DCT

2:24-cv-00130

NTECH Properties Inc v. ByteDance Ltd

Log In

I. Executive Summary and Procedural Information

  • Parties & Counsel:
  • Case Identification: 2:24-cv-00130, E.D. Tex., 05/24/2024
  • Venue Allegations: Plaintiff alleges venue is proper in the Eastern District of Texas because Defendants are foreign corporations, and under 28 U.S.C. § 1391(c)(3), venue is proper against such defendants in any judicial district.
  • Core Dispute: Plaintiff alleges that Defendants’ TikTok social media platform, including its mobile application, website, and supporting server infrastructure, infringes six patents related to systems and methods for generating and delivering personalized and customized media programming.
  • Technical Context: The technology at issue concerns the server-side architecture and logic for automatically curating and assembling streams of media content tailored to individual user data, preferences, and interactions, a foundational technology for modern content recommendation engines.
  • Key Procedural History: The complaint alleges that Plaintiff provided Defendants with direct notice of the asserted patents and their alleged infringement via a letter dated February 22, 2024, which may form the basis for a claim of willful infringement from that date forward.

Case Timeline

Date Event
2003-06-24 Earliest Priority Date ('185', '597' Patents)
2007-06-13 Earliest Priority Date ('704', '261', '947', '753' Patents)
2012-03-27 '704 Patent Issued
2014-10-28 '185 Patent Issued
2014-11-11 '753 Patent Issued
2015-12-15 '261 Patent Issued
2016-04-19 '597 Patent Issued
2017-01-01 Accused Technologies Launched
2018-03-20 '947 Patent Issued
2024-02-22 Plaintiff Sent Notice Letter to Defendants
2024-05-24 Complaint Filed

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 8,145,704

  • Patent Identification: U.S. Patent No. 8,145,704, titled “Method and system for providing media programming,” issued on March 27, 2012 (the “’704 Patent”).

The Invention Explained

  • Problem Addressed: The patent describes a media landscape where content is generated from libraries of media elements (e.g., video or audio clips) that are tagged and assembled into programming using templates, but suggests a need for improved methods of providing this programming to users (’704 Patent, col. 1:20-29).
  • The Patented Solution: The patent discloses a three-party system architecture to create personalized media streams (’704 Patent, Fig. 1). An "aggregator" first presents a list of content "feeds" to a user. The user's selection, along with user-specific data, is sent from the aggregator to a separate "publisher." The publisher then uses this information to assemble a customized media stream from its own elements and provides it to the user (’704 Patent, Abstract). This architecture separates the initial user interface and selection (aggregator) from the content assembly and personalization logic (publisher).
  • Technical Importance: This approach describes a foundational client-server model for moving beyond static, one-to-many content delivery toward dynamic, personalized media streams based on user choice and data (’704 Patent, col. 2:53-64).

Key Claims at a Glance

  • The complaint asserts independent system claim 19 (Compl. ¶52).
  • The essential elements of Claim 19 are:
    • An aggregator having a processor adapted for presenting a plurality of media streams, receiving a signal indicative of a user's selection, and providing an output signal containing data on the selection and the user.
    • A publisher having a processor adapted to receive the output signal from the aggregator, use an algorithm with the selection and user data to determine a temporal sequence of media programming, and assemble the selected media stream with other media elements into that sequence.

U.S. Patent No. 9,215,261

  • Patent Identification: U.S. Patent No. 9,215,261, titled “Method and system for providing media programming,” issued on December 15, 2015 (the “’261 Patent”).

The Invention Explained

  • Problem Addressed: The invention addresses the generation of intuitive and personalized media programming, aiming to improve upon conventional methods that lacked the ability to automatically curate content tailored to individual users in real time (Compl. ¶¶ 32-33).
  • The Patented Solution: The patent claims a method where an aggregator delivers a menu of media feeds to a user's device. Each feed comprises media elements that are explicitly "tagged in at least one library" associated with a publisher (’261 Patent, col. 15:58-64). After the user makes a selection, the aggregator provides the selection ID and client information to the publisher and initiates the display of a customized program, where media elements are selected from the tagged libraries and ordered based on the user's information and feed choice (’261 Patent, Abstract).
  • Technical Importance: The method's focus on "tagged" media elements suggests a system where content is indexed with specific metadata to facilitate automated, rule-based assembly into personalized streams based on user preferences (Compl. ¶32).

Key Claims at a Glance

  • The complaint asserts independent method claim 1 (Compl. ¶74).
  • The essential elements of Claim 1 are:
    • Delivering by an aggregator a selectable menu of media feeds, where each media element is tagged in at least one library.
    • Receiving a selection from the client device by the aggregator.
    • Providing an identification of the selected feed and client information to a publisher by the aggregator.
    • Initiating, by the aggregator, the display of customized media programming comprising elements selected from the libraries that correspond to the feed and are responsive to the user information.

Multi-Patent Capsule: U.S. Patent No. 9,923,947

  • Patent Identification: U.S. Patent No. 9,923,947, titled “Method and system for providing media programming,” issued March 20, 2018 (the “’947 Patent”).
  • Technology Synopsis: This patent claims a method for generating customized media programming where a publisher analyzes user information received from an aggregator. Based on this analysis, the publisher inserts an "element" into the media programming based on targeting criteria such as positioning, demographics, or geo-targeting, and then dynamically updates that element based on additional user information (Compl. ¶97).
  • Asserted Claims: Independent method claim 1 (Compl. ¶94).
  • Accused Features: The complaint alleges that TikTok’s recommendation engine infringes by analyzing user data (including positioning, demographic, and psychographic information) to generate and dynamically update customized media feeds (Compl. ¶¶ 103-104).

Multi-Patent Capsule: U.S. Patent No. 8,886,753

  • Patent Identification: U.S. Patent No. 8,886,753, titled “Method and system for providing media programming,” issued November 11, 2014 (the “’753 Patent”).
  • Technology Synopsis: This patent claims a method where an aggregator delivers a list of feeds to a client, receives a selection, and a publisher then acquires that feed selection from the aggregator. The publisher collects client information from the aggregator and provides the client with customized media programming based on a "publisher-initiated feed" responsive to the user's selection and information (Compl. ¶119).
  • Asserted Claims: Independent method claim 1 (Compl. ¶116).
  • Accused Features: The complaint accuses TikTok’s system of practicing this method, identifying front-end servers as the aggregator and algorithm servers as the publisher that generates customized "For You" feeds (Compl. ¶¶ 120-122).

Multi-Patent Capsule: U.S. Patent No. 9,317,597

  • Patent Identification: U.S. Patent No. 9,317,597, titled “Method and apparatus for efficient, entertaining information delivery,” issued April 19, 2016 (the “’597 Patent”).
  • Technology Synopsis: This patent describes a method of providing video programming where a server receives a user request, searches databases of video clips, and identifies both responsive clips and associated "transitional clips." The system then uses "templates" comprising variables (e.g., user information, style requirements) to define and concatenate a sequence of first video clips and transitional clips for presentation. A subsequent search can be initiated based on a user's selection of a clip from the first program (Compl. ¶137).
  • Asserted Claims: Independent method claim 1 (Compl. ¶134).
  • Accused Features: The complaint alleges TikTok's system uses a request-and-response architecture where its servers search video databases and use algorithms (templates) with user information to generate sequences of video clips for user feeds (Compl. ¶¶ 139, 144).

Multi-Patent Capsule: U.S. Patent No. 8,875,185

  • Patent Identification: U.S. Patent No. 8,875,185, titled “Method and apparatus for efficient, entertaining information delivery,” issued October 28, 2014 (the “’185 Patent”).
  • Technology Synopsis: This patent claims a method similar to the ’597 Patent, but adds a spatial display element. It describes presenting a first video program in a "first area of a display" while concurrently presenting frames from other video clips in a "second area of said display." A user can then make a selection from the second area to trigger the generation and presentation of a second video program (Compl. ¶165).
  • Asserted Claims: Independent method claim 1 (Compl. ¶162).
  • Accused Features: The complaint maps this to TikTok’s user interface, alleging that the "Following" and "For You" feeds function as the first and second display areas, respectively, where content is sequenced using templates and user data (Compl. ¶¶ 167, 172).

III. The Accused Instrumentality

Product Identification

  • The "Accused Technologies" are identified as the TikTok App, the TikTok Website, and the underlying servers and software that support the platform's functionality (Compl. ¶42).

Functionality and Market Context

  • The complaint describes the Accused Technologies as a video-sharing social networking platform that allows users to create, share, and watch short-form videos (Compl. ¶43). A core accused functionality is the platform's recommendation algorithm, which generates personalized video streams, or "feeds," such as the "For You" feed (Compl. ¶45). These feeds are assembled based on algorithmic analysis of user information, viewing history, location, and interactions with the platform (Compl. ¶44). The complaint provides a screenshot of the TikTok "For You" feed, which shows a personalized stream of videos that the app predicts the user will enjoy (Compl. p. 15). The platform is alleged to use metadata to associate video clips and to include display features like "Animated Thumbnails" (Compl. ¶¶ 46-47).

IV. Analysis of Infringement Allegations

’704 Patent Infringement Allegations

Claim Element (from Independent Claim 19) Alleged Infringing Functionality Complaint Citation Patent Citation
an aggregator having a memory and a processor, the processor adapted for presenting a plurality of media streams to a user... Defendants’ front-end servers that support the TikTok app and website, which present media streams such as trending hashtags to users. ¶56, ¶57 col. 2:50-53
receiving a signal having data indicative of a selection of one of the media streams... The front-end servers receive signals when a user performs an action, such as creating a playlist from available videos. ¶58 col. 2:53-55
and providing an output signal including data indicative of the selection and of user data... Front-end servers send an output signal with the user's selection and data to TikTok's algorithm servers. ¶58 col. 2:55-57
and a publisher having a memory and a processor... Defendants’ back-end "algorithm servers" that operate TikTok's Recommendation Engine. ¶58, ¶60 col. 2:57-58
the publisher processor adapted to receive the output signal from the aggregator, and responsive to the data indicative of the selection and of user data, access from the memory at least one algorithm for selection and sequencing of media elements... The Recommendation Engine receives user selection and data, and uses algorithms to filter and select media elements based on the user's profile, actions, and trending topics. ¶60 col. 2:58-64
select a temporal sequence of media programming... [and] assemble the selected media stream and the other media elements in the temporal sequence. The Recommendation Engine assembles the selected video clips into a temporal sequence, such as the queue of videos presented in the user's "For You" feed. ¶60 col. 15:13-18
  • Identified Points of Contention:
    • Scope Questions: A central question may be whether TikTok's distributed, cloud-based server architecture can be separated into the distinct "aggregator" and "publisher" entities as required by the claim. The complaint alleges front-end servers are aggregators and algorithm servers are publishers (Compl. ¶¶ 56, 58, 60). A defense may argue that this is an artificial distinction and that TikTok operates as a single, integrated system not contemplated by the patent's architecture.
    • Technical Questions: The complaint alleges that when a user creates a playlist, this constitutes "receiving a signal having data indicative of a selection of one of the media streams" (Compl. ¶58). It raises the question of whether selecting individual videos for a playlist is equivalent to selecting a "media stream" from a presented plurality of streams, as the claim language requires. The complaint includes a visual of the "Add to playlist" screen, showing a user selecting individual public videos (Compl. p. 22).

’261 Patent Infringement Allegations

Claim Element (from Independent Claim 1) Alleged Infringing Functionality Complaint Citation Patent Citation
delivering by an aggregator, to a client device associated with a user, a selectable menu of media feeds... TikTok's servers (acting as aggregators) deliver a menu of media feeds, such as the trending hashtags page, to a user's device. ¶78 col. 15:58-61
wherein each said media feed comprises a plurality of media elements... and each said media element is tagged in at least one library in communication with at least one publisher... Media feeds are made up of media elements (audio, visual), and the "tagged" element is met by TikTok's music library, which contains tagged music available to users. ¶79 col. 15:61-64
receiving by the aggregator from the client device a selection identifying at least one of said media feeds; The aggregator servers receive a selection when a user searches a hashtag or creates a playlist. ¶80 col. 16:1-3
providing by the aggregator, to said publisher, an identification of each selected media feed; The complaint’s theory implies that aggregator servers provide feed selection data (e.g., the chosen hashtag) to the publisher/algorithm servers. ¶80 col. 16:4-5
providing by the aggregator, to said publisher, client information comprising information about the user and information about the associated client device; The aggregator servers provide user and device information to the publisher/algorithm servers to inform the customized programming. ¶81 col. 16:6-9
and initiating, by the aggregator, for display at the client device, individual customized media programming... wherein the selected media elements... correspond to at least one selected media feed and are responsive to said user information... The algorithm generates a sequenced media program (e.g., the "For You" feed) where the video elements shown correspond to user selections and are ordered based on user information. ¶81, ¶82 col. 16:10-18
  • Identified Points of Contention:
    • Scope Questions: The complaint's theory for the "tagged in at least one library" limitation relies on TikTok’s commercial music library (Compl. ¶79). This raises the question of whether the term "tagged" in the patent's context requires a specific technical structure for automated media assembly, or if it can be read broadly to cover general metadata (e.g., artist, genre) in a commercial music catalog.
    • Technical Questions: What evidence does the complaint provide that the "aggregator" (front-end servers) performs the step of "initiating" the customized programming? The claim requires the aggregator to perform this step, but the complaint's description suggests the "publisher" (algorithm servers) generates the final feed (Compl. ¶81). The precise sequence of operations and control between the alleged system components will be a key factual issue. The complaint includes a visual of the TikTok interface showing trending hashtags, which it alleges is a selectable menu of media feeds (Compl. p. 31).

V. Key Claim Terms for Construction

For the ’704 and ’261 Patents

  • The Term: "aggregator"

  • Context and Importance: This term is foundational to the claimed two-part architecture. Its definition will determine whether TikTok's allegedly integrated platform can be mapped onto the patent's structure. Practitioners may focus on whether "aggregator" requires a structurally separate entity from the "publisher" or if it can refer to a distinct software function within a unitary system.

  • Intrinsic Evidence for Interpretation:

    • Evidence for a Broader Interpretation: The specification describes an aggregator broadly as one of "one or more sources of programming," including "web servers providing websites," an "electronic program guide," or even "software incorporated locally on a user's system" (’704 Patent, col. 3:6-12). This language may support a functional, rather than strictly structural, definition.
    • Evidence for a Narrower Interpretation: Figure 1 of the ’704 Patent depicts the "aggregator" (100) as a distinct block that communicates with, but is separate from, the "publisher" (110). This visual separation may support an argument that the two must be distinct components.

  • The Term: "publisher"

  • Context and Importance: As the counterpart to the "aggregator," the definition of "publisher" is critical to establishing the claimed architecture. The dispute will likely concern the nature of the interaction between the two, specifically whether the handoff of "selection and user data" requires a formal, discrete communication between separate systems.

  • Intrinsic Evidence for Interpretation:

    • Evidence for a Broader Interpretation: The patent states that feed generation could occur in a set-top box, with software running on the box to assemble media elements (’704 Patent, col. 5:16-21). This suggests the aggregator and publisher functions could be co-located on the same hardware, supporting a more flexible definition.
    • Evidence for a Narrower Interpretation: Claim 19 of the ’704 Patent requires the publisher processor to "receive the output signal from the aggregator." This language suggests a directional, sender-receiver relationship that could be argued to imply two distinct entities.

  • The Term: "tagged" (’261 Patent)

  • Context and Importance: This term appears to be a key distinguishing feature of the ’261 Patent's claims. The infringement allegation relies on a "commercial music library" meeting this limitation (Compl. ¶79). The construction will determine if standard metadata suffices or if a more specific technical implementation is required.

  • Intrinsic Evidence for Interpretation:

    • Evidence for a Broader Interpretation: The patent does not appear to provide an explicit definition of "tagged," which may support giving the term its plain and ordinary meaning, potentially encompassing any form of metadata association.
    • Evidence for a Narrower Interpretation: The specification discusses media assembly engines that use tags and templates to generate programming (’261 Patent, col. 1:25-33). A defendant may argue that, in this context, "tagged" implies metadata specifically structured for use by such an automated assembly engine, rather than general-purpose descriptive metadata found in a music library.

VI. Other Allegations

  • Indirect Infringement: The complaint alleges induced infringement, stating Defendants instruct and encourage users to infringe through online "Help Center" articles that explain how to use accused features like managing the "For You" feed and using hashtags (Compl. ¶¶ 68, 89). Contributory infringement is also alleged on the basis that the TikTok platform is specialized for the provision of entertaining media and is not a staple article of commerce (Compl. ¶70).
  • Willful Infringement: The complaint alleges that Defendants had knowledge of the asserted patents and their infringement at least as of February 22, 2024, the date Plaintiff sent a notice letter (Compl. ¶50). This allegation forms the basis for a claim of post-notice willful infringement.

VII. Analyst’s Conclusion: Key Questions for the Case

  • A core issue will be one of architectural mapping: can the "aggregator" and "publisher" entities, described in the patents as distinct functional blocks, be mapped onto the components of TikTok’s modern, integrated cloud-based infrastructure? The case may depend on whether the alleged separation between front-end and back-end algorithm servers is a functional reality that satisfies the claim limitations.
  • A second central issue will be one of definitional scope: can patent terms rooted in the context of early-2000s media systems, such as "tagged," be construed to cover the features of the accused platform? The viability of the infringement theory for the ’261 Patent, for instance, may turn on whether metadata in a commercial music library constitutes a "tagged" library as contemplated by the patent.
  • A key evidentiary question will be one of functional control and operation: which entity—the alleged "aggregator" or "publisher"—performs the specific actions required by the method claims? For example, the dispute over the ’261 Patent may focus on factual evidence showing whether it is the aggregator that "initiates" the customized programming for display, or if that function is performed exclusively by the publisher.