Recent decisions from the U.S. Court of Appeals for the Federal Circuit regarding damages available in design patent cases highlight the court’s divergence from its damages jurisprudence in the utility patent context. — Charlene Morrow and Sapna Mehta
The TPP’s Intellectual Property Chapter (Chapter 18) establishes complex, detailed provisions pertaining to patents, copyrights, and trademarks, as well as rules regarding procedure, enforcement, and remedies. — Ciara N. McHale and Andrew P. Bridges
Patent Office Issues New AIA Rules — Christopher P. King
The Defend Trade Secrets Act of 2016 Creates Federal Jurisdiction for Trade Secret Litigation — Sebastian Kaplan and Patrick Premo
Introducing the EUIPO – Understanding Recent Changes to EU Trademark Law — Emily Gische
A Report on Notice and Takedown in Everyday Practice — Kathleen Lu and Jennifer Stanley
By Charlene Morrow and Sapna Mehta
Recent decisions from the U.S. Court of Appeals for the Federal Circuit regarding damages available in design patent cases highlight the court’s divergence from its damages jurisprudence in the utility patent context – specifically, the lack of an apportionment requirement between patented and unpatented portions of an infringing product. While this may make design patents increasingly desirable, the Supreme Court’s decision to review the issue now raises the possibility that the discrepancy will be resolved.
A patent holder who successfully sues for infringement is entitled to compensatory damages not less than “a reasonable royalty for the use made of the invention by the infringer.” 35 U.S.C. § 284. An additional statutory provision specific to design patents provides for damages in the amount of the infringer’s “total profit” on the infringed “article of manufacture,” not less than $250. 35 U.S.C. § 289. Section 289 makes clear that while it does not preempt the patent holder’s other statutory remedies, the patent holder cannot “twice recover the profit made from the infringement.” Accordingly, a party who has secured a judgment for infringement of its design patent has two options for recovering damages: (1) seek its own lost profits or a reasonable royalty under § 284, or (2) seek the infringer’s total profits (or the $250 statutory floor) under § 289.The choice will almost always skew towards the latter option as a result of the different approaches the Federal Circuit has adopted to calculating lost profits (or a reasonable royalty) under § 284 and calculating profits gained from infringing products under § 289.
Under Federal Circuit precedent, a party seeking damages for design patent infringement under § 289 may obtain the infringer’s profits from the whole product that contains the infringing design. The court analyzed this issue in Nike, Inc. v. Wal-Mart Stores, Inc., a case concerning a patent for the design of an athletic shoe. Affirming an award of all profits from sales of the infringing shoe, the court reviewed the history of design patent remedies and the early cases that prompted design patent-specific legislation in the 1800s.The court determined that the design patent-specific legislation was enacted to remove “the need to apportion the infringer’s profits between the patented design and the article bearing the design.”
The Federal Circuit recently reiterated its holding and rationale in two additional decisions, bringing renewed attention to the issue of damages for design patents. Most notable is the court's decision in the closely watched Apple Inc. v. Samsung Electronics Co. (2015) smartphone case. In that case, Apple successfully secured a damages award for infringement of three of its design patents. Each patent claims an element of the design of Apple’s iPhone – specifically, they respectively claim: (1) the design of the “graphical user interface,” or icons on the display screen; (2) the flat-contoured front face of the phone and “bezel” around the face; and (3) the “flat, continuous and edge-to-edge transparent black front surface.” Despite the discrete aspects of the iPhone encompassed by each design patent, the jury awarded damages for Samsung’s entire profits from the smartphones found to infringe.
Samsung was unsuccessful in its post-judgment motion challenging this award, with the district court finding its position “clearly foreclosed” by Nike and Federal Circuit precedent (see the Apple Inc. v. Samsung Electronics Co. (2013) decision from the Northern District of California). Samsung appealed the damages award, arguing (1) that “basic causation principles” require that damages be limited to that portion of its profits actually attributable to the design patent infringement, and (2) that the awarded profits should be limited to the infringing “article of manufacture” rather than the entire product. The Federal Circuit rejected both arguments as imposing an impermissible requirement that the profits be apportioned. In its decision, the court cited Nike in support of the proposition that Congress had rejected such a requirement in enacting design patent-specific legislation, and further found that § 289 “explicitly authorizes the award of total profit from the article of manufacture bearing the patented design.”
Similarly, in Nordock, Inc. v. Systems Inc., the Federal Circuit vacated a damages award calculated by the jury as a reasonable royalty, reiterating its previous holding that “apportioning profits in the context of design patent infringement is not appropriate.” Clarifying the interplay between § 289 and § 284 in design patent cases, the court advised that “[o]nly where § 289 damages are not sought, or are less than would be recoverable under § 284, is an award of § 284 damages appropriate.”
The design patent at issue in the Nordock case claims the “ornamental design of a lip and hinge plate” for a dock leveler (a large mechanical device used to bridge loading dock surfaces and the surfaces of truck beds). The design of the lip and hinge plate consists of different-shaped lugs that are secured to both the “lip” (the portion that extends) and the “hinge” (the portion that attaches to a frame). In rejecting apportionment, the court emphasized that there was no evidence that the infringing party sold the lip and hinge plate as a stand-alone unit, separately from the dock leveler. Interestingly — and perhaps telling of future trends with respect to the prevalence of design patent cases — the court noted that the patent holder had applied for utility patent protection for the “lug hinge design,” but that its application had been rejected over the prior art.
In contrast to the damages analysis under § 289 — which some have criticized as granting windfalls to plaintiffs — the Federal Circuit has established a robust apportionment requirement between patented and unpatented features when calculating damages under § 284. It articulated the requisite analysis through Lucent Technologies, Inc. v. Gateway, Inc., and its progeny. Since Lucent, the court has generally required apportionment of the patented features in calculating damages, first through identification of the smallest salable unit containing the infringing invention, and then through estimation of the value or contribution of the patented invention to that salable unit versus the value or contribution of the unpatented portions.
As an example, the court in VirnetX, Inc. v. Cisco Systems, Inc., vacated a damages award for infringement of a utility patent by the “FaceTime” and “VPN On Demand” features contained in some of Apple's products, including the iPhone. The Federal Circuit found that the district court had incorrectly instructed the jury by suggesting that the smallest salable unit should be used as the base for damages, finding that this approach ignores “a fundamental concern” about “using a base that misleadingly suggests an inappropriate range.” The court emphasized that a smallest salable unit that is a “multi-component product containing several non-infringing features with no relation to the patented feature” must be further analyzed to delineate between the value of that unit attributable to the patented portion versus the other portions. Thus, the court pointed out, the calculation of damages based on the cost of an iPhone as the “smallest salable unit” failed to separate the value between “software from hardware, much less to separate the [accused] software from other valuable software components.”
This serves as a stark contrast to the treatment of the infringing smartphones in the Apple case described above. Notably, if the court in Apple had construed “article of manufacture bearing the patented design” to include only the portion of the product encompassing the patented design — in the same manner as its smallest salable unit analysis in VirnetX — the damages award would have been quite different. For example, the “article of manufacture” encompassing the claimed design of the phone’s front face and bezel could have been the device housing of the Samsung smartphones, rather than the entire Samsung smartphones found to be infringing. The Apple case thus serves as an effective example of the disparities between damages under § 284 and § 289, since the design patents at issue in that case cover particular portions of a multi-component product.
Capitalizing on that fact, Samsung filed a petition for certiorari, asking the Supreme Court to weigh in on the Federal Circuit’s application of § 289, and to consider whether an award of profits for infringement of a design patent should be limited to the profits attributable to the patented component. Samsung argued, in its request for review, that § 289 applies only to the patented portion of a product, and that in any event, the statute’s requirement that the patent holder “not twice recover” — as well as principles underlying tort liability and equitable remedies — requires a showing of causation between the infringement and the damages.
On March 21, the Supreme Court granted certiorari to consider: “[w]here a design patent is applied to only a component of a product, should an award of infringer’s profits be limited to those profits attributable to the component?” Accordingly, the Supreme Court now has the opportunity to resolve the Federal Circuit’s divergent case law regarding damages awarded under § 284 and those awarded under § 289.
By Ciara N. McHale and Andrew P. Bridges
In November 2015, the Office of the U.S. Trade Representative publicized for the first time the final, full text of the Trans-Pacific Partnership Agreement. In February of this year, 12 member nations signed the TPP. The agreement is controversial in both substance and form, as its negotiation took place in unprecedented secrecy among its participants. In addition to the U.S., the member nations are Australia, Brunei, Canada, Chile, Japan, Malaysia, Mexico, New Zealand, Peru, Singapore, and Vietnam. The TPP now awaits ratification.
The TPP’s Intellectual Property Chapter (Chapter 18) establishes complex, detailed provisions pertaining to patents, copyrights, and trademarks, as well as rules regarding procedure, enforcement, and remedies. Chapter 18’s copyright provisions have drawn both pointed criticism and strong support from commentators across the copyright community. The U.S. Trade Representative claims the IP chapter provides strong and balanced protections that align with current U.S. intellectual property law. Some commentators agree with this description, pointing out that the U.S. itself has robust copyright protection and a great interest in preserving strong IP protection internationally, given the economic importance of industries that rely upon strong copyright protection, such as motion pictures, electronic games, and computer software. On the other hand, many critics of the IP chapter take the view that the provisions are not balanced, providing significant rights to IP owners without affording sufficient protections or consideration to users or the public. Commentators have also voiced concern that the TPP cements into place on an international level provisions that are not just consistent with, but expand, U.S. copyright protections. The result is to effectively prevent change and innovation in intellectual property law in the U.S. and other member nations, including prohibitions upon Congress’s amending certain provisions of copyright law.
Indeed, while many of the IP chapter provisions are consistent with strong copyright provisions already in place in the U.S., a thorough examination of its specific copyright provisions reveals that its treatment of copyright law is emphatically anti-innovation in its locking in of current, arguably outdated rules, precluding innovations in copyright law that may require change in the form of individual departures by member nations from established legal norms. Following is a snapshot of some of the TPP’s more notable and controversial copyright provisions.
Article 18.2 articulates the IP chapter’s objectives:
The protection and enforcement of intellectual property rights should contribute to the promotion of technological innovation and to the transfer and dissemination of technology, to the mutual advantage of producers and users of technological knowledge and in a manner conducive to social and economic welfare, and to a balance of rights and obligations.
This provision survives from an earlier, more extensive list of objectives that did not make it into the final TPP. More specifically, earlier leaked drafts of this section included additional objectives relating to balancing the interests of rights holders and users, to promoting and preserving the public domain, and to ensuring that IP rights do not themselves create barriers to legitimate trade. Both the U.S. and Japan opposed inclusion of these objectives, and the final version includes only the provision related to the protection and enforcement of IP rights, with no mention of maintaining balance between the interests of rights holders and users. While the objectives provision does not establish affirmative obligations on the part of member nations, it does set the tone for the obligations that follow.
When the term of protection is calculated on the basis of the life of a natural person, Article 18.63 requires parties to implement a mandatory minimum copyright term of the life of the author plus 70 years. While this provision harmonizes duration of copyright across the member nations, it also locks in the current term in U.S. law and will increase the term in a number of the TPP signatory countries, including Canada. Those in favor of the life-plus-70-years term argue that it follows a trend toward longer terms and accounts for authors’ longer life spans and the need to incentivize investment in older works. But longer life spans, if relevant, themselves already automatically increase a copyright term because the term is based on the author’s length of life – so the use of longer life spans as justification for additional years of protection following the author’s death is therefore backwards. In addition, many commentators, including the former U.S. Register of Copyrights, have called the 1998 Sonny Bono Copyright Term Extension Act, which lengthened the U.S. term to its current standard, a mistake. Critics argue that such term extensions do not incentivize creativity and harm the public by keeping materials out of the public domain for longer periods of time and by exacerbating the problem of “orphan works.” Extending the term to 70 years will impose monetary costs on new authors wanting to make derivative works (because they will need to obtain licenses to do so) or will discourage them from making derivative works for 20 more years. Extending the term will also impose restrictions and costs on the public wanting access to works that would otherwise enter the public domain sooner.
The TPP’s fair use framework consists of two articles, one familiar and one new. Article 18.65 adheres to the well-known three-step test for exceptions and limitations in other multinational treaties. Article 18.66 significantly departs from other multinational treaty provisions. That article provides:
Each Party shall endeavor to achieve an appropriate balance in its copyright and related rights systems, inter alia by means of limitations or exceptions that are consistent with Article 18.65 (Limitations and Exceptions), including those for the digital environment, giving due consideration to legitimate purposes such as, but not limited to: criticism; comment; news reporting/ teaching, scholarship, research, and other similar purposes; and facilitating access to published works for persons who are blind, visually impaired, or otherwise print disabled.
At first glance, this novel provision is a positive step toward encouraging parties to the TPP to adopt a flexible approach to fair use analysis, as it expounds upon the traditional three-step approach to affirmatively require member nations to consider balance between rights holders and users who access copyrighted works and engage with them in new and creative ways. Thus, although the need for balance did not survive in the TPP’s IP objectives article, it appears here. But in many ways it does not go far enough. More specifically, while Article 18.66 includes helpful language and lists non-exhaustive examples of fair use from § 107 of the Copyright Act, it does not incorporate the fair use factors that govern in the U.S. This could lead to a pigeonhole rather than flexible approach.
Article 18.57 sets forth definitions for the IP chapter’s section on copyright as those provisions apply “with respect to performers and producers of phonograms.” The definitions mirror the definitions in Article 2 of the WIPO Performances and Phonograms Treaty and therefore illustrate a focus on sound recordings. In many instances, including the examples below, these definitions appear to differ from definitions in § 101 of the Copyright Act and to broaden copyright holders’ rights under U.S. law:
The preceding list is illustrative and suggests a broadening of concepts and an increase in copyright power at the international level that will likely, if ratified, reverberate to affect the development of U.S. law. This probability is underscored by the actions of copyright holders who have bypassed congressional action and sought judicial expansions of their rights by arguing that later treaties compel reinterpretation of previously existing statutes.
The TPP’s anti-circumvention measures in Article 18.68, governing treatment of technological protection measures or “digital locks,” are also controversial. The article's anti-circumvention provisions resemble those of the U.S.’s Digital Millennium Copyright Act, and suggest that liability could arise whether a technological measure serves to prevent infringement or more broadly restricts non-infringing use of or access to works. Proponents of strong technical protections again highlight the need for protections to preclude misuse of access to locked services and underlying content. But critics point out that the DMCA has stifled a number of legitimate activities and products in the U.S., and they warn that if countries apply Article 18.68 as broadly as U.S. courts have applied the DMCA, this could result in the article's misuse by businesses for the purpose of evading competition, and could otherwise stifle innovation, scientific research, and other beneficial uses of locked materials by the public.
The provisions discussed above and others in the copyright section of the IP chapter exist against a backdrop of remedies that include harsh criminal sanctions for copyright infringement. Most notably, the TPP allows for imposition of criminal penalties even in the absence of commercial motivation on the part of an infringer. Critics of increased criminalization of copyright infringement point out that this provision could lead to the incarceration or fining of users for offenses such as casual file-sharing.
Although the member nations have signed the TPP, the pact is not yet effective, and the path to implementation is not guaranteed. The TPP will take effect (1) 60 days after all 12 original signatories ratify it, or (2) in April 2018 if at least six of the 12 countries, comprising 85 percent of the combined gross domestic product of the original signatories, have ratified the agreement – which would require ratification by both the U.S. and Japan. The TPP faces a significant battle for ratification in the middle of a chaotic election year in the U.S., where critics on multiple fronts include the leading presidential candidates.
The U.S. Patent and Trademark Office recently issued an updated set of rules affecting trial practice before the Patent Trial and Appeal Board. In large part, the rules, which went into effect on May 2, 2016, were implemented as proposed on August 20, 2015. In particular, they modify the prior rules governing inter partes review, post-grant review, the transitional program for covered business method patents, and derivation proceedings that implemented provisions of the America Invents Act providing for trials before the Office.
The updated rules affect several different aspects of PTAB trial practice including sanctions, claim construction standards, standards for length of submissions, and — perhaps most notably — the types of evidence usable by the patent owner.
The updated rules amend 37 CFR 42.11 to include a Rule 11-type certification for papers filed with the Board including a provision for sanctions for non-compliance, such as for lack of adequate pre-filing investigation.
The updated rules also change the standard for claim construction in certain cases, while leaving the standard unchanged in the majority of cases. Specifically, for most cases the PTAB will continue to use the same “broadest reasonable interpretation” standard employed by the Office during prosecution, under which petitioners can more easily achieve art-based invalidation due to the broader claim scope. However, in situations where the patent at issue will expire during the PTAB review and cannot be amended, the PTAB will instead use the narrower “Phillips-type” interpretation used by the district courts.
Further, the rules governing the length of submissions have been altered. Instead of imposing a page-based limit on submissions — i.e., petitions, patent owner preliminary responses, patent owner responses, and petitioner replies — the rules now impose a word-based limit. This change incentivizes inclusion of illustrations and diagrams, as these will not count toward the word limits.
Finally, the updated rules now permit patent owners to include testimonial evidence within their preliminary responses arguing against the institution of a review. Since the previous rules permitted petitioners challenging a patent to include such testimonial evidence in their initial petition, the ability of patent owners to respond with testimonial evidence of their own would appear to level the playing field, which many view as skewed in favor of petitioners in light of the estimated 70-75% invalidation rate of PTAB reviews thus far. However, in view of the infeasibility of cross-examining the patent owner’s declarant within the compressed three-month schedules afforded by the reviews, the updated rules contain a limitation that may negate much of the value of including testimonial evidence. Specifically, the updated rules clarify that “if a genuine issue of material fact is created by testimonial evidence, the issue will be resolved in favor of petitioner solely for institution purposes, so that petitioner will have an opportunity to cross-examine the declarant during the trial.” This limitation essentially imposes a summary judgment-like hurdle that the patent owner must overcome if it wishes to prevent institution of the review. Assuming that petitioners will usually be able to point to some factual dispute, the PTAB will likely proceed to institute reviews in most cases, regardless of whether the patent owners introduced testimonial evidence in their preliminary responses.
By Sebastian Kaplan and Patrick Premo
There is now federal jurisdiction for trade secret theft. The Defend Trade Secrets Act of 2016 was signed into law on May 11, 2016 after being unanimously passed in the Senate and ratified in the House by a vote of 410-2.The DTSA became immediately effective for all misappropriation occurring after the bill’s enactment.
The DTSA creates a federal cause of action for trade secret misappropriation that largely mirrors the current state of the law under the Uniform Trade Secrets Act, which has been adopted by 48 states. The DTSA uses a similar definition of trade secrets, and a three-year statute of limitations, and it authorizes remedies similar to those found in current state laws. The DTSA also creates an ex parte seizure procedure for use in extraordinary circumstances where the party against whom the seizure is ordered “would destroy, move, hide, or otherwise make such matter inaccessible to the court, if the applicant were to proceed on notice to such person….” While the seizure may be carried out immediately, the new law provides that the court shall set a hearing not less than seven days after the issuance of the order. Finally, the law protects whistleblowers from retaliatory accusations of trade secret misappropriation, so long as the whistleblowers disclose trade secret information to government or court officials in confidence.
The DTSA will not preempt existing state law, which will preserve and afford plaintiffs' options in regards to whether to file federal or state claims and which court to select. It also notably omits any requirement that a trade secret plaintiff describe its trade secrets with particularity, which several states, including California, currently require. Significantly, the DTSA also prohibits injunctive relief based on the inevitable disclosure doctrine, which is consistent with California law and thus should not detrimentally impact employee mobility.
You and your company should consider four responses to the DTSA:
First, update your employment and confidentiality agreements to disclose the whistleblower immunity provisions in the DTSA. If you do not, your company is not eligible to recover double damages or attorney fees in trade secret litigation.
Second, reevaluate your company’s tolerance for bringing trade secrets claims. Many companies have been deterred from making claims in the past because of the uncertainties and delays in state courts. Federal courts, however, have smaller case loads, allowing them to more directly and efficiently manage such litigation.
Third, inventory your company’s trade secrets and evaluate the protections in place to maintain the confidentiality of those secrets. Preventative measures are far more effective, and less costly, at keeping your secrets safe than methods designed to try to stuff the proverbial genie back in the bottle.
Fourth, develop response plans for suspected misappropriation and for receiving a seizure order. Trade secret litigation usually moves very quickly. Having a plan for what to do in the event your secrets are stolen will prevent unnecessary delays that can compromise your rights. In addition, if you are in a very competitive space with fluid employee mobility, consider formalizing a seizure preparedness and response plan. Such a plan should aim to minimize the business disruption associated with seizure and to facilitate your lawyers' immediate response in moving for a dissolution of the seizure order and a finding of wrongful seizure.
The primary effect of the DTSA is to federalize trade secret misappropriation actions and ensure full access to the federal courts for trade secret litigants. Courts are likely to interpret provisions of the DTSA to be consistent with existing state law – but by federalizing trade secret law, Congress has paved the way for greater predictability in an area of law that has been subject to the patchwork law of 50 states.
By Emily Gische
Some of the most significant changes to the trademark law system in the European Union since its establishment in 1996 went into effect earlier this year. Brand owners should be particularly aware of the main changes outlined below:
The most visible reform involves changes in nomenclature: As of March 23, 2016, the Office for Harmonization in the Internal Market, the trademark registry for the European Union, is known as the EU Intellectual Property Office. And the Community trade mark, which provides protection in all 28 EU member states, is now known as the European Union trade mark.
The fee structure for filing EUTM applications is also changing. Previously, applicants could file one application covering up to three “classes,” or categories, of goods and services, for 900 Euros. This three-for-one pricing structure often incentivized applicants to file for more classes than necessary. The reforms introduce a tiered fee structure: one class of goods or services for 850 Euros, two classes for 900 Euros, and three classes for 1,050 Euros. This tiered structure may help reduce clutter on the registry, as brand owners will now think twice before applying to register in more than one class. Renewal fees for existing trademark registrations have also been reduced, from 1350 Euros to 850 Euros for one class, with fees for additional classes following the same structure as that of application fees.
Starting in October 2017, EUTM applicants will no longer have to represent their marks graphically. This change opens the door to possible registration of non-traditional marks, such as scent or sound marks. But, it remains to be seen how the EUIPO will interpret the new language that replaces the graphic representation requirement, as the new laws mandate that any application be represented “in a manner which enables the competent authorities and the public to determine the clear and precise subject matter of the protection afforded to its proprietor.”
The reforms also codify the June 2012 ruling by the Court of Justice of the European Union, Chartered Institute of Patent Attorneys v. Registrar of Trade Marks Case – 307/10 (known as the ”IP Translator” decision). The IP Translator decision ruled that EU trademark registrations with class heading language were deemed to cover only those goods and services falling within a literal interpretation of the class heading (i.e., what you see is what you get).
“Class headings” describe in broad terms the nature of the various classes of goods and services under the Nice Classification system covering trademark applications. For example, the class heading for a class 9 application (the class that includes downloadable software) is “scientific, nautical, surveying, photographic, cinematographic, optical, weighing, measuring, signalling, checking (supervision), life-saving and teaching apparatus and instruments; apparatus and instruments for conducting, switching, transforming, accumulating, regulating or controlling electricity; apparatus for recording, transmission or reproduction of sound or images; magnetic data carriers, recording discs; compact discs, DVDs and other digital recording media; mechanisms for coin-operated apparatus; cash registers, calculating machines, data-processing equipment, computers; computer software; fire-extinguishing apparatus.” Prior to the reforms, an application with this class heading language would have been interpreted liberally, such that a brand owner could claim rights for all goods falling under class 9, even if not specifically listed. With the new reforms, if a certain good or service is not explicitly listed in the description, it’s deemed not covered, despite the inclusion of any broad class heading language. This being said, owners of EUTMs filed before June 22, 2012, that include class heading language, now have an opportunity to add specific goods and services to reflect those actually provided under the mark. EUTM owners have until September 24, 2016, to make such changes. Brand owners should review their EU trademark portfolios and consider taking advantage of this opportunity to ensure the broadest protection possible.
By Kathleen Lu and Jennifer Stanley
Recently, professors from the University of California Berkeley School of Law’s Samuelson Law, Technology & Public Policy Clinic and Columbia University’s American Assembly issued a report summarizing three studies on takedown notices under the Digital Millennium Copyright Act regime.
The report is enlightening and very useful, with an accessible executive summary, primer on the law, and insightful analysis and recommendations helpful to anyone who deals with DMCA notices even occasionally. A few highlights follow on the lessons we can draw from the three studies published in the report:
Information about service provider and copyright holder norms is hard to find. Study 1 summarizes interviews with online service providers and copyright holders/agents. Both groups have generally proven reluctant to share information about their practices, even with researchers. Service providers worry that speaking publicly will make them targets of copyright trolls or the recipients of a deluge of takedown notices; copyright holders and agents worry that revealing some of the methods they use to find allegedly infringing work online could result in potential infringers working around those methods. But study participants emphasized how little they knew about how their peer companies act in regards to takedown notices, and expressed a wish for greater transparency in this regard. By promising confidentiality to the participants, Study 1 appears to be the first to examine the real-life practices of service providers of various sizes, as well as of copyright holders and agents – and to give companies a valuable snapshot view into how their peers act on takedown notices. One of its most notable gleanings is the revelation that, despite public focus on algorithms, filtering, and automated takedown systems, the majority of services providers still use manual methods of processing DMCA notices.
Weaknesses in automated sending algorithms lead to erroneous notices. Study 2 looks at a random sample of takedown notices and finds that those generated through the application of algorithms contained certain types of mistakes that humans would not make, such as confusing a documentary that included the word “Bees” in the title for a song by “The Bees” band, or sending thousands of takedown notices targeting websites that had been defunct for over a year. Online service providers have limited options for correcting such errors (which affect approximately 30 percent of notices), especially if they do not have the resources necessary to manually review millions of notices. Copyright holders, on the other hand, would be wise to monitor and to audit the agents they use. Copyright agent algorithms vary in quality, and while some copyright agents regularly conduct manual spot checks and correct and adjust their algorithms, others simply blast away on generic and overly broad keywords (such as, apparently, the word “bees”). Takedown notices that target fair uses, fan works, or even authorized works can generate negative press for copyright holders, cost them fans and legitimate customers, and even expose them to legal threats or action.
Individual senders are less likely to understand the law. Study 3, supported by anecdotes from Study 1, shows that individual/inexperienced senders are more likely to make fundamental mistakes in their DMCA takedown notices, including leaving out portions of the notice that are required by statute, and even misusing the notice (e.g., to address trademark, privacy, or defamation concerns rather than copyright concerns). Many online service providers, including some with automated notice-processing systems, now flag notices from new senders for manual review because they believe such notices have higher error rates. Moreover, some specific abusive senders appear to send multiple defective notices to multiple services, including one individual who accounted for half of the total notices in Study 3. The report recommends a number of steps for clarifying proper notice use to potential senders, including (1) redesigning online forms to more clearly identify the specific concerns they are intended to address (e.g., copyright infringement, trademark violation, privacy intrusion, etc.), which would help to reduce senders' erroneous reliance on DMCA notices as a universal complaint mechanism; (2) providing educational materials explaining what copyright protection is and how fair use generally works; and (3) providing clear instructions, via an easy-to-use interface, for creating a proper takedown notice containing all of the required elements.