United States
Applying US patents beyond the US border in software and networking


Pavan K Agarwal
Foley & Lardner LLP

This chapter looks at the application of US patents outside the US border, focusing on the areas of software and telecommunications.

Patents are territorial and limited to the country of issuance – right?

Patent rights apply territorially, carrying force in their country of issuance. US patent laws cover particular types of activities occurring in or otherwise affecting the United States. In 1857 the US Supreme Court declared that US patent laws “do not, and were not intended to, operate beyond the limits of the United States” (Brown v Duchesne, 60 US (19 How) 183 (1857)). A patent holder cannot normally assert US patents against another company for activities occurring only outside the United States. Companies exploit this territorial limitation to avoid US patents by moving certain activities offshore, although they often continue to reap the profits from the US market and from US involvement.

While geographical limits on enforcement can be easily applied to certain technologies (eg, manufacturing widgets), the situation becomes more complex for other technologies. Borders have become essentially transparent for technologies such as software and networking. Networking often involves communication between numerous computers located over several different networks, where servers may be located in different countries. Software is readily transmitted in various forms and between various points, so tracking transmission of a particular copy of software can be difficult. Software can be transmitted by disk or by up/downloading copies between a computer in one location and a server in a different country. The multiple locations and ease of transmissibility of these technologies complicate the application of the territoriality limits of US patents.

In three recent high-profile cases the US Court of Appeals for the Federal Circuit (often the court of last resort in US patent cases) has confirmed the expansion of US patent law to cover activities that occur in part outside the United States. Therefore, companies avoiding US patents by locating specific facilities outside the United States need to re-evaluate their decisions. Companies whose primary activities exist outside the United States, but which have components of their operations taking place within the United States, may find themselves at greater risk of allegations of infringement of US patents.

While the most direct impact of the recent case law appears to be on telecommunications and software companies, these cases potentially affect companies in all industries. The prevailing nature of computers and data transmission – with so many companies in so many industries participating at some level in creating communication paths and using specialised software codes – means that the potential impact is widespread rather than being confined to certain groups of companies.

Recent decisions expanding the territorial effect of US patent laws

Eolas v Microsoft

Involving a US$500 million-plus trial verdict, Eolas v Microsoft highlights the potential impact of exporting a single copy of software where other copies are made outside the United States. Microsoft exports its Windows operating system on a golden master disk to original equipment manufacturers (OEMs). Microsoft licenses the OEMs to make copies of the golden master disk to place Windows operating systems on each computer. The OEMs may sell the computers either back into the United States or outside the United States. The golden master disk itself never becomes a part of the computers sold by the OEMs – only the copies of the Windows operating system, all made outside the United States, reside on the computers.

US patent law states that if a party supplies all or a substantial portion of components from the United States in a manner to induce actively combining them into a patented invention outside the United States, where the patent covers the combination the exporter can be liable for US patent infringement.

Yet this US patent law provision has traditionally been applied to physical components exported from the United States with the intention of combining them outside the United States. It was enacted to prevent manufacturers from producing parts in the United States and simply offshoring the assembly of the components. However, in this case Microsoft exported only one copy and the OEMs made multiple copies overseas, as licensed by Microsoft. Could Microsoft potentially be liable for all the copies it sold to the OEMs, no matter where the computers were sold? In Eolas the Federal Circuit said yes. The court found that “the software code on the golden master disk is not only a component, it is probably the key part of this patented invention”. The Federal Circuit found a close analogy between software and hardware in function, stating that: “on a functioning computer, software morphs into hardware and vice versa at the touch of a button. In other words, software converts its functioning code into hardware and vice versa.” The court determined that providing a single copy of the golden master disk with the intention that it be copied overseas to make multiple copies meant that Microsoft could be liable for both domestic and non-domestic sales of Windows.

AT&T v Microsoft

AT&T also sued Microsoft regarding the Windows operating system. As in the Eolas Case, AT&T sought damages based on Microsoft’s non-US sales of the Windows operating system, which was copied outside of the United States from the golden master disk exported from the United States.

Microsoft raised an additional argument in AT&T: that the patent law provision at hand required it to have supplied the component from the United States. It argued that it had only supplied one copy – the golden master disk itself from the United States.

The Federal Circuit focused on the technology at issue – the software – and interpreted ‘supply’ in light of how software is distributed. The court found that a server holding copies that can be downloaded is supplying a copy. The court stated that it could not “disregard the nature of the relevant technology and business practices underlying a particular litigation”. The court specifically focused on the technology at issue in order to interpret how the patent law applied – the dissenting judge argued that this was against the normal rules of legal interpretation of patent statutes.

The court rejected the notion that differences might exist in the form of transmission of the software (ie, on disk or by electronic communication). Microsoft further argued that the decision could have a potentially damaging effect on the domestic software industry, as companies might choose to move more manufacturing facilities overseas, but the court rejected this on the grounds that its decision cannot be based on the potential loss of US jobs.

NTP v RIM

Another recent highly visible patent case (even involving letters from members of Congress protesting against a possible injunction against BlackBerrys), is NTP Inc v Research In Motion Ltd. Research In Motion (RIM) sells BlackBerrys that utilise a wireless network for their operation. Part of the system for sending emails from one BlackBerry user to another includes sending the email through a BlackBerry relay located in Canada.

The US patents at issue covered the system and method for sending messages. Even when two persons send messages to each other in the United States, the message is transmitted using the relay located in Canada. As the US patent statute at issue requires that the infringing activities have occurred in the United States, RIM argued that the use of the BlackBerry relay in Canada meant that the entire infringing activity did not occur in the United States, so there could be no infringement.

The Federal Circuit found that having one of the components – termed the “control point” by RIM – outside the United States did not preclude infringement. The court rejected a strict standard that all activities must occur in the United States, stating that: “Even though one of the accused components in RIM’s BlackBerry system may not be physically located in the United States, it is beyond dispute that the location of the beneficial use and function of the whole operable system assembly is the United States.” The court reasoned that the true location of the use of the system was in the United States, even though a critical component for using the system was located in Canada.

Practical impact on software and telecommunications

How does the Eolas situation arise for a company exporting software?

The Eolas/AT&T line of cases brings new risks for companies by interpreted expansion of the scope of US patent law. Patentees may argue that any form of exporting software from the United States can render a company liable for all copies of that software, no matter where it is copied and sold. For example, as part of a contract made wholly outside the United States, a company may manufacture a product in Japan, the production of which uses in part a software update coming from a server located in the United States. Patentees may use the Eolas/AT&T reasoning to argue that the software was exported with an intent to copy, thus creating infringement. Factual issues regarding intent become the focus of the infringement allegation.

Patentees will further explore the whole design and distribution process for the software in an attempt to find some link to the United States (eg, in writing the software code) for components otherwise sold completely outside the United States. Patentees may assert that there was some exportation of software, which was ultimately incorporated into a product sold outside the United States.

International companies must consider that patentees could seek to argue infringement of the end product, arguing that the OEM actually carried out the export by downloading the software from a server in the United States and that damages should be based on the complete end product. Coupled with the high cost of US patent litigation and the presence of forums perceived to be patent-friendly, the worldwide sales of a product become an issue, as opposed to just the US sales.

Furthermore, companies working with other companies in the development of software must consider the territorial issues, such as code-writing activities being carried out in the United States. Indemnification and other clauses in co-development or supply agreements need to be re-examined from this unique angle.

What does the NTP decision mean for telecommunications?

Networks are built from communication nodes and components based in several locations, both within and outside the United States. The Federal Circuit has rejected the bright-line rule that all the components relevant to the patent claims must be located in the United States. Instead, the court seems to look primarily at the facts and circumstances to determine the “beneficial use and function of the whole operable system assembly”. While the court’s standard does not necessarily mean that a system with any perceived benefit in the United States makes it susceptible to US patent law, moving a key component offshore no longer provides what many previously considered to be a safe haven for avoiding infringement.

On the one hand, companies can be comforted by the court’s practical assessment of where the base operation and benefits occur. On the other hand, the site of a use becomes more complicated in light of the NTP decision. Furthermore, savvy patentees that previously felt shut out because their patents covered multi-component systems that were partially outside the United States will seek to take the NTP decision and push the boundaries to see exactly what percentage of the components must be located in the United States and where the benefits exist. Companies should assess the main components of their networks and determine the appropriateness of moving further components offshore. Such a move may prove beneficial, especially where the companies can strongly mitigate risk without tremendous capital expenditures (eg, where supporting facilities already exist offshore). The additional overheads could readily offset the potential cost of a patent infringement lawsuit.

Communications intrinsically involve interaction among several parties owning various network components. Companies must analyse geographic impact and base operational issues, and draft their contracts with third parties accordingly. Companies working outside the United States could find their products subject to an infringement action brought in the United States, and they need to consider carefully exactly what they are indemnifying with special attention to geographic limitations (or requirements, depending on the viewpoint).

Expanding beyond networking and software

No one can accurately predict how the Federal Circuit may address the geographical boundaries issue beyond the particular technologies, and associated facts, involved in Eolas, AT&T and NTP. Yet companies involved in various other technologies should consider their activities and how they may affect the United States. Such an evaluation goes beyond software and communications – for example, where mass production occurs from a single base component or where various aspects of the activities occur in the United States, even though some key aspect occurs outside the United States. Grey areas exist between sending a golden master disk from the United States for copying outside the country and the mass manufacturing of widgets outside the United States, and patentees will be conceiving new arguments to cover these grey areas.

With the expected continued growth in US business methods patents, the potential impact could be even greater. The US Patent and Trademarks Office has recently declared that US patents (eg, business method patents) do not have a technological arts requirement – in other words, they do not have to be tied to a computer or other electronic device (Ex parte Lundgren, US PTO Board of Patent Appeals, October 2006). With a greater breadth of US patents in all forms of business methods, companies could become subject to patents in areas never previously contemplated, even though their actions do not occur exclusively in the United States.

Conclusion

Traditional transborder enforcement stems from enforcing a portfolio of patents in different countries, each territory being relatively separate and distinct from the others. As borders become more transparent through technology, the associated risks grow. The assertion of US patent rights against non-US activities is likely to continue to grow through the recent developments discussed above. Companies must re-evaluate their activities and contracts with third parties in order to stay ahead. Patentees may launch enforcement programmes based solely on US patents having a transborder effect.

Author Information