U.S. Court of Appeals for the Federal Circuit, In Re: Google LLC, Docket No. 19-126

Venue

Jurisdiction

Patent Infringement

Place of Business:

-       Real Property Ownership

-       Leasehold Interest

-       Possession

-       Control

-       Local Caches for Data

-       Internet Service Providers (ISPs)’ Datacenter

-       Rack Space Occupied by the GGC Servers

-       Physical Presence of an Employee or Other Agent of the Defendant

-       Agency Relationship

-       Right of Interim Control

-       Maintaining Equipment

-       Production, Storage, Transport, and Exchange of Goods or Services

Writ of Mandamus

Writ – All Writs Act

Contract Drafting

On Petition for Writ of Mandamus to the United States District Court for the Eastern District of Texas in Nos. 2:18-cv-00462-JRG, 2:18-cv-00463-JRG

Google LLC (“Google”) petitions for a writ of mandamus ordering the United States District Court for the Eastern District of Texas to dismiss the case for lack of venue. See Super Interconnect Techs. LLC v. Google LLC, No. 2:18- CV-00463-JRG, 2019 U.S. Dist. LEXIS 132005 (E.D. Tex. Aug. 7, 2019). We hold that mandamus is warranted and order that the case either be dismissed or transferred.

Super Interconnect Technologies LLC (“SIT”) sued Google for patent infringement in the Eastern District of Texas. Under the patent venue statute, 28 U.S.C. § 1400(b), “any civil action for patent infringement may be brought in the judicial district where the defendant resides, or where the defendant has committed acts of infringement and has a regular and established place of business.” SIT filed its suit after the Supreme Court’s decision in TC Heartland LLC v. Kraft Foods Group Brands LLC, 137 S. Ct. 1514, 1517 (2017), which held that “a domestic corporation ‘resides’ only in its State of incorporation for purposes of the patent venue statute,” and this court’s decision in In re Cray, Inc., 871 F.3d 1355, 1360 (Fed. Cir. 2017), which held that a “regular and established place of business” under the patent venue statute must be: (1) “a physical place in the district”; (2) “regular and established”; and (3) “the place of the defendant.”

(…) SIT’s allegation of venue was based on the presence of several Google Global Cache (“GGC”) servers, which function as local caches for Google’s data.

The GGC servers are not hosted within datacenters owned by Google. Instead, Google contracts with internet service providers (ISPs) within the district to host Google’s GGC servers within the ISP’s datacenter. When a user requests Google’s content, the ISP attempts to route the user’s request to a GGC server within its own network (within the district) before routing the request to Google’s central data storage servers (outside the district). The GGC servers cache only a small portion of content that is popular with nearby users but can serve that content at lower latency—which translates to shorter wait times— than Google’s central server infrastructure. This performance benefit is in part due to the physical proximity of the GGC servers to the ISP’s users. This arrangement allows Google to save on bandwidth costs and improve user experience on its various platforms.

At the time of the complaint, Google had entered into contracts with two ISPs to host GGC servers owned by Google in the Eastern District of Texas: Cable One Inc. (“Cable One”) and Suddenlink Communications (“Suddenlink”). The contracts provided that the ISPs would host Google’s GGC servers in their data centers. Specifically, the GGC servers are installed in the ISP’s server racks, which are cabinets that accept standard server components. Each contract states that the ISP must provide “rack space, power, network interfaces, and IP addresses,” for the GGC servers, and provide “network access between the GGC servers and the ISP’s network subscribers.” Supplemental Record, Dkt. 31, Ex. A, at 1; id., Ex. B, at 1. The contracts permit the ISPs to select the rack space for the GGC servers, but they tightly restrict the ISPs’ ability to relocate the servers without Google’s permission once a location is selected. Id., Ex. A, at 2; id., Ex. B at 2. The contracts also strictly limit any unauthorized access to the space used by Google’s servers. Id., Ex. A, at 6–7; id., Ex. B, at 5. The contracts state that the ISPs are required to provide “installation services,” i.e., installing the GGC servers in the server racks. Id., Ex. A, at 1; id., Ex. B at 1. While the contracts forbid the ISPs to “access, use, or dispose of” the GGC servers without Google’s permission, id., Ex. A, at 2; id., Ex. B at 2, they also require the ISPs to provide “remote assistance services,” which “involve basic maintenance activities” performed on the GGC servers by the ISP’s onsite technician, if requested by Google, id., Ex. A, at 1, 6; id., Ex. B, at 1, 5. It is undisputed that no Google employee performed installation of, performed maintenance on, or physically accessed any of the GGC servers hosted by Cable One or Suddenlink.

Google moved to dismiss the complaint for improper venue under 28 U.S.C. § 1406(a) and Federal Rule of Civil Procedure 12(b)(3). The district court denied Google’s motion (…) found that the GGC servers qualified as Google’s “regular and established place of business”.

Google now petitions for a writ for mandamus directing the district court to dismiss the case for lack of venue under § 1400(b). Acushnet and 17 other companies filed an amicus brief in support of Google’s petition. This court heard oral argument on December 13, 2019.

This court “may issue all writs necessary or appropriate in aid of its jurisdiction and agreeable to the usages and principles of law” under the All Writs Act. 28 U.S.C. § 1651(a). The Supreme Court has held that three conditions must be met before a writ may issue: (1) the petitioner “must have no other adequate means to attain . . . relief,” (2) the petitioner must show that the right to mandamus is “clear and indisputable,” and (3) the court must be “satisfied that the writ is appropriate under the circumstances.” Cheney v. U.S. Dist. Court for D.C., 542 U.S. 367, 380 (2004) (first alteration in original).

(…) This court has not addressed this fundamental and recurring issue of patent law. We thus conclude that mandamus is an available remedy.

Under Cray, there are three general requirements to establishing that the defendant has a regular and established place of business: “(1) there must be a physical place in the district; (2) it must be a regular and established place of business; and (3) it must be the place of the defendant.” 871 F.3d at 1360.

The first question is whether the rack space occupied by the GGC servers constitutes a “place” under § 1400(b) as interpreted in Cray. As the court in Cray emphasized, “the first requirement [under § 1400(b)] is that there ‘must be a physical place in the district.’” 871 F.3d at 1362. A “place” merely needs to be a “physical, geographical location in the district from which the business of the defendant is carried out.” Id.

Google’s petition suggests that a court’s inquiry into whether the defendant has a physical “place of business” should focus on whether the defendant has real property ownership or a leasehold interest in real property. We hold that a “place” need not have such attributes. In Cray, we rejected the notion that a “virtual space” or “electronic communications from one person to another” could constitute a regular and established place of business. 871 F.3d at 1362. Here, the GGC servers are physically located in the district in a fixed, geographic location. Indeed, Cray itself recognized that a “place of business” is not restricted to real property that the defendant must “own or lease,” and that the statute could be satisfied by any physical place that the defendant could “possess or control.” Id. at 1363 (discussing the third Cray factor). (…) see also In re Cordis Corp., 769 F.2d 733, 735, 737 (Fed. Cir. 1985) (suggesting that defendant’s employees’ homes, which were used to store the defendant’s “literature, documents and products,” could constitute a “regular and established place of business”).

(…) We agree, however, with Google’s alternative argument that under the second Cray factor, a “place of business” generally requires an employee or agent of the defendant to be conducting business at that place. This is apparent from the service statute for patent cases, now codified at 28 U.S.C. § 1694.

We conclude that a “regular and established place of business” requires the regular, physical presence of an employee or other agent of the defendant conducting the defendant’s business at the alleged “place of business.”

The question then is whether Google had an employee or agent with a regular, physical presence at its “place of business” and whether that employee or agent was conducting Google’s business. The record is clear that there is no Google employee conducting business in the Eastern District of Texas. However, there is nonetheless the question of whether the ISPs are acting as Google’s agent.

An agency relationship is a “fiduciary relationship that arises when one person (a ‘principal’) manifests assent to another person (an ‘agent’) that the agent shall act on the principal’s behalf and subject to the principal’s control, and the agent manifests assent or otherwise consents to act.” Restatement (Third) of Agency § 1.01. The essential elements of agency are (1) the principal’s “right to direct or control” the agent’s actions, (2) “the manifestation of consent by the principal to the agent that the agent shall act on his behalf,” and (3) the “consent by the agent to act.” Meyer v. Holley, 537 U.S. 280, 286 (2003).

Google contracted with two ISPs, Cable One and Suddenlink, to host its GGC servers. The contracts stated that, for each ISP, Google would provide the ISP with GGC server equipment, which the ISP would install and host in server racks within its datacenter. The contracts contemplated that the ISP would perform three functions.

First, the ISP provides the GGC servers with network access, i.e., a connection to the ISP’s customers, as well as the public Internet. The ISP provides Google with a service, and Google has no right of interim control over the ISP’s provision of network access beyond requiring that the ISP maintain network access to the GGC servers and allow the GGC servers to use certain ports for inbound and outbound network traffic. In this respect, the ISPs are not agents of Google. See Restatement (Third) of Agency § 1.01 cmt. f(1) (“The power to give interim instructions distinguishes principals in agency relationships from those who contract to receive services provided by persons who are not agents.”).

Second, the ISP performs installation of the GGC servers. The contracts with the ISPs stated that the ISP was responsible for the installation of the GGC servers, including “coordination with logistics and shipping personnel; inventory of equipment received; unpacking equipment; assembling equipment based on information and instructions provided by Google; . . . connecting equipment to power strip(s) and Ethernet cable(s); and powering up equipment & executing installation scripts configuring IP address information.” Supplemental Record, Ex. A at 6; id., Ex. B at 5. Although these provisions may be suggestive of an agency relationship, we do not consider the ISPs performing these installation functions to be conducting Google’s business within the meaning of the statute. The installation activity does not constitute the conduct of a “regular and established” business, since it is a one-time event for each server.

Third, the contracts provide that “Google may from time to time request that the ISP perform certain services” involving “basic maintenance activities” with respect to the GGC servers. Id., Ex. A at 6; id., Ex. B at 5. The contracts provided examples of these activities:

Physical switching of a toggle switch; power cycling equipment . . . ; remote visual observations and/or verbal reports to Google on its specific collocation [sic] cabinet(s) for environment status, display lights, or terminal display information; labeling and dress-up of cabling within cabinet; tightening screws, cable ties, or securing cabling to mechanical connections, plugs; replacing existing plug-in only hardware such as circuit cards with spares or upgrades.

Id., Ex. A at 6; id., Ex. B at 5. The ISP’s conduct as to these activities is permitted “only with specific and direct step-by-step instructions from Google.” Id., Ex. A at 6; id., Ex. B at 5. The ISP is also prohibited from “accessing, using, or disposing of the GGC servers, in whole or in part” without Google’s prior written consent. Id., Ex. A at 2; see also Ex. B at 2.

Although the maintenance provision, like the provision on installation, may be suggestive of an agency relationship, SIT has not established that the ISPs performing the specified maintenance functions are conducting Google’s business within the meaning of the statute. The better reading of the statute is that the maintenance activities cannot, standing alone, be considered the conduct of Google’s business.

Maintaining equipment is meaningfully different from—as only ancillary to—the actual producing, storing, and furnishing to customers of what the business offers. In 1897, Congress focused on the latter sorts of activities as the conduct of business. See 29 Cong. Rec. 1900 (1897) (statement of Rep. Lacey) (discussing venue in the context of agents performing traditional business functions, such as manufacturing, sales, or direct customer services); id. at 1902 (discussing similarities to a law conferring “jurisdiction” to sue agents of an insurance company). There is no suggestion in the legislative history that maintenance functions that existed at the time, such as the maintenance of railways or telegraph lines, constituted “conducting the defendant’s business” within the meaning of the statute. See id. at 1900–02.

(…) Clear intent of Congress in enacting the statute to restrict venue to where the defendant resides or is conducting business at a regular and established place of business, with agents there regularly conducting that business, lead us to our conclusion. The venue statute should be read to exclude agents’ activities, such as maintenance, that are merely connected to, but do not themselves constitute, the defendant’s conduct of business in the sense of production, storage, transport, and exchange of goods or services.

We conclude that the Eastern District of Texas was not a proper venue because Google lacked a “regular and established place of business” within the district since it has no employee or agent regularly conducting its business at its alleged “place of business” within the district.

To be clear, we do not hold today that a “regular and established place of business” will always require the regular presence of a human agent, that is, whether a machine could be an “agent.” Such a theory would require recognition that service could be made on a machine pursuant to 28 U.S.C. § 1694. Nor do we decide what might be inferred in this respect from Congress’ amendment to the venue statute in the AIA concerning automated teller machines. See AIA § 18(c).

(U.S. Court of Appeals for the Federal Circuit, February 13, 2020, In Re: Google LLC, Docket No. 19-126, Circuit Judge Dyk)