Apple Inc. v. Motorola, Inc.

Citations: 757 F.3d 1286; 94 Fed. R. Serv. 301; 110 U.S.P.Q. 2d (BNA) 1695; 2014 U.S. App. LEXIS 7757; 2014 WL 1646435Docket: 12-1548

Court: Court of Appeals for the Federal Circuit; April 25, 2014; Federal Appellate Court

Original Court Document: View Document

SummaryOriginalCytator Case TreatmentBackward CitatorAnalysis

EnglishEspañolSimplified EnglishEspañol Fácil

The case involves Apple Inc. and NeXT Software, Inc. as plaintiffs-appellants against Motorola, Inc. and Motorola Mobility, Inc. as defendants-cross appellants, under appeal 2012-1548 from the U.S. District Court for the Northern District of Illinois. The appeal was decided on April 25, 2014, by Circuit Judge Richard A. Posner. Legal representation for the plaintiffs included E. Joshua Rosenkranz and several associates from multiple law firms, while the defendants were represented by David A. Nelson and his team. Numerous amici curiae participated in the case, representing various organizations and corporations, including The American Antitrust Institute, the United States Federal Trade Commission, and major tech companies such as Intel Corporation and Microsoft Corporation, among others. Each amicus provided legal briefs supporting either side of the dispute, indicating the case’s significance in areas of intellectual property and antitrust law.

Apple Inc. and Next Software, Inc. filed a patent infringement complaint against Motorola, Inc. and Motorola Mobility, Inc. on October 29, 2010, in the U.S. District Court for the Western District of Wisconsin, asserting three patents. Motorola counterclaimed with six patents and both parties sought declarations of non-infringement and invalidity. The case was transferred to the Northern District of Illinois, where claim construction was completed. The district court granted summary judgment of non-infringement on certain claims and excluded most damages expert evidence, ultimately dismissing all claims with prejudice before trial.

On appeal, six patents are contested: Apple’s U.S. Patent Nos. 7,479,949; 6,343,263; and 5,946,647; and Motorola’s U.S. Patent Nos. 6,359,898; 6,175,559; and 5,319,712. The appellate court affirmed the district court’s claim construction decisions, except for the construction of Apple’s ’949 patent. It reversed the exclusion of damages evidence and the summary judgment of no damages for infringement of Apple’s patents. The court vacated the summary judgment regarding Apple’s injunction request due to the reversal of the claim construction of the ’949 patent but affirmed that Motorola is not entitled to an injunction for infringement of the FRAND-committed ’898 patent. The court addressed claim construction issues concerning several patents, concluding that the district court had erred in its construction of the ’949 patent, leading to the reversal of its summary judgment decision.

The ’949 patent addresses touchscreen control via finger contacts, detailing claims 1, 2, 9, and 10 with specific limitations. Claim 1 describes a computing device featuring a touchscreen, processors, memory, and programs for detecting finger contacts, applying heuristics to determine commands, and processing those commands. The heuristics include a vertical scrolling heuristic, a two-dimensional translation heuristic, and a next item heuristic for transitioning between items based on finger movement angles. 

The district court ruled the term “heuristic” as not indefinite, defining it as "one or more rules to be applied to data to assist in drawing inferences." It found the heuristic limitations in the claims were drafted as means-plus-function limitations under 35 U.S.C. § 112, despite lacking the term "means." The court determined that the specification provided adequate corresponding structure, specifically limiting the "next item heuristic" to include a finger tap on the device's right side.

Motorola sought summary judgment for non-infringement based on this construction. The court found that only the Amazon Kindle application, which uses the finger tap as defined, constituted infringement, leading to the granting of Motorola's motion for the remaining products. On appeal, Motorola challenged the definiteness of "heuristic" and supported the means-plus-function conclusion, while Apple contended that the absence of "means" creates a presumption against such construction. The legal question of whether claim language triggers Section 112.6 is reviewed de novo, with Section 112.6 allowing claims to be expressed as means for performing specified functions without detailing structure, covering equivalent structures as described in the specification.

The means-plus-function analysis consists of two distinct steps, though they are analytically related. First, the claim limitation must be evaluated to determine if it is in means-plus-function format by assessing whether it has "sufficiently definite structure" for a person of ordinary skill in the art, which involves examining the specification and other evidence. If the claim is deemed to be in means-plus-function format, the second step requires reviewing the specification for corresponding structure. The dissent expresses concern that the analysis improperly seeks corresponding structure before confirming the claim's means-plus-function status, potentially leading to every such claim term being deemed indefinite. However, the majority clarifies that the initial step indeed involves determining whether the claim limitation suggests definite structure, necessitating consultation of the intrinsic record. It is possible for a claim to lack sufficiently definite structure even with some corresponding structure present in the specification. Notably, not all means-plus-function limitations are considered indefinite; only those without the term "means," lacking definite structure, and devoid of corresponding structure are. In the current case, since the claims convey sufficiently definite structure, the second step of the analysis is unnecessary. A claim lacking the term "means" creates a rebuttable presumption against the application of Section 112, which may only be rebutted if the claim does not recite sufficient structure for its function. This presumption is characterized as strong and seldom overcome. The majority asserts that the choice to include "means" is significant and should not be dismissed as a minor drafting decision in the context of Section 112.6 analysis.

The absence of "means" in a claim limitation creates a strong presumption that enhances clarity and predictability for both the public and the patentee. This presumption aids the public in identifying when claim elements are limited to disclosed structures or equivalents and allows the patentee to effectively choose between means-plus-function claiming and broader structural claims. A claim drafted in broad structural terms may be more susceptible to invalidity challenges; however, the choice to draft in this manner is ultimately the claim drafter's responsibility, and courts should not rewrite claims to adopt a means-plus-function format. The focus remains on the claim language selected by the patentee, reinforcing the principle established in Phillips v. AWH Corp. that the claim language is paramount during construction.

In this context, Motorola has the burden to overcome the presumption against the application of Section 112.6. The district court made several errors, incorrectly concluding that Motorola rebutted this presumption by misapplying legal precedents requiring a detailed algorithm disclosure for means-plus-function claims. In cases where "means" is explicitly used, an algorithm must be disclosed unless a general-purpose computer suffices for the function. However, when a claim does not include "means," the relevant inquiry is whether the limitation provides sufficient structure to a person skilled in the art, which in this case, it does. Notably, the structure relevant to software claims differs from traditional mechanical structures, as software lacks physical components, making it inappropriate to seek physical structure in such claims.

The "structure" of computer software is characterized through various representations, such as algorithms, flowcharts, or explicit instructions, allowing a skilled individual to create an operative program for a designated function. Patents must disclose enough structural detail for such individuals to fulfill the specified function without requiring traditional physical structures, as doing so could render the limitations indefinite under means-plus-function analysis. A claim term conveys sufficient structure if it is defined in the patent's specification or is commonly recognized in the field. Terms used in everyday language or by professionals in the relevant art can indicate structure, as evidenced by the term “detector,” which was deemed sufficiently structural in a previous case. Structural definitions can also be derived from a limitation's operation, including its inputs, outputs, and connections, rather than merely its function. Generic terms like "means," "element," and "device" typically lack definite structure unless contextualized by the specification, as shown in cases where such terms were accompanied by detailed descriptions of their operational roles. Claims are interpreted in light of their supporting descriptions, maintaining the presumption against means-plus-function claims when adequate structural context is provided.

If a claim limitation includes a term with an established structural meaning or describes a class of structures, the presumption against means-plus-function claiming remains valid. A term does not need to specify a single structure; it can refer to a group of known structures, as demonstrated in cases like *Personalized Media Commc’ns v. Int’l Trade Comm’n*, where "detector" indicates a variety of structures. Even if all structures performing a function are described in the patent, this does not negate the presumption against means-plus-function claiming in the absence of the term “means.” Conversely, if a claim features a generic term without sufficient structural description in the claim language or specification, the presumption is rebutted. In *MIT*, the term "colorant selection mechanism" lacked sufficient structural specificity, leading to a determination that means-plus-function claiming applied.

The analysis then shifts to the claim limitations of the ’949 patent, where the term “heuristic” is found to have a known meaning and is adequately described in terms of its operation, input, and output. The function of these limitations is to identify commands based on finger contacts, with "heuristics" being defined as rules applied to data to infer results. Unlike generic terms such as "mechanism," "heuristic" is not a structureless nonce word; it conveys definite structural implications understood by skilled individuals in the field. Therefore, the district court's conclusion that "heuristic" signifies a recognized structure is upheld, emphasizing that a term used in common parlance or by experts to indicate structure will not invoke means-plus-function analysis. The functional definition of "heuristic" does not diminish its structural definiteness, aligning with precedent that many devices derive their names from their functions.

In Greenberg v. Ethicon Endo-Surgery, Inc., the court refrained from determining if the term “heuristic” alone provides enough structure to counteract the presumption against means-plus-function claiming. The claims in question do not merely mention heuristics without additional context; instead, they detail the operation of heuristics within the invention, specifying inputs and outputs. The input is defined as a finger contact, which can manifest as taps, swipes, or rolls, with variations based on the angle of contact. The claims outline multiple objectives for the heuristics, such as vertical scrolling and two-dimensional translation, and distinguish between these based on the angle of the finger's initial movement. For instance, an upward swipe within a specific angle results in one-dimensional scrolling, while one outside that angle enables two-dimensional translation. The specification provides comprehensive descriptions of how users navigate lists and content within frames, detailing gestures like finger taps and swipes, and the structural implementations of these actions. Figures from the patent further illustrate these concepts and their associated gestures. The court concludes that the claims articulate a known term meaningfully and explain their operation, thus offering sufficient structure to a skilled person in the field, thereby maintaining a strong presumption against means-plus-function claiming.

The district court's interpretation of the "heuristic" claim limitations as means-plus-function format has been reversed, and its summary judgment of non-infringement vacated. In regard to Apple's ’647 Patent, the court addressed disputes over the terms "analyzer server" and "linking actions to the detected structures." The district court defined "analyzer server" as "a server routine separate from a client that receives data having structures from the client" and "linking actions to the detected structures" as "creating a specified connection between each detected structure and at least one computer subroutine that causes the CPU to perform a sequence of operations on that detected structure." Apple contended that both definitions were incorrect; however, the court affirmed the district's construction as accurate.

The ’647 patent describes a system for recognizing structures (like phone numbers) on a touchscreen and associating actions (such as dialing the number) with these structures. Claim 1 outlines a system comprising an input device, output device, memory with program routines, a user interface, an action processor, and a processing unit. The district court upheld that "analyzer server" must be construed as separate from the client, as supported by the patent's specification and the understanding of a person skilled in the art. Apple’s proposed definition, which suggested that the analyzer server does not need to be separate from the client, was found to contradict the claim language and failed to provide evidence for a different ordinary meaning. The specification clearly delineates the analyzer server as a distinct component within the system architecture.

Apple's proposed construction omits the term "analyzer server," effectively altering the claim language by disregarding this key term. Legal precedent favors claim constructions that maintain the meaning of all terms, as established in cases such as Merck Co. v. Teva Pharms. and Pause Tech. LLC v. TiVo, Inc. The district court's interpretation aligns with the ordinary meaning of "server" and is supported by the specification, leading to the affirmation of its construction of "analyzer server."

Additionally, the district court accepted Motorola's interpretation that "linking actions to the detected structures" means creating a specific connection between each detected structure and at least one computer subroutine, which Apple contests. Apple argues against the "specified connection" limitation and claims that the language should merely require "associating" actions with structures. However, the distinction between "associating" and "linking" is emphasized in the specification, indicating that linking denotes a more definitive relationship than associating. The specification differentiates between the two terms, illustrating that linking occurs after detecting structures based on specific patterns, underscoring that a "specified connection" does not restrict the claims to just the use of pointers mentioned in the specification.

A pointer, defined as a term in computer engineering, refers to a mechanism that stores a computer memory address, facilitating the linking of actions to detected structures. The specification clarifies that upon detecting a structure, the analyzer server utilizes conventional pointers to associate actions with it. The parser links grammar-associated actions to the identified structure by retrieving pointers from a grammar file, which direct the system to the corresponding actions in an associated actions file. While the district court noted that linking is achieved through pointers, it did not restrict claims solely to pointers, interpreting linking as requiring a "specified connection." This interpretation aligns with the specification's distinction between linking and associating. The court also rejected Apple's assertion that each structure must be linked to multiple actions, emphasizing that the claims allow for at least one action per structure, as illustrated in Figure 4 of the patent, which shows a single action linked to a structure.

Regarding Apple's ’263 Patent, which pertains to a system for real-time data processing, the dispute centers on whether the "real-time application program interface (API)" must operate in real-time or simply facilitate real-time processing by other subsystems. The district court concluded that the API needs only to facilitate real-time processing, defining "real-time API" as one that enables real-time interaction between subsystems. Motorola contends that this interpretation eliminates the "real-time" aspect from the claim; however, the court affirms its construction.

A signal processing system is designed to facilitate multiple real-time services for various independent client applications and devices. The system includes a host CPU that operates with an application program and a device handler program, along with an adapter subsystem that collaborates with the CPU and devices. It features a real-time signal processing subsystem capable of executing various data transformations through real-time signal processing operations. A real-time application program interface (API) connects the subsystem to the real-time signal processing subsystem, enabling interoperability with real-time services.

The district court clarified that for a system to be considered "real-time," it must meet specific response-time constraints to avoid performance degradation. Motorola argued that the district court's interpretation excluded the API from functioning in real-time, asserting that the inclusion of "real-time" in claim 1 implied the API should operate in real-time. However, the court maintained that the patent specification identifies the API merely as an interface that transmits commands to the real-time engine responsible for processing, without imposing real-time requirements on the API itself.

The specification indicates that the real-time engine executes tasks such as text-to-speech conversion and video processing, while the API facilitates communication between applications and the real-time engine by supplying the necessary parameters for processing. Consequently, the district court concluded that the API does not need to perform real-time processing itself, and its role is limited to interaction with the real-time engine, which handles the time-sensitive data processing. Thus, the court affirmed that the API's design does not contradict the real-time functionality of the system, as it can still operate in real-time without being mandated to do so.

The claims and written description of the Motorola ’559 patent necessitate that the API support the real-time engine's functionality. The district court's construction of claim 5 requires that the multiplication of the outer code by the inner code occurs only after the formation of both codes is complete. This interpretation led to the court granting summary judgment of non-infringement to Apple, a decision Motorola is appealing. The relevant method described in claim 5 includes steps for generating preamble sequences in a CDMA system, specifically emphasizing the sequential nature of forming the outer and inner codes before multiplication. While generally, steps in method claims need not follow the written order, the specifics of grammar and the patent's specifications necessitate a sequential approach in this instance. The district court's ruling aligns with the claim language, which implies that "the" inner and outer codes must be fully formed prior to their multiplication, as supported by both the claim's wording and the patent's specification.

The mobile station generates a preamble sequence by multiplying the outer code with the inner code, as stated in the ’559 patent. The district court's claim construction, supported by the clear language of the claim and specification, is affirmed. This construction underpins the court's summary judgment ruling of non-infringement, making further examination of the term “a set of orthogonal codewords” unnecessary. Consequently, the court affirms Apple’s motion for summary judgment regarding non-infringement of the ’559 patent, based on the requirement that the steps of claim 17 be executed in the specified sequence.

Regarding Motorola’s ’712 patent, which outlines a data communication encryption system, the district court found that the claimed "transmit overflow sequence number" (TOSN) is never sent to the receiver in the claimed system. Since the equivalent in the accused products is transmitted, summary judgment of non-infringement for Apple was granted, and this ruling is upheld. Claim 17 details a method for cryptographic protection of a data stream, involving the assignment of a packet sequence number, updating the TOSN based on this number, and encrypting the packet prior to transmission. The specification clarifies that both a packet sequence number and an overflow number are essential for encryption and decryption. The packet sequence numbers increment until a maximum is reached, after which they reset, with the overflow counter tracking these resets. The TOSN is designated on the transmission side, while the receiving side uses a receiving overflow sequence number (ROSN). The specification notably does not describe the transmission of the TOSN or any overflow number, indicating that the receiver utilizes the packet sequence number it receives combined with a generated ROSN for decryption. This understanding aligns with the district court's conclusion that the TOSN is never transmitted.

Claims in question refer to a Transmission Overflow Sequence Number (TOSN) rather than a generic overflow sequence number, as explicitly outlined in the specification. The specification details the transmission of certain system aspects, like the data packet and packet sequence number, but does not imply that the TOSN is included in this transmission. The necessity of the TOSN being transmitted is negated by the receiver's role in generating a Receiver Overflow Sequence Number (ROSN). Transmitting both the TOSN and packet sequence number would compromise security by allowing an eavesdropper to intercept the full key needed for message decryption, contradicting the invention's purpose of enhancing security by not transmitting the entire key.

Motorola's statements during the prosecution of a related Japanese patent reinforce this interpretation. They emphasized that the overflow sequence number is never transmitted, further arguing that this non-transmission prevents interception of the overflow sequence number, thereby providing greater security. These assertions were consistently made by Motorola and relate to both the Japanese application and the U.S. patent (the ’712 patent), which share the same specification.

The relevance of Motorola’s later statements to the earlier issued claims is supported by precedent, indicating that remarks made during the prosecution of related patents can inform claim construction. Specifically, statements made in a foreign patent application may be relevant if they pertain to a patent's prosecution history. However, statements from unrelated applications are generally not relevant for claim construction. In this case, the Japanese application is part of the prosecution history of the U.S. patent in question, making its statements applicable to the claim construction process.

The court examined arguments made by the patentee before the European Patent Office (EPO) and found a "blatant admission" by the defendant that supported the court's claim construction. It referenced several precedents, including Tanabe Seiyaku Co. v. U.S. Int’l Trade Comm’n, which emphasized caution against relying on foreign patent application prosecutions for claim construction. Statements from Motorola during the prosecution of the '893 patent in Japan were deemed relevant, as they clarified that the TOSN "is never transmitted" to the receiver, enhancing system security. The court noted that these statements were made in an official context where Motorola had an incentive to accurately define its invention's scope. The related patents shared a common PCT application and identical specifications, reinforcing the construction derived from the specification. The district court's reliance on Motorola's statements was upheld, leading to the affirmation of its grant of summary judgment for Apple regarding non-infringement of the '712 patent.

In the context of damages, both parties contested the admissibility of each other's expert testimony during a Daubert hearing. The district court excluded most expert evidence related to damages for infringement, but on appeal, Apple and Motorola argued for the admissibility of their respective damages experts for certain patents. The appellate court found that the district court erred in excluding this expert evidence and reversed the decision.

The Federal Rules of Evidence, particularly Rules 702 and 703, govern the admission of expert testimony, with key guidance from the Supreme Court's decisions in Daubert v. Merrell Dow Pharmaceuticals, Inc. and Kumho Tire Co. v. Carmichael. The Daubert case established that the admissibility of expert testimony should focus on the principles and methodologies used, rather than the conclusions drawn. Kumho Tire reaffirmed that district courts hold a "gatekeeping obligation" for all expert testimony. 

Rule 702 specifies that an expert's qualifications must derive from knowledge, skill, experience, training, or education, and their testimony must aid the trier of fact, be based on adequate data, derive from reliable methods, and apply those methods appropriately to the case facts. Rule 703 allows experts to base opinions on facts that they are aware of or have observed, even if those facts are not admissible, provided their probative value significantly outweighs any prejudicial effects.

District courts can exclude expert evidence if it is founded on unreliable methods or insufficient data. The focus of the gatekeeping function is on the expert’s methodology, not the correctness of their conclusions, which are left for the jury to evaluate. Judges should not substitute their judgment for that of the fact finder or assess the credibility of experts. The jury is responsible for weighing the soundness of the expert's analysis and the validity of their conclusions, even if those conclusions are subject to doubt.

Judicial gatekeeping in determining the admissibility of expert testimony focuses on excluding unreliable methods, particularly crucial in patent damage cases. The Seventh Circuit has established that questions about the relevance and reliability of facts used to calculate reasonable royalties are for the jury, provided the methodology is sound and the evidence is sufficiently related to the case. Estimating a reasonable royalty is inherently imprecise, often resulting in a range of acceptable values and potentially multiple reliable methodologies. Various approaches—such as using comparable licenses, assessing infringed features against market equivalents, or evaluating benefits against non-infringing alternatives—can yield admissible testimony, each with its own strengths and weaknesses. Disputes over the relative merits of these methodologies are appropriate for trial rather than grounds for inadmissibility. 

In reviewing the district court’s damages rulings, the Seventh Circuit applies its legal standards for admissibility and assesses the district court's adherence to proper frameworks for admitting expert testimony, reviewing for abuse of discretion. Specifically regarding Apple’s ’949 patent, the court found that the district court improperly excluded Apple’s expert damages testimony due to reliance on an incorrect claim construction, which necessitates reversal and remand.

The ruling has been vacated, determining that the patentee is entitled to damages, and the case is remanded for reconsideration based on a modified claim construction. The district court previously erred by providing incorrect claim constructions to the jury and not fully considering the asserted claims' scope. Additionally, it improperly questioned Apple’s expert's conclusions and substituted its opinion without adequately assessing the reliability of the expert's methods or the sufficiency of the underlying facts.

Apple's expert, Brian W. Napper, analyzed the Magic Trackpad to estimate the value of the asserted claims of the ’949 patent. The Trackpad is a touchpad that replaces a mouse for Mac computers and recognizes multiple finger gestures for various commands. Napper asserted that the Trackpad's touch gestures are comparable to the claimed features of the patent, supported by the testimony of Dr. Stephen Feiner, another Apple technical expert.

To isolate the value of the Trackpad’s touch features, Napper compared its price to that of a traditional mouse, attributing the price difference to the Trackpad's touch features and wireless connectivity. Since wireless connectivity is not part of the asserted claims, he discounted his royalty estimate accordingly. Napper also reduced his royalty calculation to exclude additional Trackpad features not covered by the claims. Ultimately, he arrived at a reasonable royalty estimate and compared it to royalty rates Motorola had paid for related technology, concluding that Motorola would likely pay a higher royalty for a license to the asserted claims due to the technology's similarity and the nature of the licensing parties.

Napper testified that Motorola could have opted to exclude the asserted functionality from its products, which would result in a less functional touchscreen and negatively affect user experience, leading to consumer frustration. He concluded that this alternative would not provide a reasonable basis for calculating damages. The district court reviewed Napper’s expert report and excluded all his proposed testimony related to the ’949 patent, largely due to its incorrect claim construction that confined the claims to products with a pre-loaded Kindle application utilizing a tap gesture. The court determined that the critical aspect of the ’949 patent involved using a tap on the right side of the screen to turn pages in a Kindle book on a cell phone. This narrow interpretation resulted in the court dismissing much of Napper’s analysis as irrelevant to the asserted claims. Despite recognizing that Napper's report addressed the broader scope of the claims, the court ultimately found his testimony regarding the vertical scrolling feature unreliable. It concluded that comparisons to a Magic Trackpad were insufficient for estimating the value of the tap gesture or vertical scrolling feature. The court's focus on isolated claim limitations, rather than the comprehensive scope of infringement, was a significant error. It did not assess whether Napper employed reliable principles or sufficient data to value the entire scope of the asserted claims, which is crucial for determining damages under 35 U.S.C. § 284.

Claim constructions define the scope of infringement and the applicable damages, prompting the need to evaluate whether Napper's methodology was reliable and legally supported. The court found that Napper employed sound principles and methods, contrary to the district court's ruling. His approach involved comparing the claimed features of the technology with those of the Trackpad, isolating the value of similar features while accounting for non-claimed aspects. Napper's resulting reasonable royalty was based on comparisons to related licenses, and he argued that removing the asserted features would harm consumer value. His testimony was backed by Apple’s technical experts and was grounded in reliable application of methods. The district court's dismissal of Napper's testimony, claiming it failed to quantify specific improvements in gestural control, was deemed incorrect; Napper's analysis was specifically focused on the asserted claims. The district court did not provide evidence of flaws in Napper's methodology, which included identifying comparable features, isolating their value, and rationalizing royalty rates based on similar agreements.

The court accepted the testimony of Apple’s technical expert regarding the similarity of the Trackpad’s features to the asserted features without questioning its credibility. It did not identify any deficiencies in Napper’s method of cost apportionment or justify why its alternative valuation method was preferable. The existence of other reliable royalty estimation methods does not automatically render Napper’s approach inadmissible. The court criticized Napper for not estimating the amount a consumer would pay to avoid unsuccessful swiping, but Napper's methodology did account for what consumers pay for comparable features. Importantly, the statute necessitates a determination of a "reasonable royalty," not a direct consumer price. While the consumer valuation of the infringing feature can inform royalty estimation, it is not mandatory for admissibility. Motorola can challenge the Trackpad benchmark's accuracy, but such arguments pertain to evidentiary weight, not admissibility. Any technical differences or flaws in Napper’s analysis can be addressed during cross-examination. The district court's ruling on Napper’s estimated value precluded Motorola from effectively rebutting Napper’s opinion with its own evidence, undermining the adversarial process. The admissibility of Napper’s testimony should focus on the reliability of his methods, while the accuracy of the valuation should be settled by the jury. The district court improperly determined admissibility based on its own valuation views, which should have been left for the fact-finder.

The district court's analysis was flawed due to its incorrect claim construction and failure to fully assess the scope of infringement. It improperly questioned the factual basis of Napper’s testimony instead of focusing on the reliability of his methods and data. Consequently, the exclusion of Napper's expert testimony regarding the ’949 patent was reversed. Additionally, the district court excluded Apple’s damages evidence related to the ’263 patent, asserting that Napper relied on information from a technical expert, Dr. Polish, which led to a summary judgment in favor of Motorola that denied Apple damages. The court's decision to exclude Napper's testimony was found to be erroneous, as Napper identified two design-around approaches for the ’263 patent, ultimately concluding that Motorola would incur higher costs through one method, leading to a reasonable royalty estimate. The district court's rationale for excluding Napper's testimony was based on a hypothetical conversation suggesting reliance on an engineer from Apple was inappropriate, a stance that was deemed incorrect since experts often rely on the expertise of others in related fields.

In the context of expert testimony, courts recognize that experts may rely on the opinions or data from other experts within their field, as established in cases such as **Dura Automotive Systems of Indiana, Inc. v. CTS Corp.** and **Monsanto Co. v. David**. Under **Rule of Evidence 703**, experts can utilize information they are made aware of if it is the type that professionals in their field would reasonably depend upon in forming their opinions. This rule does not restrict admissibility based on the source of the information or whether the source is associated with one of the parties involved.

Patent damages experts frequently draw from technical knowledge outside their specific area when assessing design alternatives or the significance of infringing features, which often involves complex technical questions related to infringement scope. In the case of **Apple Inc. v. Motorola, Inc.**, the district court incorrectly determined that Apple could not support its damages calculation with insights from its hired technical expert, Dr. Polish, citing potential bias due to his employment by Apple. However, while bias in expert testimony is a valid concern, it pertains to the weight of the testimony rather than its admissibility. The potential for bias can be addressed through cross-examination and the presentation of opposing expert testimony.

The court emphasized that the fact an expert may have a financial interest does not automatically disqualify them, so long as they provide an objective analysis. Any issues regarding bias should be considered by the trier of fact who can evaluate the conflict of interest. If Motorola contests the validity of Dr. Polish’s suggestions, they can challenge this through cross-examination and their own expert’s testimony, reinforcing that bias affects evidentiary weight but not admissibility.

The district court suggested that Dr. Polish's recommendation of a specific replacement chip was not credible expert information outside litigation. However, it was argued that Rule 703 requires expert testimony to be connected to the case's facts, and that it is reasonable for an expert like Napper to seek technical advice from specialists when valuing a patent. Such reliance on technical information from Apple or its experts is common and maintains intellectual rigor similar to courtroom standards. Consequently, the court reversed the exclusion of Napper's testimony regarding the ’263 patent and vacated the summary judgment on damages.

Regarding Motorola’s ’898 patent, the district court excluded all testimony related to damages and granted summary judgment to Apple, ruling Motorola was entitled to no damages. While affirming the exclusion of testimony from Motorola's licensing expert, Charles Donohoe, the court erred by excluding other expert testimony that employed reliable methods to analyze comparable licenses. Motorola’s damages expert, Carla S. Mulhern, analyzed license agreements with major U.S. cellular phone makers, excluding Apple, and found that Motorola previously received a royalty rate of around 2.25% for its entire standard-essential patent portfolio. Mulhern also reviewed licenses between Apple and third parties related to cellular communications technology, which also commonly included cross-licenses. The court reversed the exclusion of Mulhern's remaining testimony and vacated the summary judgment.

Apple's royalties from its agreements with Motorola were consistent with industry standards. Expert Mulhern distinguished the value of Motorola's ’898 patent from its overall standard-essential patent (SEP) portfolio, noting that individual patents may hold varying values based on their significance to the technology and standards. He referenced economic studies indicating that while many patents hold minimal value, a few are highly valuable. During hypothetical negotiations, a licensee would expect to pay less for a few patents than for an entire portfolio of SEPs. Mulhern asserted that the ’898 patent was a significant innovation and disproportionately valuable within Motorola's portfolio. He emphasized that licensing typically involves large portfolios rather than individual patents, yet acknowledged that negotiations would likely require licensing a subset of these patents, leading to a "nonlinear" royalty structure where initial patents would command higher rates—estimated at 40%-50% of the total portfolio rate. This estimate was supported by testimonies from Motorola and Ericsson licensing experts, as well as historical practices from IBM. Additionally, Mulhern suggested a conservative estimate placing the ’898 patent at 5% of Motorola's essential patents, based on a linear valuation method. He noted that Apple could have avoided infringement by limiting the iPhone's release to the Verizon network, which did not utilize the GMS/UTMS networks.

Mulhern concluded that the ’898 patent was essential for GMS/UTMS mobile networks, indicating that Apple needed a license to use these networks. She noted impracticalities with Verizon's options but did not estimate the cost difference for Apple between pursuing a Verizon phone versus an AT&T phone. The district court found that Motorola's proposed royalty rates met fair, reasonable, and non-discriminatory (FRAND) conditions but excluded Mulhern's testimony due to her failure to analyze plausible alternatives, particularly the potential contract with Verizon. Moreover, Mulhern did not provide evidence to substantiate that contracting with Verizon would cost Apple an additional $347 million.

Regarding Donohoe's testimony, the district court criticized his lack of basis for determining the correct nonlinear royalty rate and noted he had no knowledge of the specific portfolio that included the ’898 patent. His 40-50% figure was deemed too general and not applicable to the case's details. The court emphasized that proof of damages must connect directly to the claimed invention. Donohoe's admission of ignorance about the relevant portfolio weakened his reliability, leading to a conclusion that his testimony, even when incorporated into Mulhern's analysis, lacked necessary ties to the facts of the case. Although the theory presented by Donohoe and Mulhern—that initial patents in a portfolio may command higher royalties—was not inherently unreliable, their failure to link it to the specific patent rendered the expert testimony inadmissible. The court affirmed the district court's decision to exclude this portion of Mulhern's testimony based on Donohoe's declaration.

Disagreement exists with the district court regarding Mulhern's testimony, which, although lacking a direct cost estimate for Apple's choice of AT&T over Verizon, involved an analysis of the feasibility of contracting with Verizon. The district court's dismissal of Mulhern's conclusions—specifically that contracting with Verizon was impractical and did not require specific valuation—can be challenged at trial through cross-examination and expert evidence from Apple. The admissibility of an expert's damages testimony does not hinge on valuing every non-infringing alternative, and the district court overlooked that Mulhern constructed a cost estimate typically used in patent damage calculations. The testimony references established licensing practices and comparable licenses, which are deemed reliable for estimating patent value. The second Georgia-Pacific factor highlights that the rates paid for comparable patents are relevant, and established royalty rates usually serve as the best measure for a reasonable royalty, reflecting market conditions and the cost of non-infringing alternatives. Actual licenses are critical indicators of a reasonable royalty, as they demonstrate the economic value of the patented technology. The comparability of Motorola's licenses to determine a reasonable royalty is a matter of evidentiary weight rather than admissibility, and Mulhern's analysis utilized reliable principles and methods.

The court determined that the jury could accurately assess the value of the ’898 patent and the cost of non-infringing alternatives, relying on the expert testimony of Mulhern, with the exception of her reliance on Donohoe's declaration. Consequently, the court reversed the district court's exclusion of Mulhern's testimony and vacated the summary judgment concerning damages for infringement. Following the exclusion of damages expert testimony, both parties sought summary judgment on damages and injunctive relief, leading the district court to rule that neither party was entitled to damages or an injunction. However, based on the reversal of admissibility decisions, the court vacated the summary judgment regarding damages for Apple's ’949 and ’263 patents and Motorola’s ’898 patent. The court affirmed that Apple was not entitled to damages for the ’647 patent and that Motorola was not entitled to an injunction. In reviewing Motorola's motion for summary judgment on the ’647 patent, the court disagreed with the district court's conclusion that Apple could not rely on any evidence to establish damages. The finding that a royalty estimate may have factual flaws does not equate to a legal conclusion of zero reasonable royalty.

Patents at issue in this case are presumed valid and infringed, necessitating the court to award damages of at least a reasonable royalty per 35 U.S.C. § 284. The court is required to award damages once infringement is established, even if expert evidence is lacking, as established in Dow Chemical Co. The statute mandates that damages cannot be less than a reasonable royalty, and if a patentee's evidence does not support a specific royalty estimate, the fact finder must still identify a reasonable royalty based on the record. A presumption of entitlement to damages arises from the fact of infringement, which violates the patentee's right to exclude. The Federal Circuit affirms that the statute obligates the court to determine damages despite difficulties in establishing the exact amount. If a patentee fails to prove lost profits, it is justified for the court to award no lost profits but still provide a reasonable royalty. The court can reject extreme figures proposed by the parties and select a reasonable royalty based on the overall evidence, potentially awarding a low or nominal amount if the patentee's proof is insufficient.

A fact finder can only award zero damages if the evidence supports such a conclusion. Instances where no evidence exists to justify a damages award or where a defendant deemed the patent valueless at the time of infringement may warrant a zero royalty award. However, a hypothetical negotiation typically assumes some royalty would be agreed upon. There is no known case that supports a zero royalty award in the context of proven infringement. At the summary judgment stage, a zero royalty can only be awarded if there are no genuine material facts suggesting a non-zero royalty is reasonable. If a patentee raises a factual issue about entitlement to a non-zero royalty, summary judgment must be denied. The mere failure of a patentee to prove the accuracy of its royalty estimate does not justify a zero royalty award.

In the case at hand, Apple's damages expert, Napper, estimated a royalty based on the time HTC took to design around the same patent. The ITC had issued an exclusion order against HTC for infringing the same patent at issue. Napper concluded that Motorola would be willing to pay a $52 million royalty based on the operating profit loss from the exclusion order. However, the district court discredited this testimony, citing differences in claim construction between the ITC and the court, and a lack of sufficient analysis of the differences between HTC and Motorola. The court ultimately determined that Apple did not demonstrate any damages, as it failed to prove the accuracy of its royalty estimate and did not present admissible evidence supporting its claim. Consequently, the record did not justify a summary judgment of no damages.

Motorola failed to demonstrate that a zero reasonable royalty is the only viable option, as it provided no evidence or arguments to substantiate this claim, despite asserting that Apple's expert evidence is inadmissible. There is no indication in the record that Apple would accept no payment for Motorola's infringement or that Motorola believed the ’647 patent had no value at the time of infringement. In contrast, Apple presented expert evidence indicating that it would take four months to design around the ’647 patent and estimated the profits Motorola lost during this period based on its sales figures for the accused devices. Additionally, Apple argued that the features of the ’647 patent drive consumer demand for its products.

The existence of factual disputes regarding the reasonable royalty amount, specifically the flaws identified by the district court in Apple's HTC-based royalty estimate, does not legally establish a zero royalty. The court is required to approximate the damages owed to the patent owner, and the absence of expert evidence from Apple does not preclude a finding of a reasonable royalty. The reasonable royalty concept implies that a valid, infringed patent holder suffers legal damages at least equivalent to a lost licensing opportunity. Expert testimony is not mandatory to recover damages, and a qualified witness may provide insights on what constitutes a reasonable royalty under the circumstances.

Motorola failed to prove that zero is a reasonable royalty for the infringement of the ’647 patent, lacking evidence to support its claim. Conversely, Apple provided admissible evidence warranting a non-zero royalty, and the existence of factual flaws in Apple's estimate does not justify concluding that zero is reasonable. Consequently, the appellate court reversed the district court's summary judgment in favor of Motorola.

Regarding Apple's request for an injunction, the district court ruled that Apple had not established a causal link between the alleged irreparable harm and Motorola's infringement of the ’949 and ’647 patents. This ruling was based on an overly narrow interpretation of the claims of the ’949 patent. The appellate court's reversal of the claim construction alters the infringement scope relevant to Apple's injunction request. It emphasized that the causal nexus requirement is meant to demonstrate that a patentee suffers irreparable harm due to infringement, and that the court can consider aggregate harm from all infringing features rather than evaluating each patent individually. The court recognized that multiple patent infringements may bolster a patentee's case for an injunction if they contribute to consumer demand or irreparable harm, while also weighing public interest in maintaining access to non-infringing features. Therefore, the appellate court vacated the district court's summary judgment regarding Apple's injunction request due to the reversed claim construction.

As for Motorola's request for an injunction related to the ’898 patent, the appellate court noted that it reviews the district court's injunction decisions for abuse of discretion and grants of summary judgment de novo. It referenced the Supreme Court's eBay decision, which outlines factors to consider before issuing a permanent injunction, including the necessity for the plaintiff to demonstrate irreparable injury and inadequacy of legal remedies.

A remedy in equity is justified based on the balance of hardships between the plaintiff and defendant, and a permanent injunction would not harm the public interest. The ’898 patent is a standard-essential patent (SEP), requiring Motorola to license it on fair, reasonable, and non-discriminatory (FRAND) terms. The district court ruled that enjoining Apple from infringing the patent would only be appropriate if Apple refused to pay a FRAND-compliant royalty, as Motorola's commitment to FRAND implies that monetary compensation is sufficient for licensing the patent. The court emphasized that it erred if it applied a blanket rule against injunctions for SEPs, asserting that the framework established by the Supreme Court in eBay is adequate to address the unique characteristics of FRAND patents. 

A patentee under FRAND commitments faces challenges in proving irreparable harm; however, an injunction may be warranted if an infringer refuses a FRAND royalty or delays negotiations. Nonetheless, a mere refusal to accept any license offer does not automatically justify an injunction, especially if the offer is not FRAND-compliant. The public interest includes encouraging participation in standard-setting organizations and preventing the overvaluation of SEPs. The court concluded that Motorola is not entitled to an injunction for the ’898 patent's infringement, as its FRAND commitments indicate that monetary damages are sufficient for compensation, and it failed to show that Apple's infringement resulted in irreparable harm. Motorola's claims of Apple's refusal to negotiate and delay are noted, but the overall context suggests that adding Apple as a licensee would not cause significant harm given the existing market participants using the technology.

Negotiations between Apple and Motorola have been ongoing, with no evidence indicating that Apple has unilaterally refused to agree to a deal. Consequently, the district court's summary judgment ruling that Motorola is not entitled to an injunction for infringement of the ’898 patent is affirmed. The appeals court confirmed the decision to affirm in part, reverse in part, and vacate in part, with each party bearing its own costs. A dissenting opinion by Chief Judge Rader highlights concern over the denial of Motorola's injunction request, arguing that the record presents a genuine dispute regarding Apple's status as an unwilling licensee, resulting in irreparable harm due to its infringement of the patent. Rader emphasizes the complexity of assessing patents adopted as industry standards, noting that the valuation of technology in this context complicates the determination of whether a licensee is unwilling. The majority opinion acknowledges that while establishing irreparable harm in standard patent contexts can be challenging, an injunction is warranted if an infringer unreasonably delays negotiations or refuses a FRAND royalty. The dissent further discusses the economic implications of "hold out" versus "hold up" scenarios, suggesting that factual analysis is necessary to differentiate these situations and evaluate the underlying sources of value related to the patent and its standardization.

Evidence presented creates a material fact dispute, highlighting the district court's failure to distinguish between value derived from inventive contributions and that from standardization. This oversight undermined the court's ability to properly apply the eBay test for injunctions. The court considered Motorola’s commitment to license the ’898 patent under FRAND terms as conclusive, without addressing whether the offered royalty was genuinely fair and reasonable. The implications of Apple potentially refusing a FRAND license or delaying negotiations were not adequately examined by the court, which could have justified an injunction. The district court acknowledged conflicting evidence regarding Apple’s willingness to negotiate but did not allow Motorola to substantiate its claim that Apple's alleged unwillingness indicated irreparable harm. Therefore, the case should be remanded for further development of the record. The dissenting opinion emphasizes the need for a thorough factual inquiry into these issues, while also clarifying that the author supports the majority opinion on other matters but diverges on specific claims related to the ’949 patent and the summary judgment regarding Apple's injunction request.

The majority opinion in the case of Apple Inc. v. Motorola, Inc. concludes that the "heuristic" claim terms in the ’949 patent possess sufficient definite structure, thus avoiding means-plus-function treatment. However, there is disagreement regarding this interpretation. The dissent emphasizes that a means-plus-function analysis consists of two phases: first, determining if a claim term is in means-plus-function format under 35 U.S.C. § 112, ¶ 6, followed by identifying the corresponding structure described in the specification. The dissent argues that the majority conflates these phases, which could eliminate the distinction between terms that are not indefinite and those that qualify as means-plus-function, thereby undermining the integrity of means-plus-function claims.

The dissent further critiques the majority’s criteria for determining sufficient structure, which includes the presence of a structural definition in the specification or known in the art, outlining algorithms, or describing operational details. It contends that the majority's reasoning implies that any term with corresponding structure in the specification cannot be classified as means-plus-function, leading to an unreasonable outcome where all such terms would be deemed indefinite. The dissent asserts that, based on the correct legal framework, the heuristic limitations should indeed be classified as means-plus-function limitations, highlighting a fundamental misunderstanding in the majority's approach to the issue.

A court must thoroughly examine the specification to find corresponding structures to prevent a claim term from being treated as means-plus-function. The majority references several cases, such as Typhoon Touch Technologies, Inc. v. Dell, Inc., to argue that a claim can avoid means-plus-function status by including an algorithm. However, in Typhoon Touch, the term "means for cross-referencing" was undisputedly a means-plus-function limitation, and the court's search for an algorithm was to avoid indefiniteness. Similarly, in Finisar Corp. v. DirecTV Group, Inc., the court assessed whether sufficient algorithmic structure was disclosed for a computer-implemented means-plus-function claim. The majority's interpretation of these cases mirrors the error it attributes to the district court regarding Aristocrat Technologies Australia Party Ltd. v. International Game Technology.

The absence of "means" in disputed claim terms typically suggests they are not means-plus-function limitations, as established in Inventio AG v. ThyssenKrupp Elevator Ams. Corp. However, the heuristics limitations described in the claims involve specific functions, raising the question of whether adequate structure for these functions is present, which could override the presumption against means-plus-function treatment. The key consideration is whether the term is recognized as describing structure rather than being a non-specific term. The court in Personalized Media Communications highlighted that terms like "detector," while broad, have structural significance, in contrast to generic or unclear terms. Similarly, terms such as "computing unit" and "soft start circuit" have been deemed to convey sufficient structure to avoid means-plus-function classification, unlike terms such as "colorant selection mechanism" and "mechanism for moving said finger," which do not.

Welker Bearing, 550 F.3d at 1095-97, indicates that the term "heuristic" lacks sufficient structural specificity to avoid means-plus-function treatment, similar to vague terms like "mechanism" or "element." The district court's interpretation of "heuristic" as "rules to be applied to data to assist in drawing inferences" fails to convey identifiable physical structure, unlike terms such as "computing unit" or "circuit." This leads to the conclusion that heuristic limitations should be treated as means-plus-function limitations. The author argues against the majority's view that the "heuristic" limitations in the '949 patent do not invoke means-plus-function analysis, suggesting a need to revisit this issue for computer-implemented inventions. The author compares two claim limitations: one explicitly described as a means-plus-function limitation, while the other, using "next item heuristic," was deemed outside this treatment despite lacking additional structural clarity. This discrepancy allows broader claims without corresponding structural limitations as outlined in the patent specification, which only details two specific heuristics (a right-to-left swipe and a tap on the right side). The majority's interpretation risks enabling broad functionality claims without adherence to the constraints of 35 U.S.C. § 112, particularly when functional claiming in software patents is concerned, as highlighted by Professor Lemley’s commentary on the subject.

Many software patent claims have not been treated as means-plus-function claims, leading to the issuance of overly broad patents. The inclusion of generic structures like “a computer” or “a processor” allows claims to dominate the function they perform, circumventing restrictions from the 1952 Patent Act on functional claiming. Consequently, these patents often reflect not the actual invention but rather the functional outcome, resulting in claims that are not commensurate with what was invented. This issue creates patent thickets, where broad patents overlap, complicating the landscape for innovation. Concerns have been raised about this approach to means-plus-function claims, particularly regarding computer-implemented inventions. Although the court's majority opinion did not rely on generic terms, it still resulted in an expansive interpretation of Apple's '949 patent, encompassing the function of heuristics rather than the specific heuristics described. 

The speaker concurs with the district court's classification of heuristics as means-plus-function limitations and addresses the corresponding structure for the term “next item heuristic.” The patent specifies two examples: a tap on the right side of the touchscreen and a right-to-left swipe. The district court, however, rejected the swipe as corresponding structure, asserting that it overlapped with another claimed heuristic. It concluded that the same finger movement could not represent two commands. Therefore, the court limited the corresponding structure for the next item heuristic to a tap on the right side of the screen. Apple argues that the district court did not adequately consider that the heuristics might function in various scenarios.

The specification of the ’949 patent describes the functionality of the photo album application, highlighting that a user can view the next image by swiping left. If only a portion of the image is shown, a swipe gesture translates the visible part according to the gesture's direction, which varies based on whether the full image or a zoomed-in section is displayed. There is no requirement in the claim language for the “two-dimensional screen translation heuristic” and “next item heuristic” to be applied in the same context. Therefore, the district court's limitation of the corresponding structure to a finger tap on the right side of the screen should be reversed, along with the summary judgment of non-infringement for products using a "swipe" gesture as a next item heuristic.

Regarding expert testimony, there is agreement with the decision to reverse the district court's exclusion of Apple’s expert, Brian Napper, due to its incorrect claim construction. The district court criticized Napper for not isolating the consumer value of the “tap for next item” functionality and for overlooking alternative royalties. The majority found the district court's categorization of the “next item heuristic” as a means-plus-function claim limiting it to a right-side tap to be erroneous. While affirming that the “heuristic” claims should be viewed as means-plus-function limitations, there is disagreement on limiting the corresponding structure to a right-side tap. The district court's exclusion of Napper's testimony was not seen as an abuse of discretion, as it deemed his reliance on the Magic Trackpad for valuation to be unreliable due to insufficient comparability.

Napper's proposed testimony was deemed unreliable for determining the value of the vertical scrolling feature of a product, as it failed to provide evidence of consumer willingness to pay to avoid unsuccessful swiping. The court noted that while precedent allows for using the cost of benchmark products to assess infringement value, the Magic Trackpad, which Apple acknowledges lacks the functionality claimed in the '949 patent, served as a questionable starting point for Napper's analysis. Consequently, the discounts Napper applied to estimate value appeared arbitrary, leading the court to uphold the district court's discretion in excluding Napper's testimony.

Regarding Apple's request for injunctive relief, the district court granted summary judgment denying Apple an injunction for infringement of the '949, '263, and '647 patents. Although the majority vacated this decision due to a reversal in claim construction for the '949 patent, the dissenting opinion argued for affirming the summary judgment for all three patents. The dissent emphasized that Apple must demonstrate a causal link between Motorola's infringement and its alleged irreparable harm, specifically showing that lost market share and downstream sales resulted directly from the infringing features. The district court found that Apple failed to provide evidence establishing this connection.

Apple's argument regarding potential infringement does not establish that such infringement harmed its sales, market share, or consumer goodwill. The district court noted that Apple's claims about Motorola's phones mimicking the iPhone do not constitute a legally recognized harm separate from competition. Apple asserts that it has a policy against licensing the three asserted patents, which is relevant to determining irreparable harm, yet there are no precedents linking a licensing policy directly to proving irreparable harm. 

Apple's evidence, including consumer surveys, fails to demonstrate that the features claimed in the patents specifically drive consumer demand for Motorola’s products. For the ’949 patent, the surveys indicate a general preference for superior touchscreen interfaces but do not relate to specific patented features. Testimonies from Motorola executives about consumer preferences are deemed insufficient for establishing a causal link between patent infringement and market harm. Similarly, for the ’263 patent, while Apple cites studies showing its role in the success of its devices, there is no evidence that these factors drive demand for Motorola's products. Furthermore, evidence of Motorola's beliefs regarding the benefits of using Apple's technology does not establish market harm attributable to infringement.

Evidence presented by Apple is inadequate to demonstrate a causal link between Motorola’s inclusion of the allegedly infringing feature and the harm Apple claims to have suffered. Specifically regarding the ’647 patent, although Apple references Motorola's acknowledgment of the patent’s "structure-detection feature" as a key and differentiating aspect, and cites a newspaper article characterizing it as an appealing new feature in Apple products, this does not establish the necessary nexus. Consequently, the district court's determination that Apple did not raise a genuine issue regarding whether these features drive consumer demand for Motorola’s products is upheld. Therefore, Apple fails to prove that Motorola’s infringement has caused it irreparable harm, which means it does not satisfy the eBay standard for obtaining injunctive relief.

Regarding Motorola's request for injunctive relief related to the ’898 patent, while there is agreement on the denial of an injunction, there is dissent concerning the majority's implication that an alleged infringer's refusal to negotiate a license warrants an injunction post-infringement finding. It is emphasized that there should not be a blanket rule preventing injunctions for FRAND-committed patents, but rather that FRAND commitments should be integrated into the eBay analysis. While it is accepted that an injunction may be justified if an alleged infringer unreasonably delays negotiations or refuses a FRAND royalty, there is opposition to the notion that merely refusing to negotiate should justify an injunction. It is argued that an infringer has the right to contest the validity of a FRAND-committed patent without facing punitive measures for slow negotiations.

An alleged infringer may opt to pay a FRAND (Fair, Reasonable, and Non-Discriminatory) license to avoid the costs of litigation, including potential treble damages, attorney’s fees, and the risk of higher damage awards than the FRAND rates if found liable for willful infringement. Courts may award damages exceeding a reasonable royalty to address bad faith negotiations by the infringer, compensating the patentee for harm caused. Despite these considerations, monetary damages are typically deemed sufficient for a FRAND patentee's injuries, suggesting that pre-litigation conduct should not influence the availability of injunctive relief. Injunctions may be warranted if a patentee cannot collect court-ordered damages due to the infringer being judgment-proof or refusing to pay, but these circumstances do not apply in this case. Consequently, the court supports the district court's finding that Motorola did not demonstrate irreparable harm or inadequate damages, affirming the denial of Motorola's request for injunctive relief concerning the '898 patent. The author concurs in part and dissents in part.