a potential
registrant who tries to register a matching domain name will receive a claims
notice setting out the registered mark with some details. Just a notification:
does not block registration but makes them aware and hopefully they can make an
informed decision. They might be deterred
based on the knowledge of rights out there.

Q: where are these TMCH registrations coming from?

Barnett: Deloitte has a report: a lot of these TMCH records
come from the US, Europe. There are a
number of Asian jurisdictions represented, but majority are US/Europe. Large corporations
tend to be the ones recording multiple marks, but smaller/medium businesses can
use it too.

Wong: Yes; likely to see other countries’ submissions go up
b/c of increasing awareness of these kinds of protections. Most big companies
don’t enter all their brands.

King: it’s not cost effective to enter all TMs, but the
crown jewels deserve TMCH entry. There was an educational period, but people
are aware of this now.

Rebecca Tushnet, Harvard Law School

New legal or quasi legal systems are often set up assuming
that they will be used against, and
not by, bad actors, because that’s
the problem that is most salient before the new system exists. Unfortunately, online is like offline: bad
actors find ways to game the system whether that’s by disseminating fake news
or by claiming rights far beyond what they actually have, and we need to take
that into account in designing our systems. Devil is also in the details: when a policy is
written, it may change substantially depending on who is implementing it and
with what expectations.

After Sunrise, there is a Trademark Claims period: when we
talk about Trademark Claims notice, we mean the notice that goes to anyone who
attempts to register a term that is an exact match to a mark in the TMCH. If
they proceed to a final domain name registration then the TMCH mark owner also
gets a notice so that it can take further action if it chooses to do so. This second notice is called a NORN in case there
weren’t enough acronyms around.

Top ten queried entries in the TMCH that trigger a Trademark
Notice when there’s a registration attempt: Cloud (which is a registered Swiss
trademark for pens, registered by a trademark lawyer), smart, love, luxury,
nyc, forex, hotel, one, london, abc. Other terms that reporting has identified
as being in the TMCH: Cash, “RICH,” “CREDIT, the word “the,”
VIP, global, hotels, mobile, prime, uk, virtual, “Christmas”, “Insurance”,
“Dating,” “Discount,” Direct, social, construction, BUILD, BET, VACATION, and
WEDDING. Not exactly Microsoft or Xerox, though those are likely also in the
TMCH.

The Trademark Clearinghouse adopts a “lowest bar” approach
to registration, whereby if a mark can be protected under any jurisdiction, and
in any context, it is eligible for inclusion globally under a universal
standard of protection. Doesn’t matter if the mark isn’t eligible for
protection in certain jurisdictions – geographic indicators. Huge dispute elsewhere in ICANN and in TM law
over whether geographic indications are “marks,” but resolved in implementation,
outside of agreement by participants, in favor of including GIs. Also doesn’t matter if a word is considered
generic for the relevant goods or services, or for some other goods or
services, in certain jurisdictions. If it’s protected anywhere it’s protected
everywhere for everything.

Moreover, although not by design, the TMCH accepts design marks
– Once a mark has been accepted, it is protected as a word mark without
reference to any distinguishing design features, since domain names are a
text-only medium. So even when the trademark itself does not protect the textual
elements as text, the TMCH will extract the textual elements for protection, as
Deloitte confirmed when we asked them what they’d do with various design and
stylized marks. Cars by Disney, Music by
Parallel Music Entertainment – all eligible as word marks for trademark
protection.

For example: the American Physical Society publishes a
journal called, naturally enough, “Physics”. Their US registration for the logo
used on that publication was required to disclaim the word “Physics”. However,
since the EU does not have an explicit disclaimer procedure, even though the
mark is just as limited in the EU—it is a figurative mark—they used their
figurative EU registration to get their TMCH entry for the word “Physics” per
se.

Q: Implementation v. policy development.

Wong: One of the most fascinating aspects of the job is
looking how a policy is written, which goes through multistakeholder
consultation process, a little bit like treaty negotiations. The output might be fairly general and that’s
what happened. The original policy recommendations were really really general.
Protect legal rights of others. Everything we’re talking about today was
developed as minimum protections during implementation of the very general
policy recommendation. Had to be turned into something workable. Now there’s a policy working group looking at
all of this and will hopefully develop policy recommendations that can be
implemented. Where things go too far
they can be restrained, where not far enough extended.

Q: was this issue discussed in policy development then?

Wong: there was a very general policy; with respect to specific
points about GIs and text, the group didn’t list all the different things that
could or couldn’t be protected, but during implementation it was made very
explicit that no jurisdiction should be excluded. It was felt that neither ICANN nor Deloitte as
the validator are in a position to determine which marks are protected
trademarks. So the rules were made
fairly generic.

RT: Where there are gray areas, to say “we’re not making a
judgment” actually admits of two approaches: you can say it’s in or it’s
out. Both of those approaches are “nonjudgmental”—and
you really could have thought that “all TMs are in” clearly excluded GIs
because there’s an entire class of people who think that GIs aren’t trademarks
and have a different ordinal priority when it comes to rights conflicts. And
where you have already a really expansive definition of rights (protected
anywhere, protected everywhere; registered for any good, covered for all
goods/services) then further expansive decisions have collateral consequences.

Barnett: Different jurisdictions approach things from different
standpoints, especially with respect to GIs; the clearinghouse was intended as
a baseline so no one would be left behind by the way they were accepting
certain rights. Concrete hypothetical: if Italy protects parmesan as a GI in national
law, based on the underlying concept: we did discuss that rights based in
national law should be protected. [“Rights”
is not “trademark rights” which was the actual language. Rights of publicity is
another easy example where rights but not TM rights exist.] There are a lot of nonspecific new gTLDs but
you still might encounter consumer protection issues e.g., for .web, .online, .llc.
Can’t tie Delta Airlines brand to exclude those general websites. On design marks: where there’s a textual
element w/stylization—going back to policy rationale, this took expansive
approach b/c the underlying approach is to protect consumers against
confusingly similar uses in the DNS. If
someone registered cars.whatever and uses it in bad faith to target Disney’s
Cars, it’s appropriate to have that in the TMCH for Sunrise and Claims. It’s up
to the registrant to make an informed decision.

Kleiman: Rules in consensus policy adopted by ICANN. Unanimously adopted by the GNSO council and
Board. Two things it says: the name of the RPM should be the TMCH (not “IP
clearinghouse as proposed”) to signify that only TMs should be included. Should
be required to include “text marks.” As
a parenthesis: design marks provide protection for letters and words only in
the context of the design or logo and the group was under a mandate not to
expand trademark rights. We also debated
the secrecy of the TMCH and decided against it: the notice function is
important. What happened? This goes way
beyond implementation: they accepted GIs, which WIPO says aren’t TMs; they
accepted design marks; they decided to make the TMCH secret. Stopped making
sense of “implementation.”

Wong: The implementation that Wong described started in
2008, and the policy was “protect the legal rights of others.” Kleiman’s team
(quoted) happened during implementation. Took 4-5 years and there were multiple
discussions. Coming back to consensus-building, at each step, all proposals
were put out for feedback from stakeholders. Where we ended up was the Applicant
Guidebook, reflecting a consolidated history of 5 years of implementation
stemming from v broad policy. Not saying we got it right or wrong but that’s
where we ended.

King: We protect TMs from any jurisdiction and not just from
the US. No one country’s law rules the
internet, so it’s important for this community to allow TM registrations under
the laws of all jurisdictions.

Barnett: Expanding rights: legit brand owners don’t want people having rights
they don’t have. Consumer confusion and bad faith are the ultimate tests. If cars.whatever
is confusingly similar and used and bad faith it shouldn’t be ok. Depending on
the use, Disney might win a UDRP case.

RT: Letting everything in is a great way to get much, much
further from the consumer protection basis of trademark. And every axis on
which you decide to allow more in has effects on how much further away from protecting
consumers you are. We’re already abstracting away from national rights and from
goods and services. Cars is a great
example b/c the TMCH/Sunrise have already abstracted away from consumer
confusion and bad faith. How many cars.whatever
websites would you have to look at before you found one that was infringing? (URS/UDRP
is the right solution for the bad faith ones.)
And they’re all impacted by Sunrise and Claims.

Q: Global database of recordals would be really useful. So
let’s talk secrecy.

King: Putting people on notice: TMCH does put potential new
domain name owners on notice that they may infringe if they proceed. That notice can then be evidence in a
potential case. 90% bounce rate of those
who put a domain in their cart, received a notice, then didn’t proceed to final
registration. That’s great for
deterrence! TM is a clear way to cry foul but it’s impractical to go to federal
court in every infringement case.

RT: Abandonment rates are really unclear: 90% rate might be mainly
for queries by automated programs, but we really don’t know. Also, that 90% rate
refers to queries for cloud, hotel, one, london, abc—the idea that they were
planning on being cybersquatters is simply not plausible. We were unable to
survey actual cybersquatters and it’s hard for me to believe that someone who
wants a phishing site is going to be deterred by a notice, but better data
might convince me otherwise; the point is that making claims about deterrence
is simply not supported by the current data.
We were able to survey ordinary registrants and their understanding of
the notice was not much better than chance.

Barnett: there are issues around gaming the register. We
want a clean TMCH as much as we want a clean PTO register. There is a mechanism that allows challenges
to records in the clearinghouse—you can challenge a denial if you’re a TM
owner, or a third party can challenge [if they learn of it]. If you’re
attempting to register in good faith, then there is a mechanism to get that
deactivated.

Confidentiality: there’s conflicting views on transparency.
The idea of keeping it confidential is important b/c TM owners make strategic
decisions. That reflects information from the brand owner; don’t want to
provide road map to what gaps exist.

Farley: people monitor other countries’ registration systems
for information already.

[How would this “abuse of transparency” work exactly? The abuse scenario is that they look at the
TMCH, look at the PTO’s registration database, and register the domain names
that aren’t in both? There is zero evidence that this is how anyone does
anything, and zero explanation how knowing from the TMCH that Coca Cola might possibly
consider Coke a more valuable mark than Vault would change anyone’s behavior. Moreover, trying to register the domain names
would itself reveal the relevant information to the putative abuser.]

Phil Corwin, working group co-chair at ICANN. (1) Confidential, but if I want to know whether
ACME has registered, I can simply go during the first 90 days and find out whether
it’s recorded. We suspect that some
parties have done that. [So the point of the secrecy is ….?] We don’t know how many queries were like
this. (2) Domain names can be trademarked, as in booking.com. (3) We are talking here about new gTLD policy
not consensus policy: to get that through ICANN you need consensus—unanimous or
near-unanimous; 80% support isn’t going to be good enough. [Clarification:
consensus policy would mean that Sunrise, Claims, URS would all be applied to
every gTLD, including the legacy TLDs like .com.]

RT: [Great points; (3) means that policy decisions made in
implementation are likely to stick.]

Q: a brand is not a TM.
Delta.sex isn’t a problem if it isn’t confusing. You don’t have rights against it even if you
don’t like it.

Q from PTO person: GIs are not marks protected by
statute/treaty. Way back when, we were thinking about Olympics, Red Cross etc. for
those categories. Has been handed her
head for conflating GIs with marks—offends Europeans/WIPO. However, GIs can be
protected as trademarks whether unregistered or registered if they are serving
as TMs. This is a burning issue for the PTO. We are very concerned that when it
says “marks,” “TMCH,” we want to be talking about marks and not GIs. Is this still on the table with the working
group at ICANN?

Q: how does GDPR affect rights protection mechanisms as we revamp
WHOIS in a different PDP (policy development group)?

Michael Karanicolas: Manual input has determined some of the
content of the TMCH, e.g., inclusion of Christmas and the like. Not a great
oversight mechanism, though. Hard to get a broad understanding of how the
system is working, which is vital to accountability and overall assessment: is
abuse widespread or limited? Corwin
mentioned that if there isn’t consensus the status quo prevails, which is particularly
problematic where the implementation drifted from the rules as originally
designed, specifically for accepting design marks [and GIs]. Turns the original rules on their head. Allows beneficiaries to block “consensus” even
though that was the original consensus. Leads to larger Qs about ICANN and its
role in internet governance. Transparency is fundamental to what ICANN does. And it’s fundamentally problematic when used
to maximize rights beyond what TM would grant (global rights, goods/services).

Wong: Yes, it is under consideration that GIs would be treated
differently. Part of our discussions. “Mark
protected by statute or treaty” but the word mark isn’t defined. May not be about whether something is a GI
but whether it is a mark. [But if GIs
are not trademarks … I believe my logic class had something to say about this.]

Barnett: the wording of the claims notice can be improved,
we all agree.

King: “cloud” example—be careful b/c one of the most
valuable TMs in the world is Apple.
[Though note it’s not at the top of the list of those queried in the
TMCH the way “cloud” is, so a bit of a red herring, no pun intended.] GDPR makes things a lot harder.

from Blogger https://ift.tt/2NuYEMe

Posted in Uncategorized | Tagged ICANN and the New Top-Level Domains part 1 conferences, presentations, trademark | Leave a comment

Misleading advertising that’s already leaving the market can’t support finding irreparable harm

Posted on February 25, 2019 by rtushnet

Danone, US, LLC v. Chobani, LLC, No. 18 Civ. 11702 (CM), 2019
WL 760040 (S.D.N.Y. Jan. 23, 2019)

Danone sued Chobani for advertising that Chobani’s kids’
drinkable yogurt product, Gimmies Milkshakes, contains “33% less sugar than the
leading kids’ drinkable yogurt” – which all agreed was a reference to Dannon’s
Danimals Smoothies. Although Danone was
likely to succeed on the merits, it did not show irreparable harm and could not
secure a preliminary injunction.

Danimals are sold in 3.1 fluid ounce bottle serving sizes
which currently contain 9 grams of sugar per 3.1 oz. serving, although three of
them used to contain 10 grams of sugar. (Dannon reduced the sugar content in
those three flavors in mid-2018, but used up its old packaging for the new
products; this packaging was available at least through January 2019 in NYC.)
Chobani’s Gimmies are sold in packs of 6 4-ounce servings. Until early January
2019, each bottle of “Cookies & Cream Crush” and “Bizzy Buzzy Strawberry”
contained 9 grams of sugar (now 8), while the “Chillin’ Mint Chocolate” flavor
contained 7 grams of sugar. On a per ounce basis, Gimmies had less sugar (by a
gram or two) than did Danimals.

front with 33% less sugar claim

back with mouseprint

On the front, top, and back of the packaging, the Gimmies
packaging says that Gimmies contains “33% less sugar than the leading kids’
drinkable yogurt.” “On the front and top of the packaging, the claim appears in
reasonably readable typeface and is in no way qualified.” On the back, it’s
smaller and asterisked, referring the consumer to two footnotes that are found
below the nutrition facts panel, “in typeface so small that it is barely
legible.” Those footnotes read: “i. *chobani® Gimmies™ Milkshakes: avg. 8g
sugar; leading kids’ drinkable yogurt; avg. 12 g sugar, per 4 fl oz serving;
and ii. **Chobani® Gimmies™ Milkshakes: net 4 fl oz; leading kids’ drinkable
yogurt: net 3.1 fl oz.”

No matter what, 7 or 8 grams of sugar is not 33% less than 9
grams of sugar, or even than 10 grams of sugar. “Put otherwise, a single
serving of Gimmies does not have 33% less sugar than a single serving of
Danimals, no matter the flavor.” Chobani argued that its disclaimers disclosed
the necessary information, but doing so required—and the disclaimer didn’t
disclose that it required—a lot of math, including averaging all three flavors
of Chobani’s products (8.333) without knowing the grams of sugar in the flavors
other than those in the package the consumer was looking at, rounding that number to the nearest gram, then
comparing it to the sugar in Danimals, which is 9 grams per serving. Additional calculations were required to
produce a hypothetical 4-ounce serving of Danimals (11.6 grams, which Chobani
rounds up to 12 grams). (“And as the Court noted at the TRO hearing, parents do
not ordinarily give a child 1.33 servings of yogurt anything.”) At that point, the 33% less claim became “true.”

Chobani offered a simpler theory in litigation: dividing the
number of grams of sugar in a single serve bottle of each drink by the number
of ounces in the bottle yields 2 grams of sugar per ounce for Gimmies, and 3
grams of sugar per ounce of Danimals. “However, this rather more elegant and
easily comprehended ‘ounce for ounce’ comparison is not suggested by anything
on the packaging – especially not Chobani’s barely legible footnotes. And that
is not how Chobani calculated its claim.”

Dannon’s extrinsic evidence came from a marketing expert. Based on his review of academic literature,
which the court considered an appropriate source, he concluded that consumers
were not likely to attend to Chobani’s disclaimers, primarily because of their
location on the packaging. He further opined that consumers would not likely
understand that Chobani’s “33% less sugar” claim required “an averaging of
Chobani’s three flavors, a scaling up of Dannon’s serving size, and a rounding
of fractional sugar contents to the nearest whole number[.]” He also conducted
an online consumer survey and concluded that significantly more consumers would
interpret Chobani’s “33% less sugar” claim as referring to the sugar content of
a bottle of Gimmies versus a bottle of Danimals – not to an ounce of Gimmies
versus an ounce of Danimals.

The online survey used a hypothetical involving two ice
cream brands with different single-serve sizes. Consumers were asked about
their understanding of a claim of “25% less fat” by the larger cup.
Specifically, the survey asked if they were more likely to assume that the
comparison above is made with respect to fat content per cup or per ounce; and
second, if they were more likely to assume that the comparison above is made
with respect to each flavor independently or the average of all three flavors. There
was a “don’t know/not sure” option. Forty-four percent of respondents thought
that the comparison statement “25 percent less fat than [hypo]” referred to the
fat content per mini cup of ice cream, while only 28% of respondents interpreted
it as a per-ounce claim. Similarly, 45% thought that the comparison statement
referred only to the specific flavor in question, while 27% of respondents chose
an average across flavors. Only 10%
thought both that the comparison was per ounce and across flavors. Chobani’s expert criticized the study, but conducted
none of her own.

The court found that Dannon’s evidence ruled out a literal
falsity claim but proved a misleadingness claim. Dannon proved that multiple interpretations
were possible and that 10% of respondents would understand the claim completely
(which doesn’t seem like enough to me, but ok).
Still, “read in the easiest and most straightforward way, the
advertising on the Gimmies package is not accurate. A bottle of Gimmies has
either 7 or 9 grams of sugar; a bottle of Danimials has either 9 or 10 grams of
sugar. Even taking the biggest difference (7 grams to 10 grams), without any
averaging, there is not 33% less sugar in a serving of Gimmies over a serving
of Danimals.”

Chobani offered no persuasive evidence of consumer
reception, and Dannon did. Indeed, the survey was actually favorable to Chobani
by offering respondents a menu suggesting per ounce and flavor averaging
options; “had the questions been open-ended, it is plausible that some
respondents would not have contemplated that interpretation.” The questions were close-ended but not
therefore leading because they didn’t suggest a specific answer. In this case, too, no control group was
necessary to guard against bias or pre-existing beliefs because it wasn’t that
kind of study.

Ultimately, reasonable purchasers who saw “33% less sugar
than the leading brand” on the front of the box “cannot be expected to study
the back of the packaging in the detail necessary to discover the cryptic,
microscopic footnoted disclosures … never mind figure out what needs to be ‘averaged’
with what and perform the multiple calculations needed to make sense of that
claim.” The court credited “numerous studies across different types of products
reveal[ing] that consumers are unlikely to notice or pay attention to Chobani’s
hard-to-find and hard-to-read ‘disclosure’ footnotes,” and these particular
disclosures were also incomprehensible. No yogurt-specific study was required
to show likely success on these points.

Chobani made a useless argument that it couldn’t be liable
because it complied with FTC/FDA rules. Multiple problems with that: (1) Pom Wonderful says the Lanham Act is
different. (2) Read carefully, the rules did not clearly authorize, much less
require, Chobani’s choices here. “The relevant FTC Guideline about ‘Comparative
Nutrient Content Claims’ states that comparative nutrient content claims must
not contain ‘misleading implications’ in order to survive FTC scrutiny” and
should also make clear the basis for the comparison, which Chobani didn’t do.
And the FDA regs cited by Chobani, to the extent they governed relative claims,
didn’t seem to allow Chobani to average the sugar content of its various
flavors.

Ultimately, though, the court was guided by common sense:

[A] parent walking down the dairy
aisle in a grocery store, possibly with a child or two in tow, is not likely to
study with great diligence the contents of a complicated product package,
searching for and making sense of fine-print disclosures in asterisked
footnotes, and looking for flavors other the one(s) s/he wishes to buy (which
may or may not be on the shelf) in order to perform multiple mathematical
calculations – all in order to confirm the truth or falsity of a claim that is
of dubious veracity, and that could easily have been replaced with the simple
and truthful statement, “My product has less sugar per ounce than his product.”
Nor does the law expect this of the reasonable consumer. With yogurt or any
other product, plain vanilla ads and labels tend to work best.

As a result, Dannon also showed a likelihood of success
under its NY GBL § 349(a) claims. Chobani argued that Dannon’s competitive
injury was insufficiently consumer-oriented, but that’s not true. Misleading
parents about the sugar content of a product intended for their children is a
consumer harm: the public “has a strong interest in receiving accurate
information, especially when it comes to products marketed specifically for
children.”

However, Dannon failed to show irreparable harm. Though some
courts have found irreparable harm with a showing of competition + “a logical
causal connection between the alleged false advertising and the plaintiff’s own
sales position,” that couldn’t happen here. The evidence indicates that, in the
month Gimmies launched, Dannimals’ share of the yogurt product market expanded. This wasn’t an absolute increase—sales of kids’
drinkable yogurt were generally down during the holiday season—but it did
indicate lack of harm to Dannimals’ market position. “This data strongly suggests
that neither Danimals’ sales position nor its brand equity has suffered
irreparably.” Danone also didn’t offer
any evidence that it lost sales, or that the lost sales were attributable to
the 33% less sugar claim versus the fact that Gimmies has somewhat less sugar
than Dannimals and a larger serving size. It was purely conjectural to say that
Danone lost sales as a result of the
misleading claim.

Anyway, lost sales would be remediable at law. Danone argued
that its injury was to its good reputation as a purveyor of healthy and
nutritious food for children. But such
“purely conclusory testimony” proved nothing.
“[T]here is no denying the fact that Danimals has more sugar per fluid
ounce than Gimmies does.” Even if the
ads were misleading, there was no evidence or other reason to think that
Danone’s reputation would be more
injured by the misleadingness than by the truth, and Chobani had a right to
injure Danone with the truth.

[There is an interesting set of ideas about but-for
causation and appropriate baselines in here; this issue comes up a lot in
various ways in advertising law, from whether a difference between the truth
and the ad claim is merely puffery or is potentially actionable to materiality
to harm causation, as here. One can
imagine a different legal regime where the advertiser acts at its peril in
making false claims, even if it can show that the truth would also have worked.
Presumably the advertiser chose falsity over truth because it thought the
falsity would sell better, so we might rely on the advertiser’s own choices and
not its post-hoc defenses. But that’s not the legal regime we have.]

Even worse for Danone, it developed that it had been selling
Dannimals in old packaging that overstated the sugar content of its product. If
Danone was willing to tolerate that because of the expense of making new
packaging, it couldn’t turn around and accuse Chobani of irreparably harming it
on the same metric.

In addition, prompted by the lawsuit, Chobani changed
Gimmies, reducing the sugar in two flavors and changing the packaging. Flavor-specific
labels now advertise that they contain “30% less sugar*” “*than the leading
kids drinkable yogurt and disclose the comparison “Gimmies: 2g sugar per fl.
oz.; leading kids’ drinkable yogurt: 2.9g sugar per fl. oz.” This easy to
understand comparison will be on the front and the back. The remaining old stuff will be sold or reach
its expiration date by the end of March. Thus, an injunction would have nothing
to redress. Although this will take a few weeks, Danone’s own use of stale
packaging made its irreparable harm claim unpersuasive.

Relatedly, the balance of hardships favored Chobani because of
the expense/difficulty of repackaging existing product (over 2.8 million
bottles would expire before they could be relabeled) and the special harm of
pulling a new product from the shelves before it’s fully established, which
would damage relations with retailers.

Nor would a preliminary injunction help vindicate the public
interest given the ongoing reformulation and repackaging.

from Blogger https://ift.tt/2BUXIMy

Posted in Uncategorized | Tagged Misleading advertising that's already leaving the market can't support finding irreparable harm false advertising, remedies | Leave a comment

Selling copyright-infringing content isn’t trademark infringement or false advertising

Posted on February 25, 2019 by rtushnet

Joint Stock Co.“Channel One Russia Worldwide” v. Russian TV
Co., 2019 WL 804506, No. 18 Civ. 2318 (LGS) (S.D.N.Y. Feb. 21, 2019)

Channel One produces and broadcasts TV programming in the
Russian Federation and its neighboring states, and granted third parties the
“exclusive right to broadcast and re-broadcast [its] Programming and other
copyrighted materials in the United States.” Russian TV allegedly runs a
website through which it provides unauthorized access to the programming in the
United States in exchange for a subscription fee, having decrypted or acquired
unencrypted signals carrying the programming. The Russian TV website also sells
set-top boxes (STBs) that facilitate unauthorized access to the Programming. The website invites customers to “Watch
Russian TV online” and advertises over 200 television channels; the complaint
didn’t show use of Channel One’s marks on the website, but when unauthorized
users stream the programming using Russian TV’s services, Channel One’s marks
appear at the corner of the screen.

After dismissing a couple of defendants whose involvement
wasn’t sufficiently pled, the court easily found that the complaint stated
claims for direct copyright infringement and (in the alternative)
secondary/contributory liability. In terms of unauthorized reproduction, it was
reasonable to infer that the programming data remained on Russian TV’s US
equipment for “at least several minutes” as required under Cartoon Network because the complaint alleged that defendants
“store” Channel One’s programming signals “via equipment, including computer
servers, located in the United States.” For secondary liability, given the
allegations that Russian TV offered its streaming services “at far lower prices
than lawfully licensed services,” it was plausible that Russian TV knew or had
reason to know it was selling programming acquired through infringing activity
and materially contributed to the unauthorized reproduction “by creating demand
for it, advertising access to it on the Russian TV website and distributing it
to subscribers.”

Trademark and false advertising claims were, however,
dismissed. The complaint didn’t plausibly plead that Russian TV’s acts were
likely to confuse consumers as to whether Channel One originated or sponsored
Russian TV. Nor were facts alleged to show that consumers are likely to be
confused about whether Russian TV or Channel One was the source of the programming.
Instead, the allegation was merely that Russian TV displayed Channel One’s
marks. “[A]s a general rule, the Lanham Act does not impose liability for the
sale of genuine goods bearing a true mark even though the sale is not
authorized by the mark owner because such a sale does not inherently cause
confusion or dilution.”

Comment: this analysis skips over a lot of detail that is
fought out in other cases, particularly the various Phoenix Entertainment cases, but it gets to the right result. There’s no such thing as a “counterfeit”
digital signal in the way there can be a counterfeit purse, as long as the
signal contains the content that is promised (here Channel One content). A deepfake might be a counterfeit in the
purse sense, but that’s not what’s alleged. The fact that the goods/services
are intangible shouldn’t change this result (especially since the pixels on
your screen don’t “come from” the broadcaster in any physical sense; the
directions for how to configure them ultimately have the content creator as
their source, but only through multiple intermediaries—and that’s true whether
or not there’s an unauthorized rebroadcaster in the middle). Dastar
would be another way to get to this result, but it’s unnecessary where the
court can see clearly the core issue, which is copyright.

False advertising: the complaint failed to plead anything
that was false and advertising. Russian TV’s website states “Watch Russian TV
online[.] Over 200 … channels.” “Although this may be advertising, it is not
alleged to be false.” The actual broadcasts of Channel One content weren’t
“advertising or promotion” nor were they alleged to be false. Even the display
of Channel One trademarks at the corner of the screen when consumers streamed
the content wasn’t “commercial advertising or promotion” because that wasn’t an
organized campaign by defendants to penetrate the relevant market. To the
extent the false advertising claim was based on the theory that Russian TV
misled consumers by “making it appear that Defendants have the right to
rebroadcast [Channel One] Programming and Channels,” that didn’t work because
claims about licensing aren’t misrepresentations about “the nature,
characteristics, qualities, or geographic origin of … goods or services,” as
required by §43(a)(1)(B).

The DMCA anti-trafficking claim was also dismissed. The DMCA
relevantly bars people from providing or otherwise trafficking in any
“technology, product, service, device, component, or part thereof, that is …
designed or produced for the purpose of circumventing a technological measure.”
17 U.S.C. §§ 1201(a)(2), (b)(1). But the complaint carefully didn’t plead that
the STBs provided to customers themselves circumvented any encryption, as
opposed to receiving the results of prior circumvention. Accessing what’s
already been obtained is not circumvention.

State law claims for unfair competition and infringement
were, of course, preempted by the Copyright Act. Channel One tried to argue
that Russian TV was “passing off the Russian version of Channel One Programming
as the edited version of the Channel One Programming that is licensed to cable
companies and IPTV providers.” First, relevant underlying facts weren’t alleged
in the complaint, but more importantly, that wasn’t a passing off theory: “this
alleged misrepresentation does not deceive consumers into believing they are
watching Channel One programming when they are in fact watching Russian TV
content.” So this theory was still preempted.

from Blogger https://ift.tt/2U7Bmyp

Posted in Uncategorized | Tagged dastar, dmca, false advertising, secondary liability, Selling copyright-infringing content isn't trademark infringement or false advertising copyright, trademark | Leave a comment

Fair use amicus for Google/Oracle cert petition

Posted on February 25, 2019 by rtushnet

Pam Samuelson’s amicus focusing on copyrightability was excellent and I hope it helps persuade the Court to finally take up this important issue in computer software. I worked on a fair use amicus to point out that the Federal Circuit’s override of a jury verdict in favor of fair use was also a big problem.

from Blogger https://ift.tt/2BMq6QT

Posted in Uncategorized | Tagged Fair use amicus for Google/Oracle cert petition copyright | Leave a comment

Second Circuit summarily reverses bad Prevagen decision on statistical proof

Posted on February 21, 2019 by rtushnet

FTC v. Quincy Bioscience Holding Co., 17-3745-cv(L) (2d.
Cir. Feb. 21, 2019)

Quincy sold Prevagen dietary supplements, claiming (1) that
the supplements improve memory and provide other cognitive benefits, (2) that
these effects are clinically proven, and (3) that the products’ active
ingredient “supplements” brain proteins that are lost with age. The FTC alleged
that Quincy conducted a randomized, double-blind, placebo-controlled study that
contradicted these representations and showed no statistically significant
improvement in the memory and cognition of participants taking Prevagen over
participants taking a placebo. Quincy subsequently “conducted more than 30 post
hoc analyses of the results” of the study, and “the vast majority of these post
hoc comparisons failed to show statistical significance.” While the study showed
a “few positive findings on isolated tasks for small groups of the study
population,” these findings allegedly did not “provide reliable evidence of a
treatment effect.” The district court, in a poorly reasoned opinion, dismissedthe complaint because the post hoc analyses must have shown something. The court
of appeals summarily reversed.

“The FTC has stated a plausible claim that Quincy’s
representations about Prevagen are contradicted by the results of Quincy’s
clinical trial and are thus materially deceptive in violation of the FTC Act
and New York General Business Law.” For example, Quincy claimed broadly that in
a clinical study “Prevagen improved memory for most subjects within 90 days,” yet
this wasn’t true: the study “failed to show a statistically significant improvement
in the treatment group over the placebo group on any of the nine computerized
cognitive tasks.” Not only did this make plausible that it was deceptive to claim
that “the majority of people” experience cognitive improvement from taking
Prevagen, but the FTC also stated a claim that Quincy’s representations that
this cognitive improvement is clinically supported are deceptive. Further, the
FTC alleges that Quincy’s claim that the active ingredient in Prevagen, apoaequorin,
“enters the human brain to supplement endogenous proteins that are lost during
the natural process of aging” was false. Quincy’s “safety studies show that
apoaequorin is rapidly digested in the stomach and broken down into amino acids
and small peptides like any other dietary protein.” “Drawing reasonable
inferences in favor of the FTC, as we must, the FTC plausibly alleged that
Quincy’s representations about Prevagen’s active ingredient entering the brain
are false.” The district court erred in dismissing the complaint and declining
to exercise supplemental jurisdiction over New York’s claims.

The court doesn’t go into detail on the bad statistical
understanding underlying the district court’s mistakes, but you can read
the amicus brief I signed or let xkcd do it.

from Blogger https://ift.tt/2T8imSV

Posted in Uncategorized | Tagged false advertising, ftc, Second Circuit summarily reverses bad Prevagen decision on statistical proof consumer protection | Leave a comment

Show me your teeth: dentist’s before-and-after closeups aren’t copyrightable for want of originality

Posted on February 21, 2019 by rtushnet

Pohl v. MH Sub I, LLC, No. 4:17cv181-MW/CAS (N.D. Fla. Jun.
20, 2018)

H/T Daniel Kegan. Dr. Pohl took before-and-after photographs
of his cosmetic dental work for his Florida cosmetic dentistry practice’s
website. He took such photos of “Belinda” in 2004. “The photos consist of two
direct shots of the patient’s teeth—one before the dental work and the other
after the dental work. The patient is revealing her teeth and, in both shots,
the photo consists of her teeth, her lips, and a small area around the mouth.”

the photos at issue

He sued defendants for designing/developing
seven websites on which the images appeared without his permission. Although the deposit copy of the site he used
for his copyright registration wasn’t in the record, the jury could find that
it contained the images at issue.

The court then found that the before-and-after photos
weren’t copyrightable because no reasonable jury could find the photos were
sufficiently creative or original. Though the bar was low, the photos failed to
pass it. Highly informative product
photos intended to sell the product through accurate depiction may lack
sufficient creativity to be protected, and that was what happened here. Such
photos “lack any creativity or originality primarily because they serve a
utilitarian end—to identify goods or services that a viewing customer can
expect from the business.”

Pohl argued that he was responsible for “selecting the
camera, posing the subject matter, and determining the lighting and photo angle
before taking the photographs.” But none of this involved creativity. He didn’t remember what kind of camera he
used—even whether it was digital or film.
He didn’t remember whether Belinda was sitting or standing for the
photos. “To the extent he posed her for
the camera, it was to tilt her head, lift her chin up or down, instruct her to
smile, or to tell her to look at the camera. As for lighting, there is no
creativity in merely having sufficient lighting in the room ….” For the angle,
he testified that he moved “the camera in and out until I get it in focus,” which
the court deemed “the most rudimentary and basic task for photographers since
the era of the daguerreotype.” The entire process took no more than five
minutes.

If utilitarian photos of Chinese food in different patterns
are uncopyrightable, as a previous case held, then these photos are clearly
uncopyrightable; “directing a subject to smile and moving a camera to focus on
a portion of the subject’s face” has even less of a “creative spark.”

from Blogger https://ift.tt/2IvEEKn

Posted in Uncategorized | Tagged Show me your teeth: dentist's before-and-after closeups aren't copyrightable for want of originality copyright | Leave a comment

	La puissance des dét… on Package size can be false…
	Sava v. 21st Century… on bad influence: claims against…
	Heightened Constitut… on Initial thoughts on Elste…
	Heightened Constitut… on Initial thoughts on Elste…
	ChatGPT Ist In Den B… on I was on Hard Fork to talk cop…

Rebecca Tushnet

copying without infringement, falsity without materiality in adhesive case

Omission of side effects in lash “cosmetic” ads was plausibly false & misleading

Belmora doesn’t replace 43(a)’s usual requirement of protectable subject matter in standard trade dress case

ICANN and the New Top-Level Domains, panel 2

ICANN and the New Top-Level Domains part 1

Misleading advertising that’s already leaving the market can’t support finding irreparable harm

Selling copyright-infringing content isn’t trademark infringement or false advertising

Fair use amicus for Google/Oracle cert petition

Second Circuit summarily reverses bad Prevagen decision on statistical proof

Show me your teeth: dentist’s before-and-after closeups aren’t copyrightable for want of originality

Recent Posts

Recent Comments

Archives

Categories

Meta