Property, intangible

a blog about ownership of intellectual property rights and its licensing


  • What We Don’t Know in Warhol – Who the Infringer Is and By Doing What

    I expect to see a lot of reports that the recent Supreme Court decision in Andy Warhol Foundation for the Visual Arts, Inc. v. Goldsmith held that Andy Warhol’s Prince series of 16 works, or at least the “Orange Prince” that was licensed to Condé Nast, is an infringement of Lynn Goldsmith’s photograph. In fact, what the infringement was, and by whom, is a mystery.

    The Supreme Court described the infringement this way:

    Here, Goldsmith’s copyrighted photograph has been used in multiple ways. The Court limits its analysis to the specific use alleged to be infringing in this case – AWF’s commercial licensing of Orange Prince to Condé Nast – and expresses no opinion as to the creation, display, or sale of the original Prince Series works.

    But granting a license isn’t reproducing, distributing, creating a derivative work, displaying, or performing, so merely granting a license shouldn’t be an infringement. Assume this hypothetical – Condé Nast acquires a copy of the Warhol without AWF’s knowledge or assistance and Condé Nast then sought a license when it decided to commercialize the work. AWF could have said “You’re licensed to the Warhol, you’re on your own for Goldsmith.” In that case, AWF would not have committed any infringing act, either directly or indirectly.

    This is not to say that AWF didn’t infringe, only that granting a license – signing a contract – can’t be it. If AWF made a copy of Orange Prince to give to Condé Nast, that could be an infringing act. But the most obvious use – reproducing it as the cover of the magazine – wasn’t AWF’s infringement. That exercise of the exclusive right was by non-party Condé Nast, not AWF. That might make AWF secondarily liable for infringement, but that’s a different analysis.

    What do we know from the decisions? AWF had filed a declaratory judgment action, asking the court to declare that the creation of the Prince series was not an infringement.1 Goldsmith then counterclaimed for infringement. However, she had statute of limitations problems with alleging that the creation of the original works was infringing, so she alleged that AWF’s granting of licenses was the infringement.

    On cross-motions for summary judgment, the district court punted on infringement:

    AWF argues that it did not infringe Goldsmith’s copyright to her Prince Photograph because none of Warhol’s Prince Series works, including the work licensed to Condé Nast in 2016, are substantially similar to the Goldsmith Prince Photograph under the “ordinary observer test” for substantial similarity. But the Court need not address this argument because it is plain that the Prince Series works are protected by fair use.

    So we have two different theories for infringement, Warhol’s original creation of the Prince Series and the licensing of Orange Prince. The court of appeals followed the same path, although disagreeing with the district court’s evaluation of fair use and further finding that, as a matter of law, the Warhol work infringed the Goldsmith photograph.

    So we just don’t know what the conduct was that was infringing. But how can one do a fair use analysis without knowing what the use was against which to measure “fair”? This decision is probably best thought of as clarifying the first fair use factor, but nothing beyond that.

    Andy Warhol Found. for the Visual Arts, Inc. v. Goldsmith, No. 21-869, 2023 WL 3511534 (U.S. May 18, 2023)

    Andy Warhol Found. for the Visual Arts, Inc. v. Goldsmith, 11 F.4th 26 (2d Cir. 2021)

    Andy Warhol Found. for the Visual Arts, Inc. v. Goldsmith, 382 F. Supp. 3d 312 (S.D.N.Y. 2019)

    Creative Commons License
    This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.


    1. AMF seems to have also made a questionable assumption, which is that if the Warhol work was created lawfully, every later use of it will also be lawful. 
  • Exclusivity for Exactly What?

    Does an exclusive agent for a photographer have standing to bring a copyright infringement suit on behalf of that photographer? The Ninth Circuit has said yes; the Northern District of Georgia says no.

    Plaintiff Creative Photographers, Inc. (“CPi”) represented non-party photographer Ruvén Afanador. The defendants are accused of infringing the copyright in one of his works:

    Black and white photographic portrait of Ruth Bader Ginsberg on the left; a derivative work with her in an orange robe made up of small paint strokes and a background made from cutouts of text, presumably cut from her legal opinions. The lefthand image is captioned "Figure 1: The Afanador Work, titled 'Ruth Bader Ginsberg" and the righthand image is captioned "Figure 2: An example of the Torres Works, Titled 'A Judge Grows in Brooklyn'"

    In the agreement between CPi and Afanador the relevant language was this:

    You retain [Plaintiff] as your exclusive agent to sell, syndicate, license, market or otherwise distribute any and all celebrity/portrait photographs and related video portraits, submitted to us by you and accepted by us for exploitation for sale or syndication during the term of this Agreement (the “Accepted Images”). You must be the sole owner of the copyright for all such photographs and may not offer any celebrity/portrait photographs for sale or syndication to or through any other agent, representative, agency, person or entity during the Term of this Agreement.

    Based on this language, CPi brought suit in its own name against the defendants. Is this language that transfers to CPi one of the exclusive rights of copyright that would give it statutory standing to file a copyright infringement lawsuit?

    The court relied on an earlier district court opinion in the same circuit, Original Appalachian Artworks, Inc. v. Schlaifer Nance & Co., 679 F. Supp. 1564 (N.D. Ga. 1987), for the proposition that “a grant of the right to act as an ‘exclusive agent’ for a copyright owner in authorizing others to use a copyrighted work is a mere contractual right, not a copyright enforceable under the Copyright Act.” (emphasis in original).

    The CPi agreement granted it the right “to sell, syndicate, license, market or otherwise distribute” the Afanador work. CPi claimed this language was “a clear, explicit and exclusive grant of [§] 106 rights” in the Afanador work.

    The court disagreed:

    It is true that in Original Appalachian, SN & C did not tie their standing argument to a specific § 106 right; instead, SN & C contended that “the expansive language describing the [§] 106 rights was broad enough to encompass SN & C’s exclusive right to authorize other to us the OAA’s copyright. … However, … [t]he [CPi] Agreement “explicitly and conclusively” makes Plaintiff Afanador’s agent but its terms do not otherwise expressly convey to Plaintiff an exclusive license to any of the § 106 rights. Other than reciting the text of the Agreement and asserting that the language confers § 106 rights, Plaintiff simply has not provided this Court with legal authority or other argument to interpret this language in any other manner.

    The district court found support in several cases out of Colorado (one blogged here) brought by stock photo agency Viesti, which reached the same conclusion that the agency agreement did not give the agent standing for an infringement suit.

    The Ninth Circuit has, though, in Minden Pictures, Inc. v. John Wiley & Sons, Inc. held that a photographers’ agent had standing for a copyright infringement suit. I’ve always struggled with this decision and the Georgia court here distinguished it based on the different wording of the agreements at issue in the two cases:

    However, Minden is distinguishable. The agency agreements in that case contained clear language indicating that they transferred to Minden an exclusive right (rather than merely appointed Minden as an exclusive agent): “The [a]greements also confer upon Minden ‘the unrestricted, exclusive right to distribute, License, and / or exploit the Images … without seeking special permission to do so.’ ” The Agreement in this case lacks any such language. In Minden, too, it seemed fairly clear that the agency agreements constituted some kind of license; the analysis hinged on whether that license was exclusive or nonexclusive. Here, though, while the Agreement certainly makes Plaintiff an exclusive agent to perform for Afanador certain functions, the Court does not agree with Plaintiff that the Agreement, on its face, constitutes an exclusive license. Finally, the Court is also more inclined to find authority from this circuit, like Original Appalachian, persuasive than a case from another circuit altogether.

    CPi had a second standing problem. CPi characterized the defendants’ works as “derivative works” in the complaint, but the agency agreement didn’t grant any license to CPi for derivative works:

    Although Plaintiff asserts in the First Amended Complaint that it holds an exclusive right to prepare derivative works, this is a legal conclusion that the Court need not accept as true. In any case, the Agreement clearly shows to the contrary. Thus, even if Plaintiff held an exclusive license under the Agreement, it does not appear that Plaintiff holds an exclusive license to the right—the preparation of derivative works—that it alleges Defendants infringed.

    Case dismissed with leave to amend. I don’t know if we’re headed to a circuit split; I think it depends on whether the 9th Circuit doubles-down on its holding or finds a way to walk it back. But its reasoning doesn’t seem to be getting traction with any other courts.

    Creative Photographers, Inc. v. Julie Art, LLC, No. 1:22-CV-00655-JPB (N.D. Ga. Mar. 13, 2023)

    Creative Commons License
    This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.

  • A Copyright Chicken and Egg Problem

    As told in the complaint, plaintiff Gail Taylor nee Bridges started her concert photography career in high school, with some of her photographs published in a 1975 high school yearbook. By 1980, she had at least 1000 photos of the most famous artists and bands of the era. She stored the photos at her parents’ house.

    In 1981 Taylor married the late Timothy Aaron, whose sister, Myra Nix, and niece, Andrea Johns, are defendanats in the lawsuit. Taylor and Aaron divorced in 1989. In September 1989, Taylor moved to San Antonio and left her furniture and photos at her parents’ house. She discovered the photos were missing in 1990 and contacted her ex-husband; he admitted to entering Taylor’s parents’ house and taking furniture but denied having the photographs. He said that all the concert photos he had were stolen from his truck and never recovered.

    Aaron passed away in 2014. Defendants Nix and Johns found a box of Taylor’s photos, including Taylor’s concert photos and personal and childhood photos, in a shed on the ex-husband’s property. Johns contacted Taylor and agreed to return the photos. She did not though; instead, Nix, Johns and co-defendant Trina Hill formed defendant Mystery Ship LLC. Mystery Ship registered the copyright in some of the photos, including ones of Freddy Mercury and REM, and allegedly sold a license to publish at least one of them.

    Taylor filed a lawsuit in Gwinnett County (GA) Superior Court for the return of the photographs. She thereafter filed a lawsuit in federal court for a preliminary injunction, although the basis for a federal claim could have been a little crisper. The defendants filed a motion to dismiss the federal lawsuit on several grounds, including that the suit was a claim for return of the physical property, not copyright infringement, and, if it was a claim for copyright infringement, the claim should be dismissed because Taylor had not registered the copyright in the photos.

    On the defendants’ first theory, the court held that the complaint stated a claim under the Copyright Act:

    Although Plaintiff’s Complaint focuses on the relief sought—a temporary and permanent injunction—rather than a substantive federal claim, it is ostensibly a copyright infringement action. Plaintiff cites 17 U.S.C. § 408(a) and 17 U.S.C. § 502(a) as the bases for her suit.

    However, while § 502 allows a court to grant injunctive relief on a copyright claim, one must have registered the copyright before bringing a lawsuit. Taylor did not have any copyright registrations—she couldn’t, because she wasn’t in possession of the photographs.

    Taylor argued that Foundation for Lost Boys & Girls of Sudan, Inc. v. Alcon Entertainment, LLC, No. 1:15-cv-00509-LMM (N.D. Ga. Mar. 22, 2016) provided an exception to the registration requirement. In that case, the plaintiff also did not have possession of the photographs and therefore could not register the copyright. In Lost Boys the lawsuit was not dismissed, though, because there were co-pending counts that would allow the plaintiff to take possession of the copies and therefore be able to register the copyright. But recall that Taylor first filed a state law claim for return of the photographs. By the time the federal court decided the motion to dismiss, the state court jury had decided against her on her claim for return of the photos.

    Therefore, unlike the plaintiffs in Foundation for Lost Boys & Girls, Plaintiff’s opportunity to register the copyrights has already been foreclosed. And without any valid substantive claim to sustain her action, Plaintiff cannot get injunctive relief.

    The court also distinguished an earlier Eleventh Circuit decision, Pacific & Southern Co. v. Duncan, 744 F.2d 1490 (11th Cir. 1984). In Duncan, the court granted an injunction against future works, which by definition lack a copyright registration. However, Duncan “allowed injunctive relief to be sought prior to registration because the plaintiffs had already shown infringement of a registered work and demonstrated that future infringement was both likely and predictable based on the series of works at issue” (quoting Fourth Est. Pub. Ben. Corp. v. Wall-Street.com, LLC, 856 F.3d 1338, 1342 (11th Cir. 2017)). Here, though,

    Plaintiff admits that she has not registered any copyrights. She has not indicated that she continues to create similar works, and she has not alleged potential infringement of future works. Instead, she alleges only that Defendants have already misused her photos, taken several decades prior, by registering and selling them without her permission and that they may continue to infringe her rights in the same existing set of photos, which is already in Defendants’ possession.

    The case was therefore dismissed. With cold comfort, the court advised her

    that if Plaintiff does acquire her photographs at a later date, she may be able to register copyrights and then bring an infringement action, at which point she would be able to recover for any infringement that occurred before registration.

    The verdict form for the state law case had claims for conversion, theft of lost or mislaid property, unjust enrichment, and tortious interference. None of the defendants were the ones who took the photos or misled Taylor about their existence, so it’s understandable why the claims for return of the photos failed. But it seems undisputable that Taylor owns the copyright and would likely have a successful infringement claim, if she can only get the darn things registered. This seems like a statutory gap, even though the fact patterns seems quite likely to arise—anytime an artist creates a single copy, such as a painting or sculpture, without remembering to take a photo before selling it.

    I’m not sure why Taylor couldn’t register the copyright in photos that the defendant had registered, or why she didn’t plead infringement claims for the one that reportedly was licensed by the defendant. She should have had access to the images for the registered works, either from whomever might have published them or by obtaining copies of the deposit from the Copyright Office. She may have a successful stand-off though—if, every time the defendants publish a photo, Taylor then registers the copyright in the photo and sues them and the party that was granted the license, the defendant can’t build much of business on the photos. There should be some bargain that they can strike.

    Taylor v. Mystery Ship, LLC, No. 1:22-CV-01617-LMM (N.D. Ga. Dec. 19, 2022).

    Creative Commons License
    This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.

    Update May 5, 2023:

    For some reason a comment is not visible. The comment, from Trina-Marie Hill, is as follows:

    This article is inaccurate.
    Mystery Ship was formed before any of the defendants were in contact with Gail Taylor.
    Also, Johns contacted Gail Taylor to return the personal photos of Taylor.
    The trial ended in October and Mystery Ship and all defendants were found not guilty on all charges.
    The Federal Copyright case was dismissed.

  • The Smallest of Details

    We’re all generalists for one thing or another. We have specialized knowledge in some field, but the practice of law isn’t so neat that everything you do falls within your core knowledge. Those who trade in contract work may cover many different types of deals and will necessarily rely on forms to take care of specialized requirements. An example is “goodwill” in a trademark assignment – woe be the drafter who omits it, because it may invalidate the entire assignment. In Inventist, Inc. v. Ninebot, Inc., it was a similar detail that was the plaintiff’s downfall.

    Non-party Shane Chen was the inventor of utility and design patents for an electric unicycle sold as the Solowheel. There is no dispute he was the inventor of several patents, but there were several assignments that are the monkey wrench in the works.

    Chen assigned the patents to plaintiff Investist, Inc., which owned them until August 3, 2017. After the original complaint was filed, Chen entered into an Operating Agreement with a Chinese company to establish a joint venture. The Operating Agreement defines “Intellectual Property” as the “Proposed Patents and the brand of Solowheel trademarks, websites, domain names and proprietary technology.” If the agreement terminated, “the ownership of any Intellectual Property transferred from [Shane Chen] to the Joint Venture shall immediately and automatically revert back to [Shane Chen].” Pursuant to the agreement the patents were assigned to Solowheel, Inc. and Solowheel substituted as plaintiff.

    The Operating Agreement terminated and the patents in dispute were transferred back to Chen by two written assignments dated March 2, 2019 and September 19, 2019. Chen then assigned the patents back to plaintiff Investist on October 21, 2019. Investist was substituted back as plaintiff for Solowheel. Defendant Ninebot claimed that Investist’s damages were limited to any infringement occurring after October 21, 2019 and the court agreed.

    The written assignments of the patents assigned “any and all rights, title and interest in and to the patents and patent applications” to Chen. However,

    Plaintiff does not dispute that [Solowheel] never assigned Chen the right to sue for past infringement but argues that Chen separately acquired that right by virtue of the Operating Agreement’s reversion provision. Specifically, Plaintiff argues that, upon the dissolution of the joint venture in 2019, all of the intellectual property rights Chen had transferred to that entity, including the right to sue for past infringement of the patents, reverted back to Chen pursuant to that provision.

    So did the reversion on the Operating Agreement transfer the right to recover for any past infringement?

    The Court finds that the Operating Agreement’s reversion provision did not, by its terms, grant Chen the right to sue for past infringement. As noted above, a party may sue for past infringement transpiring before it acquired legal title if a written assignment expressly grants the party a right to do so. For example, in Messagephone, the Federal Circuit held that an agreement to assign the entire right, title, and interest to a patent did not encompass the right to sue for past infringement due to the absence of any explicit language conveying that right. Messagephone, Inc. v. SVI Sys., Inc., 2000 WL 1141046, at *4-5 (Fed. Cir. Aug. 11, 2000).…

    The Operating Agreement’s reversion provision simply provides for the reversion of “any Intellectual Property.” In defining the “Intellectual Property” that Chen transferred to the joint venture, the Operating Agreement included, without any reference to past infringement, “all the Proposed Patents and the brand of Solowheel trademarks, websites, domain names and proprietary technologies.” Absent from the Operating Agreement–including its terms governing the initial transfer to the joint venture, and the reversion back to Chen–is any reference to the right to sue for past infringement. As such, Chen did not acquire the right to sue for past infringement of the patents by virtue of the Operating Agreement’s reversion provision.1

    I can see how this could be an easy one to miss. A definition of “Intellectual Property” generally doesn’t include claims too. If you are only thinking that you want to return the parties to the former status quo, it might not dawn on you that there were rights left behind. But here, the lawsuit was already pending when the business deal was made, so the gap is more surprising.

    Investist, Inc. v. Ninebot, Inc., No. 3:16-cv-5688-BJR (Jan. 18, 2023)

    Creative Commons License
    This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.


    1. I’m a little confused because the complaint was filed before the assignment to the joint venture. It seems that Investist should still have a damages claim for infringement before the assignment to the joint venture – perhaps it does and the pre-assignment damages aren’t disputed by the defendants. 
  • Can They Do That? Dungeons & Dragons Edition

    I hadn’t paid a lot of attention to the furor over a proposed license change to the Dungeons & Dragons game by Wizards of the Coast until I read an article on Ars Techica. The original license, called the Open Game License Version 1.0a, is what I would call an open culture license, granting liberal use of copyrightable content.1 The license was inspired by the free software movement and the GNU General Public License. As explained by Ars Technica, “this wasn’t just altruism on WotC’s part; Dancey said the license would encourage the kind of network externalities that would make the D&D rules system more popular, thus increasing sales of the game’s core rulebook and allowing others to profit off of content based on that system.”

    However, a new license for Dungeons & Dragons, the Open Game License Version 1.1, was recently leaked and the outrage commenced. Under the new license, WotC was demanding royalties for large commercial exploitation of Dungeons and Dragons. WotC has since walked back the release of the license. The similarity of the original game license to open source licenses is what caught my attention.

    The grant in OGLv1.0a was of a “perpetual, worldwide, royalty-free, non-exclusive license with the exact terms of this License to Use, the Open Game Content.” The license also says “Wizards or its designated Agents may publish updated versions of this License. You may use any authorized version of this License to copy, modify and distribute any Open Game Content originally distributed under any version of this License.” These terms strongly parallel those found in open source software licenses, a field in which I practice.

    The draft of the new license says that “this agreement is … an update to the previously available OGL 1.0(a), which is no longer an authorized license agreement.” In the new license, WotC makes it clear that use of Open Game Content that was available for use at no cost prior to the adoption of the new license would now bear a royalty for large commercial exploiters. In other words, in the new license WotC was claiming that OGLv1.0a was no longer valid and the “perpetual” grant in it was terminable at will. Can they do that? No one has thought that to be the case in open source licensing, so this was, to me, a novel effort.

    And … I don’t think so. We’ll start with enforceability of the terms of the original agreement against WotC. EFF’s original analysis was that the license was a “bare” license, i.e., a grant of rights by WotC that is not enforceable as a contract. They now have updated their guidance to state that it likely is an enforceable contract.

    However, even a promise cannot be rescinded if there has been reliance on it. “A promise which the promisor should reasonably expect to induce action or forbearance on the part of the promisee or a third person and which does induce such action or forbearance is binding if injustice can be avoided only by enforcement of the promise.” Restatement (Second) of Contracts § 90. People making a living by exercising the rights granted to them under the original license certainly were relying on the promise. The grant was perpetual, meaning it had no defined end. Trying to revoke or terminate the grant where the user is otherwise fully in compliance with the conditions on the promise would be a basis for making a promissory estoppel claim against WotC.

    If it’s a contract, the original agreement did not have any provision allowing for termination for convenience, only for cause. Trying to revoke or terminate the grant without cause would give the user grounds for a breach of contract claim.

    WotC anticipated this problem. In OGLv1.0a, the Open Game Content could only be used pursuant to the terms of “any authorized version of this License.” (§ 9) In the open source software realm, this is understood to mean that a user can rely on this version of the license or a later version – that is, the user has the option of relying on the original terms or instead availing themselves of favorable changes to the license. They also therefore have the power to reject a subsequent version of the license that has unfavorable terms, as is the case here. The word “authorized” was probably in the OGL1.0a to make sure no one tried to create a modified version of the OGL license not approved by WotC, or use a different license entirely, and claim that it applied to the Open Game Content.

    But draft version 1.1 of the license is trying to hack this construction. The new license says “the previously available OGL 1.0(a) … is no longer an authorized license agreement.” By “deauthorizing” the original license, they are claiming that it is no longer available as a license for the Open Game Content anymore. Maybe it works, I don’t know, hacks work sometimes (GPL, looking at you). But it is undoubtedly contrary to the intent of the original license and original license drafters.

    There may be contract formation problems too, but I don’t know how WotC plans to roll it out. The new license says “by using Licensed Content in this manner, You agree to the terms of this agreement.” I don’t know how, though, WotC will be able to show that the user was aware of, and assented to, the new license rather than using the game content based on the permission granted in the original license.

    I hope their effort isn’t legally effective, although we’ll probably never know since they claim to be rewriting the license. But it might disrupt far more than the D&D world if their effort succeeded.

    Possible or not, there is also the question “should they do that”? In my opinion, that’s a resounding “no.” They have built an empire by open licensing and now are reneging. We see this effort fairly often in open source licensing, where someone wasn’t thoughtful about their original license choice and how much they were giving away, or they deliberately used a permissive license to gain a following but later decided they wanted a piece of others’ pies, or they are pressured by investors into a more robust monetization and can’t think of any route to monetization other than exploitation of copyright. But I’ve not seen anyone so bold as to think they could terminate the original grant. It’s a complete betrayal of a very devoted following. Despite WotC’s guarded and limited mea culpa, they have damaged their reputation and lost the trust of gaming companies who are threatening to abandon D&D. They’ve shown that they don’t think they have any obligation to maintain a 20-year promise that hundreds of thousands of people trusted. The draft contract also has some terms designed to heavily favor their own rights and disenfranchise users.2 It looks very much like the “rights ratchet” that happens in open source. I think WotC has overplayed their hand.

    Creative Commons License
    This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.


    1. The Electronic Frontier Foundation explains why the original grant may not actually be granting any rights at all and points out how you might be worse off even with the original license than you would otherwise be with no license at all. 
    2. They are giving themselves a license to the user’s own original works (§ X), although in their mea culpa they say they aren’t going to do that. For non-commercial users, they are now imposing copyleft on the user-generated content when the original license was a permissive one. (§ V). They require that anyone making more than $50,000 report to WotC, although they won’t owe any royalty – for now. 
  • It’s Best If the Registrant Files the Lawsuit

    This is something that I probably shouldn’t have to blog about, but here we are.

    Plaintiff Palm Beach Concours LLC filed a complaint against defendant SuperCar Week, Inc. for: Count I, trademark infringement in violation of 15 U.S.C. § 1114 (infringement of a registered trademark); Count II for unfair competition in violation of 15 U.S.C. § 1125(a)(federal unfair competition); Count III for Florida common law trademark infringement; and Count IV for Florida common law unfair competition.1

    I have no idea what either party does or what the alleged infringement was—the opinion didn’t get that far. SuperCar filed a motion to dismiss under Federal Rule of Civil Procedure 12(b)(1) for lack of subject matter jurisdiction and under Federal Rule of Civil Procedure 12(b)(6) for failure to state a claim upon which relief can be granted. What’s the problem?

    Palm Beach Concours hadn’t suffiently alleged ownership of a trademark. There was a state trademark registration, but spot the problem:2

    Palm Beach Concours FL trademark registration

    The record owner is Sidney Vallon, the co-owner (with his wife) of Palm Beach Concours, but the trademark is not registered by Palm Beach Concours itself. As explained by the magistrate judge, “If Plaintiff is not the owner of the Mark, it has not suffered a constitutionally-sufficient injury-in-fact, even if Defendants are infringing on the Mark.” The court dismissed the complaint under Rule 12(b)(1) and declined to exercise supplemental jurisdiction over the state law claims.

    A further problem was that Count I was for infringement of a federally registered mark and the plaintiff did not claim to own a federally registered mark. Count I was thus also dismissed for failure to state a claim.

    Palm Beach Concours asked that Vallon be joined as a real-party-in-interest under Fed. R. Civ. P. 17(a)(3), but no go:

    On the current record, the Court cannot exercise any judicial power other than to dismiss the case without prejudice. That means the Court cannot allow joinder under Rule 17(a)(3). Plaintiff’s remedy is to file a separate motion for leave to file an Amended Complaint that cures the jurisdictional defect. See SDFL Local Rule 15.1.

    And a practice note: the court also dismissed the complaint as “an improper shotgun pleading” because

    Counts II, III, and IV each incorporate all prior counts by reference. Weiland v. Palm Beach Cnty. Sheriff’s Office, 792 F.3d 1313, 1321 (11th Cir. 2015) (“The most common type [of shotgun pleading] – by a long shot – is a complaint containing multiple counts where each count adopts the allegations of all preceding counts, causing each successive count to carry all that came before and the last count to be a combination of the entire complaint.”). A court confronted with a shotgun pleading should sua sponte require repleading. Id. at 1321 n.10.

    Yikes. The complaint was only ten pages with only four counts, all of them essentially for trademark infringement. Perhaps it was meant as a lesson, since the plaintiff’s lawyers didn’t appear to have done their homework on the elements of the various claims.

    But let’s pause for a moment. The plaintiff perhaps could have stated a claim for federal unfair competition. A claim can be brought under § 43(a) (the basis for Count II) by anyone “who believes that he or she is or is likely to be damaged by such act,” which might include the licensee. J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition § 32:12 (5th ed. Dec. 2022 update). So it may be possible to allege that Palm Beach Concours was a licensee that was harmed, but the complaint didn’t. Even if alleged, it’s likely that the owner would have to be joined as a necessary party, but it saves the filing date.

    This was the magistrate’s report and recommendation and the parties have until January 13, 2023 to file objections.

    Palm Beach Concours, LLC v. SuperCar Week, Inc., Civ. No. 22-80888-CV-DMM (Dec. 30, 2022)(report and recommendation)

    Creative Commons License
    This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.


    1. No idea why they didn’t allege infringement of a state registered trademark under Fla. Stat. § 495.131
    2. And no, it’s not “incontestable.” Complaint § 12. That’s a federal concept not used in the Florida trademark statute. 
  • A Scheme to Steal Software?

    I’ve been meaning to write about this case for a long time, and was finally inspired by the “Order Finding Defendants Entitled to Attorneys’ Fees,” entered after the case ended on a motion to dismiss, which tells you it went resoundingly badly for the plaintiff. Long-read warning.

    Since the case ended on a motion to dismiss, the facts are as claimed by the plaintiff, which I’ve assembled from the decisions, the complaints and the motions. The individual defendants are people who participated in the development of the NGINX (pronounced “Engine-X”) web server software while employed by non-party Rambler Internet Holdings LLC. Rambler is Russian technology company and search engine. Plaintiff Lynwood Investments CY Limited acquired any rights Rambler had in the software in 2015.

    According to Lynwood, defendant Igor Sysoev, a Russian national residing in Russia, developed NGINX in the scope of his employment, beginning in 2001. In 2004 he released a portion of the NGINX software under an open source license (to be exact, the 2-clause BSD license, aka the Simplified BSD or FreeBSD License) allegedly without authorization from Rambler. The case refers to different versions of NGINX, “Open Source NGINX” or “NGINX Software,” “NGINX Enterprise,” which was the “entire web server enterprise,” i.e., the NGINX open source software and presumably other components, and “NGINX Plus,” a proprietary version of NGINX Enterprise with extra features.

    In the original complaint, Sysoev and other individuals named as defendants (the “Team”), “conspir[ed] to steal” NGINX Enterprise. They created legal entities, successfully obtained investment, and starting in December 2011 quit Rambler.

    According to Lynwood, “Sysoev and Konovalov each misrepresented to Rambler that they were separating from Rambler to form a new company that would provide support services to the existing open source NGINX Software for third-parties.… Sysoev and Konovalov fraudulently concealed from Rambler that they were contemplating the prospect of potentially providing add-on services to the NGINX Software in the future. … In an attempt to avoid triggering suspicion at Rambler, Sysoev and Konovalov couched their statements as aspirational and deeply protective in nature.”

    Defendant F5 acquired NGINX and it various corporate entities in March 2019 for $670 million. Lynwood claims to have learned of the defendants wrongful actions sometime after that, when a former NGINX team member blew the whistle to Rambler.

    The corporate defendants in the case were F5, the NGINX companies and NGINX investors. Lynwood originally brought 26 claims against the many defendants, but the court ordered Lynwood to select ten claims to litigate through trial. Lynwood selected civil conspiracy, breaches of duty (or aiding and abetting them), tortious interference (contract and prospective business advantage), and direct copyright infringement. Motions to dismiss followed.

    On a motion to dismiss filed by the corporate entities, the court dismissed all claims, with leave to amend, for three reasons:

    First, these claims are untimely because they stem from the Team’s alleged theft of the NGINX Enterprise from Rambler, which occurred in 2011, nearly a decade ago. Second, the allegations of a fraudulent scheme to steal the NGINX Enterprise from Rambler, which underlie all these claims, do not satisfy the heightened pleading standard of Federal Rule of Civil Procedure 9(b). Third, each of these claims fails to state a claim.

    With respect to the copyright infringement claim, the court’s full analysis was:

    Lynwood has failed to plausibly allege that the F5 Entities and the Outside Investors copied the constituent elements of the work. To state a claim for copyright infringement, the plaintiff must “identify which Defendant is alleged to have infringed which particular copyright.” In the instant case, Lynwood does not identify which defendants allegedly infringed its copyright. Instead, Lynwood lumps all the defendants together without stating which entity or person copied the work.

    The individuals also filed a motion to dismiss alleging lack of personal jurisdiction, which the court denied. However, the court dismissed the claims on statute of limitations grounds as it had done with the corporate defendants, with leave to amend.

    The plaintiff refiled an amended complaint and all defendants filed a consolidated motion to dismiss for failure to state a claim. Lynwood now alleged that:

    Rambler “made a business decision to permit Sysoev to continue releasing Open Source NGINX … because such releases highlighted the technical achievements of Rambler and its employee Sysoev, which inured to Rambler’s benefit for attracting top software programmers.” Lynwood further alleges Rambler’s Chief Technology Officer (“CTO”), Konovalov [a co-defendant], “repeatedly misrepresented to Rambler that Open Source NGINX had no independent commercial value or prospects for monetization” and “that its only use for Rambler was to solve Rambler’s internal technical issues … and to burnish Rambler’s reputation as a worldwide leader in technology innovation.” As a result, Lynwood alleges, “Rambler never undertook efforts to monetize or commercialize Open Source NGINX” and “did not object to Konovalov and Sysoev earning an income off of Open Source NGINX” when they later separated from Rambler in 2011.

    Lynwood dropped the civil conspiracy claim, so nine claims remained. Lynwood still could not state facts that survived the statute of limitations for most of the claims, but did succeed with two—a claim for tortious inference with contract, based on NGINX’s continuing sale of the NGINX software, and the copyright infringement claim.

    The reasoning on the copyright infringement claim is a little disjointed. When considering whether the plaintiff had a timeliness problem, the court considered the gravamen an ownership question:

    The statute of limitations for copyright infringement is three years, see 17 U.S.C. § 507(b), and where, as here, “creation rather than infringement is the gravamen of an authorship claim, the claim accrues on account of creation, not subsequent infringement, and is barred three years from ‘plain and express repudiation’ of authorship” …

    The defendants alleged repudiation when Sysoev released Open Source NGINX without authorization from Rambler and with a copyright notice saying “Copyright (C) 2002-2004 Victor Syseov.” He also gave an interview to Free Software Magazine in 2012 allegedly claiming ownership of NGINX.

    Under US law the copyright notice might be considered an express repudiation, but we have a twist. Syseov is a citizen of Russia and wrote the software while in Russia. Under choice of law, the law of the country with the most significant relationship to a work controls the question of ownership. The court held that Russia had the most significant relationship to the property and therefore Russian law governed. So while in the US the copyright notice states the name of the owner of the copyright, under Russian law the notice states the name of the owner or the author. The copyright notice therefore was not an express repudiation by Syseov. His claim in the magazine article also wasn’t express repudiation; repudiation must be communicated to the claimant and Lynwood did not allege that Rambler or Lynwood ever saw the article.

    As a result, “it is not apparent on the face of the AC [Amended Complaint] that the Fourteenth Claim for Relief is barred by the statute of limitations.”

    Instead, the court dismissed it under Rule 12(b)(6) for failure to state a claim. In the 12(b)(6) analysis, the court’s attention is no longer on ownership, but on who is, or who is not, doing the infringing. The first decision informed Lynwood that, on refiling of the complaint, the plaintiff

    “must identify which defendant is alleged to have infringed which particular copyright,” found Lynwood “lumped all the defendants together without stating which entity or person copied the work.” Moreover, Judge Koh found Lynwood’s allegations that Runa Capital and E. Ventures “bought into the plan and funded it,” rather than “copied the work,” to “belie the notion that all defendants copied the work.”

    In dismissing the claim a second time,

    The Court first notes that Lynwood fails to adequately identify in the AC which particular work or works were copied. Lynwood now identifies the copyrighted work as “Pre-2012 NGINX Software,” which it defines as “NGINX Plus and Open Source NGINX, in both source code and executable form, conceived and/or developed before the end of 2011, when Sysoev left the employ of Rambler.” Although Lynwood argues “the term ‘Pre-2012 NGINX Software’ is precise because it identifies the specific iterations of Open Source NGINX and NGINX Plus that Sysoev wrote as works for hire under Russian law while he was employed by Rambler,” no such specific iterations can be identified from the paragraphs Lynwood cites in support of such allegation. Moreover, Lynwood’s definition fails for the additional reason that it includes software “conceived” but not developed at Rambler. See L[aws] v. Sony Music Ent., Inc., 448 F.3d 1134, 1140 (9th Cir. 2006) (noting “copyright protects original works of authorship fixed in any tangible medium of expression”).

    We’ll stop here. Typically, the copyright registration would define the scope of the claim. The court didn’t mention it, but here there was no copyright registration. Lynwood alleged in the complaint that the first publication was in Russia, and therefore no registration was required.

    Nevertheless, this description sounds pretty specific to me. Surely the pre-2011 version of at least the open source portion of the the software code exists—that’s kind of how open source works. This seems specific enough that the exact scope of the copyrighted work should be fleshed out in discovery, not be the basis for dismissal.

    Continuing, Lynwood also maintained in the amended complaint the sin of claiming copyright over software that was “conceived,” but I guess not written, by Sysoev, another reason the court gave that the description of the infringed work wasn’t specific enough.

    Lynwood also still failed to describe more clearly who was doing the infringing. It lumped together all the individuals, all the NGINX entities, and F5 (but not the investors), claiming that they collectively introduced 20 or more releases of NGINX Plus. But “such collective allegations do not suffice to plead a viable infringement claim.” The court therefore dismissed the copyright infringement count, and the tortious interference with contract claim was also dismissed under Rule 12(b)(6).

    Because Lynwood had been warned of the possibility in the earlier dismissal, this time the repleaded, but insufficient, claims were dismissed with prejudice. The parties voluntarily dismissed the remaining claims with prejudice. Lynwood has appealed the court’s dismissal.1

    But I’m still confused by the copyright infringement claim. The works allegedly infringed were “NGINX Plus and Open Source NGINX.” Taking NGINX Plus first, in all the tellings of the story the software was proprietary. I could get on board with the theory that Rambler owned it as a work made for hire, Lynwood acquired it from Rambler (although they didn’t know they did until the whistleblower told them about the secret development) and that NGINX’s exploitation is a copyright infringement of the Rambler copyright.

    But what I can’t get onboard with is that NGINX was infringing the copyright by using the Open Source NGINX. In the Amended Complaint, Lynwood states Rambler “made a business decision to permit Sysoev to continue releasing Open Source NGINX … because such releases highlighted the technical achievements of Rambler and its employee Sysoev, which inured to Rambler’s benefit for attracting top software programmers.” This seems simple enough—even if Rambler didn’t make the initial decision to license the software under an open source license, it ratified the decision by continuing to allow the license to be used. An open source license is perpetual, so thereafter NGINX couldn’t infringe the copyright in it, assuming that they met the very trivial license compliance requirements of the FreeBSD license.

    In the amended complaint Lynwood alleges that Rambler was bamboozled into releasing NGINX as open source software by defendant Konovalov, who was the chief technology officer. Lynwood’s narrative arc is that Lynwood purchased Rambler in 2019 and shortly thereafter, through a whistleblower, learned of a scheme operating from 2004 through 2011 thus: Sysoev developed the NGINX web server software; he and a whole crew of schemers recognized that it had commercial value; they hoodwinked Rambler into making it available under an open source license; they developed proprietary additions surreptitiously while still employed by Rambler; they departed Rambler and, although they disclosed that they were starting up a business to support the open source NGINX software, they didn’t disclose enough of what they were up to for Rambler to see that it was no good; and payday! NGINX therefore was wrongfully licensed under an open source license and Lynwood is entitled to $750 million AND the defendants have to destroy all copies of the software, including Open Source NGINX. Take note that all of this is a windfall for Lynwood, which presumably acquired Rambler (or the software, it’s not entirely clear) knowing that the NGINX software, even though owned by Rambler, was under a permissive open source license and therefore of little book value. But now they think that they can wind it all back to 2004 before the first release of NGINX.

    Maybe it is all true, but I have a hard time believing Rambler was so naive, and could be so fooled by one person, that the open source licensing, upon which the whole case largely rests, was the result of some unlawful act. Instead, it’s more plausible that Rambler was a search engine provider and didn’t care so much about the NGINX software because that’s not how they directly generated revenue. Rambler recognized that releasing the software under an open source license had the indirect benefit of attracting developers. That’s a very solid, knowing rationale for licensing the software under the FreeBSD license. Lynwood would have to get past that before it could unwind the open source license.

    And the final decision in the canon is an award of attorneys’ fees to the defendants on the copyright infringement claim. Where the defendant is the prevailing party, no registration is required. Latin Am. Music Co. v. Am. Soc. of Composers, Authors and Pubs. (ASCAP), 642 F.3d 87, 90 (1st Cir. 2011).

    Here, the Court finds Lynwood’s copyright claim, although not frivolous, was objectively unreasonable. First, Lynwood, from the outset, should have been aware of the grounds on which its copyright claim was dismissed, namely, that asserting a copyright infringement claim without properly identifying the copyrighted work at issue, or which defendant infringed which particular copyright, rendered its chances of success slim to none. … In particular, Lynwood, despite having such deficiencies pointed out and being afforded leave to amend, continued to include in its definition of the copyrighted work software “conceived” but not developed, see 17 U.S.C. § 102(b) (stating “[i]n no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery”), failed to “identify which defendants allegedly infringed its copyright” (see March 25, 2021, Order, 30:26-31:1), and, most importantly, was unable to “identify with specificity which work or works were copied.”

    Deterrence was necessary:

    Even in the absence of frivolous or unreasonable claims, a successful defense can further the interests of the copyright act by defeating an over-aggressive copyright prosecution exemplified by pursuit of grossly overbroad monetary and injunctive relief. … Here, defendants argue, Lynwood pursued exactly that, specifically, by seeking $750 million and destruction of all copies of NGINX software. In support thereof, defendants have submitted evidence, to which Lynwood makes no objection, that any loss Rambler arguably incurred was no more than around $810,000, a sum constituting less than one percent of the $750 million in damages Lynwood claimed, and, as defendants also point out, Lynwood filed this lawsuit only after F5–a publicly traded U.S. company–acquired NGINX for $670 million in 2019.”

    Additionally, defendants contend, Lynwood … pursued this lawsuit with the stated aim of effectuating the destruction of the NGINX software, which destruction, defendants assert, would have significantly impeded creative works and potentially disrupted a significant portion of web serving across the worldwide web, given that NGINX, which NGINX Inc. has operated … for the last decade, is the worlds’ most widely used web server, on which millions of people and businesses rely.

    Lynwood does not dispute the above contentions.

    (Quotation marks omitted.) The court therefore awarded attorneys’ fees in an amount to be determined. While I disagree that the copyright claim was objectively unreasonable for the reasons the court gave, I agree it was pursued with the aim of destroying the NGINX software. I’m not sure that copyright damages was the right hook for damages, but probably the easiest.

    So who’s the one stealing?

    Lynwood Investments CY Limited v. Konovalov, Civ. No. 5:20-cv-03778-LHK, Order Granting Motions to Dismiss With Leave to Amend (N.D. Cal. Mar. 25, 2021)

    Lynwood Investments CY Limited v. Konovalov, Civ. No. 5:20-cv-03778-LHK, Order Granting Motion to Dismiss With Leave to Amend (N.D. Cal. Mar. 30, 2021)

    Lynwood Investments CY Limited v. Konovalov, Civ. No. 3:20-cv-03778-MMC, Order Granting Motion to Dismiss (N.D. Cal. Aug. 16, 2022)

    Lynwood Investments CY Limited v. Konovalov, Civ. No. 3:20-cv-03778-MMC, Amended Order Finding Defendants Entitled to Attorneys’ Fees; Affording Defendants Leave to Supplement Motions as to Amount; Setting Briefing Schedule (N.D. Cal. Dec. 20, 2022)

    Lynwood Investments CY Limited v. Konovalov, App. No. Civ. No. 3:20-cv-03778-MMC , Notice of Appeal (N.D. Cal. Sept. 14, 2022)

    Creative Commons License
    This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.


    1. The trademark infringement claim seems like the most viable to me, but it’s been conceded. Lynwood alleged trademark infringement in its complaints but didn’t assert those claims when the court limited their claims. Lynwood also filed petitions to cancel the NGINX trademark registrations after the original complaint was filed, but withdrew the petitions after the federal complaint was dismissed. Maybe Lynwood believes that there wasn’t business value it depriving NGINX of the ability to call the software NGINX. 
  • Even an Exclusive Licensee Can’t Sue the Trademark Owner for Infringement

    This might strike you as odd. An exclusive licensee of a copyright can sue the copyright owner for infringement, Essex Music, Inc. v. ABKCO Music & Recs., Inc., 743 F. Supp. 237, 241 (S.D.N.Y. 1990), and a exclusive licensee of a patent can sue the patent owner for infringement, Ortho Pharm. Corp. v. Genetics Inst., Inc., 52 F.3d 1026, 1030 (Fed. Cir. 1995). Not so for trademarks.

    It’s a complicated licensing story that I’ll simplify. Since 2009 plaintiff Camelot SI, LLC has been a licensee of the brand SHARPER IMAGE. In 2013, Camelot and Icon NY Holdings LLC, the then-trademark owner, entered into a Manufacturing License Agreement granting Camelot a non-exclusive license to manufacture and/or source SHARPER IMAGE-branded products. At some point—the decision isn’t clear about the timing and whether it was before or after the Manufacturing License Agreement—Icon filed for bankruptcy. In 2014, the parties entered into a Website and Catalog Rights Purchase Agreement. While the Manufacturing Agreement was for the manufacture and sale of goods with the SHARPER IMAGE brand, the Purchase Agreement was for the branding of e-commerce and catalog services.

    As part of the Purchase Agreement, Camelot was granted an exclusive license to use the mark for the e-commerce channel of trade and Icon could sell SHARPER IMAGE-branded goods only through non-SHARPER IMAGE branded online retail sales or marketing platforms. As summarized by the court, “For more than a decade, Camelot has been the only party authorized to sell SHARPER IMAGE branded products through SHARPER IMAGE branded online retail sales or marketing platforms. Camelot has operated the SHARPER IMAGE branded website since 2009 and has owned it since 2014.”

    In 2016 defendant ThreeSixty Brands Group LLC acquired from Icon the SHARPER IMAGE trademarks, the Manufacturing Agreement, and the Purchase Agreement. Nevertheless, as described by the court on a motion to dismiss (that is, taking the plaintiff’s facts as true) “[d]espite ThreeSixty’s agreement to offer SHARPER IMAGE branded products only through non-SHARPER IMAGE branded e-commerce services, and in direct violation of Camelot’s exclusive license of the Trademarks in connection with e-commerce services, ThreeSixty is operating several SHARPER IMAGE branded online retail sales and marketing platforms, and has launched a second brand known as ‘Sharper Tomorrow.’”

    One of ThreeSixty’s competing stores was a microsite on Amazon, which claimed that “Shaper Tomorrow” was the next iteration of Sharper Image, and directed all traffic from it to its own Amazon store. ThreeSixty also filed complaints with Amazon about what it alleged were counterfeit products on Camelot’s Amazon site, thus causing Amazon to remove the items and mark Camelot’s account as “under review.”

    Camelot sued ThreeSixty for unfair competition under Section 43(a) of the Lanham Act and New York state law, as well as some supplemental state law claims. But, the unfair competition claims will not lie. Here is the problem:

    To state a claim under the Lanham Act for trademark infringement, unfair competition, and false designation of origin, a plaintiff must establish that: (1) [plaintiff] owns a valid mark entitled to protection under the Lanham Act; (2) defendant used the protected mark in commerce, without plaintiff’s consent; and (3) defendant’s use of that mark is likely to cause consumers confusion as to the origin or sponsorship of the defendant’s goods.”

    Camelot couldn’t get past (1). The court recharacterized the requirement as having “a valid commercial interest” in the trademarks. Relying on a similar fact pattern decided by the Court of Appeals for the Second Circuit over two television series called “Mutant X” and “X-Men,”

    The [X-Men] district court dismissed the unfair competition claim premised upon false designation of origin. It reasoned that Marvel, the owner of the marks, is the “origin” of the series (and the film) within the meaning of trademark law (i.e., the source of the goodwill inhering in the trademarks that Marvel licensed to Fox). Because Fox did not own the marks, it could “claim no impairment of goodwill in the X-Men property associated with the X-Men trademark.”

    Camelot, like Fox, cannot claim any impairment in the goodwill associated with the Trademarks. Indeed, the plain language of the Agreements shows the opposite: that all goodwill inures to ThreeSixty.

    A theory of “unfair competition” just does not stretch this far:

    Moreover, contrary to Camelot’s interpretation, courts have made clear that “under § 43 of the Act, there is no specific federal cause of action for unfair competition.” “Instead unfair competition under the Lanham Act is a category of claims consisting primarily of causes of action for false designation of origin and false advertising.” Accordingly, Camelot cannot escape the Lanham Act’s ownership interest requirement.

    Additionally, case law makes clear that as owner of the Trademarks, ThreeSixty is the sole “source” that consumers associate with Camelot or MerchSource’s use of the marks. Thus, there simply can be no confusion as to their origin.

    All is not lost for Camelot, though. The federal unfair competition count was the only one supporting federal court jurisdiction because Camelot had not pleaded diversity jurisdiction, but the parties later agreed there was diversity jurisdiction. Camelot had also pleaded breach of contract, tortious inteference with prospective economic advantage (for lost customers), tortious interference with contract (the Amazon contract), and unjust enrichment. The tortious interference claims were also dismissed, but with leave to amend.

    Camelot SI, LLC v. ThreeSixty Brands Group LLC, No. 21 Civ. 8232 (ER) (S.D.N.Y. Sept. 30, 2022).
    This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.

  • Who ChatGPT Thinks Owns a Trademark

    I asked ChatGPT to “please write a blog post about how to decide who owns a trademark.” Here goes:

    Deciding who owns a trademark can be a complex legal issue, and it’s important to consult with an experienced trademark attorney to ensure that your rights are protected. In general, the owner of a trademark is the person or entity who uses the trademark to identify the source of their goods or services.

    Trademark ownership is typically established through use, rather than registration. This means that if you are the first person to use a particular trademark in connection with your goods or services, you are likely the owner of that trademark. However, registering your trademark with the appropriate government agency can provide additional legal protection and make it easier to enforce your rights in the event of a dispute.

    There are a few key factors to consider when determining who owns a trademark. First, it’s important to consider who was the first to use the trademark in connection with their goods or services. The person or entity who was the first to use the trademark is typically considered the owner.

    Second, it’s important to consider the likelihood of confusion. If two parties are using the same or similar trademarks for similar goods or services, there may be a likelihood of confusion among consumers. In this case, the party who was the first to use the trademark is typically considered the owner.

    Third, it’s important to consider the strength of the trademark. A strong trademark is one that is distinctive and easily recognizable. A weak trademark, on the other hand, is one that is generic or descriptive and not easily distinguishable from other trademarks. If two parties are using the same or similar trademarks, the party with the stronger trademark is typically considered the owner.

    In summary, determining who owns a trademark can be a complex legal issue. The key factors to consider are who was the first to use the trademark, the likelihood of confusion, and the strength of the trademark. If you have questions about trademark ownership, it’s best to consult with an experienced trademark attorney.

    Creative Commons License
    This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.

  • Estopped By A Wrong Decision

    The Federal Circuit issued two companion decisions involving the same patent license arrangement, one that affected at least three district court decisions. Uniloc USA, Inc., Uniloc Luxembourg S.A., and Uniloc 2017 LLC, which I’ll just refer to as “Uniloc,” sued Apple, Motorola Mobility and Google for patent infringement. However, Uniloc had received funding from Fortress Credit Co. LLC and as part of the funding agreement Uniloc gave Fortress a patent license:

    Subject to the terms and conditions herein and in the Purchase Agreement, Licensor hereby grants to Licensee a non-exclusive, transferrable, sub-licensable, divisible, irrevocable, fully paid-up, royalty-free, and worldwide license to the Licensed Patents, including, but not limited to, the rights to make, have made, market, use, sell, offer for sale, import, export and distribute the inventions disclosed in the Licensed Patents and otherwise exploit the Licensed Patents in any lawful manner in Licensee’s sole and absolute discretion solely for the benefit of the Secured Parties (“Patent License”), provided that Licensee shall only use the Patent License following an Event of Default.

    Uniloc defaulted, although Fortress never claimed default and never exercised the license.

    The district court in the Apple case (Alsup, J.) held that, because of the default, Uniloc did not have standing for a patent infringement claim. As described by the Court of Appeals:

    The district court held that the Unilocs lacked standing to sue for infringement because Fortress had the theoretical right to sublicense the asserted patent to the alleged infringer, and the Unilocs, as a result, lacked the exclusionary right necessary to confer standing.

    Motorola Mobility and Google argued in their respective cases that the Apple decision collaterally estopped Uniloc from claiming that it had standing. The Federal Circuit agreed, and none of Uniloc’s theories why they shouldn’t be estopped—collateral estoppel wasn’t timely raised; Motorola Mobility waived the argument by characterizing the Apple decision as “not binding on this Court”; and Uniloc didn’t have an incentive to litigate the Apple judgment because it had a settlement opportunity with Apple—didn’t stick. The Motorola Mobility case was dismissed altogether but the Google case was not, to be discussed in a moment.

    Here’s the thing—the decision in the Apple lawsuit about Uniloc’s standing was just wrong. The appeals court decision, written by Judge Dyk, extended some grace to the district court:

    We recognize there is considerable force to Uniloc’s argument that, even if Fortress had been granted a license and an unfettered right to sublicense, Uniloc would have Article III standing. Patent owners and licensees do not have identical patent rights, and patent owners arguably do not lack standing simply because they granted a license that gave another party the right to sublicense the patent to an alleged infringer. But, we need not resolve the question of whether a patent owner who granted a right to sublicense lacks standing here.

    But Judge Lourie, in “additional views,” wanted to make sure that there was no doubt about it:

    I join the opinion of the court in all respects, except that I believe the paragraph at page 8, lines 11–23, beginning “[w]e recognize there is considerable force to Uniloc’s argument that, even if Fortress had been granted a license and an unfettered right to sublicense, Uniloc would have Article III standing,” is a regrettable understatement. In my view, there is more than considerable force to the argument; it is clear that Uniloc still had the right to sue unlicensed infringers after it granted the license.

    We normally do not opine on issues that are not necessary to decide a case, and our panel soundly affirms the district court on the ground of estoppel. But here, I believe the district court so misconstrued the license issue that something further needs to be said about it.

    The district court, respectfully, incorrectly dealt with this issue as one of determining what is an exclusive license, citing cases on whether an exclusive licensee alone has standing to sue without joining the owner of a patent. That is not the case before us. …

    The grant of a non-exclusive license with the right to sublicense, as here, gives the licensee the right to sublicense others. But the patentee still retains the right to sue unlicensed infringers. A non-exclusive license only grants a licensee freedom from suit; it does not divest the licensor of its right to sue or license other parties, or to practice the patent itself. See 14B Chisum on Patents 6240 (2022) (“A license may amount to no more than a covenant by the patentee not to sue the licensee for making, using or selling the patented invention, the patentee reserving the right to grant others the same right.”) …

    It is true that the licensee could preempt such a suit by granting a sublicense, immunizing the purported infringer. But that is a far cry from holding that the patent owner, simply by having granted a non-exclusive license with the right to sublicense, loses the power to sue an unlicensed infringer.

    Thus, while agreeing in full that Uniloc loses its appeal by being estopped from suing Motorola because it settled its suit with Apple rather than appealing it, it is not because it lost the power to sue by granting Fortress a non-exclusive license with the right to sublicense.

    But one sentence was all it took to dispose of Uniloc’s argument that the wrongness was a reason not to find there was collateral estoppel:

    [T]o the extent Uniloc argues that collateral estoppel should not apply because the decision in the Apple case is incorrect, this is simply not a proper basis to deny collateral estoppel. Generally, collateral estoppel cannot be denied because the decision was incorrect. See 18A Wright & Miller § 4465.2, at 750–51 (“[P]reclusion cannot be defeated simply by arguing that the prior judgment was wrong.”).

    However, Google was sued after Uniloc and Fortress entered into a Termination Agreement, terminating the problematic patent license Uniloc had granted. Google argued that the Termination Agreement didn’t terminate the license to Fortress because the license was irrevocable. The court of appeals shot that down quickly:

    On its face the License Agreement describes the license as “irrevocable.” But this does not suggest the license is irrevocable by mutual agreement. The term “irrevocable” in its context clearly refers to the license’s being “irrevocable” by the licensor.

    Under the relevant case law, the term “irrevocable” does not suggest that the license could not be eliminated by mutual agreement. Cases construing the term “irrevocable” agree that the term means only that the irrevocable thing cannot be unilaterally revoked by the party that granted the benefit. See In re Zimmerman (Cohen), 236 N.Y. 15, 139 N.E. 764, 766 (1923) (“The word ‘irrevocable,’ here used, means that the contract to arbitrate cannot be revoked at the will of one party to it …. It does not mean that the agreement to arbitrate is irrevocable by the mutual agreement or consent of the parties.”); Silverstein v. United Cerebral Palsy Ass’n, 17 A.D.2d 160, 232 N.Y.S.2d 968, 970-71 (1962) (“[L]ike any contract, the irrevocable offer may only be modified, released or rescinded by agreement of the parties. It cannot be unilaterally withdrawn, revoked or rescinded by the offeror.” (citations omitted)); Barclays Bank D.C.O. v. Mercantile Nat’l Bank, 481 F.2d 1224, 1238 (5th Cir. 1973) (noting that the grantor of an irrevocable letter of credit “could not modify the irrevocable credit without [the grantee’s] consent”); In re Huntington, ADV 11-4015, 2013 WL 6098405, at 8 (B.A.P. 9th Cir. Oct. 29, 2013) (noting that an irrevocable assignment cannot be revoked by one party, but can be revoked by mutual consent of all parties); *Carbonneau v. Lague, Inc., 134 Vt. 175, 352 A.2d 694, 696 (1976) (concluding that an irrevocable license was terminated by a voluntary agreement between all parties).

    Google even conceded that its theory didn’t fly but floated some other arguments, none of which worked either. It claimed that the survival clause in the original patent license, which said that “[a]ny rights … which by their nature survive and continue after any expiration or termination of this Agreement will survive and continue and will bind the Parties …” meant the license survived. This was a no-go; while acknowledging that there might be some cases where a license grant might survive, this wasn’t one of them. Granted sublicenses also survived termination of the license agreement, but this exception showed that not all licenses did survive. Google has to litigate the patent suit.

    Perhaps Uniloc could have avoided this outcome by asking that the decision in the Apple case be vacated, but it did not. Collateral estoppel, for a wrong decision, still shut down their case.

    Uniloc USA, Inc. v. Motorola Mobility, No. 2021-1555, 2012-1795 (Fed. Cir. Nov. 4, 2022)
    Uniloc 2017 LLC v. Google LLC, No. 2021-1498, and a bunch more (Fed. Cir. Nov. 4, 2022)
    Uniloc USA, Inc. v. Apple, Inc., No. C 18-00358 WHA (N.D. Cal. Dec. 4, 2020)
    Creative Commons License
    This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.