Microsoft v. AT&T: Extraterritorial Enforcement of US Patents

Microsoft Corp. v. AT&T Corp., 550 U. S. ____ (2007).

In a 7-1 decision, the Supreme Court held that Section 271(f) of the Patent Act does not extend to cover foreign duplication of software.

There is no such thing as a world-wide patent. Rather, patent law is territorial. A US patent covers infringing acts that occur in the US but generally disregards extraterritorial activity. The lone statutory exception is Section 271(f) of the Patent Act, which calls for infringement liability for the unauthorized supply of "components" of a patented invention for "combination" abroad.

Todays business reality has considerably dulled the teeth of 271(f) in the area of manufacturing — few physical components are shipped abroad from the US.  Through Microsoft and others, the US continues to be a leader in software development and the export of software.  Of course, Microsoft does not export its software by loading millions of CDs on a cargo ship. Rather, it sends only a few copies to foreign OEM distributors who load copies of the software onto PCs for sale.

AT&T is the assignee of a patent covering a computer for encoding and compressing recorded speech. In the US, a computer with Microsoft WIndows installed infringes the patent. The question in this case, is whether the foreign installed software can be considered a "component" supplied from the US under 271(f). 

The question before us: Does Microsoft’s liability extend to computers made in another country when loaded with Windows software copied abroad from a master disk or electronic transmission dispatched by Microsoft from the United States? Our answer is "No."

As can be seen from the various amici briefs, reasonable minds can differ regarding two statutory elements: (1) whether software — without being tied to a physical medium — can be classified as a "component;" and (2) whether the foreign-made copy was "supplie[d]" from the US. Abiding by its principle against extraterritorial application of laws — especially patent laws — the Supreme Court answered both of these questions in a restrictive fashion:

  • Abstract Software: Until it is expressed as a computer-readable "copy," e.g., on a CD-ROM, Windows software indeed any software
    detached from an activating medium remains uncombinable. . . . Abstract software code is an idea without physical embodiment, and as such, it does not match §271(f)’s categorization: "components" amenable to "combination."
    [DDC: Software is never without a physical embodiment?]
  • Supply of Copies: [T]he very components supplied from the United States, and not copies thereof, trigger §271(f) liability when combined abroad to form the patented invention at issue. Here, as we have repeatedly noted,  the copies of Windows actually
    installed on the foreign computers were not themselves supplied from the United States.
    [Therefore no liability for foreign-made copies]

Dissent by Justice Stevens: In my view, Justice Stevens – in a solo dissent – has the more sensible view of component.

[I]f a disk with software inscribed on it is a "component," I find it difficult to understand why the most important ingredient of that component is not also a component.


Important recent 271(f) cases:

  • NTP v. Research in Motion, (271(f) “component” would rarely if ever apply to method claims).
  • AT&T v. Microsoft, 414 F.3d 1366 (Fed. Cir. 2005) (271(f) “component” applies to method claims and software being sold abroad);
  • Union Carbide v. Shell Oil (Fed. Cir. 2005) (271(f) “component” applies to method claims).
  • Eolas v. Microsoft, 399 F.3d 1325 (Fed. Cir. 2005) (271(f) “component” applies to method claims and software);
  • Pellegrini v. Analog Devices, 375 F.3d 1113 (Fed. Cir. 2004) (271(f) “component” does not cover export of plans/instructions of patented item to be manufactured abroad);
  • Bayer v. Housey Pharms, 340 F.3d 1367 (Fed. Cir. 2003) (271(g) “component” does not apply to importation of ‘intangible information’).

22 thoughts on “Microsoft v. AT&T: Extraterritorial Enforcement of US Patents

  1. 22


    Of course company M would NOT be liable in your scenario. U.S. patent law has never been designed to prevent export of patented technology per se. Instead, patent law is designed to encourage public disclosure, on the theory that wer’e all better off.

    Your concerns ignore two factors: First, a US company can get patent protection in those foreign markets. (Of course, patent protection might not be very effective in some of those markets, but that’s a separate issue.) Second, US companies are not the only innovators in the world. If patent law worked as you suggest, then US companies would be barred from using innovations developed and patented elsewhere.

    The Supreme Court got this one right, although perhaps not for the best reasons. It’s a shame they didn’t grant cert on NTP v. RIM and clean up the extraterritorial mess that the Federal Circuit made of that one.

  2. 21

    just for the sake of clarity

    in your opinions , if a company M would build products, if a company A would hold patents in the US on some parts of the products build by company M.

    would the company M be liable if it would ship the blueprint explaining how to manufacture the A company to component to a partner X to manufacture and sell abroad ?

    in my understanding of the ATT vs Microsoft Ruling, Company M would not be liable of any infringement since the blueprint is not, in itself, a component of the A patented component only a description of how to create it….

    in which case, the ruling simply killed section-271! Now a US company can, for its export market, set an agreement with a foreign company located where another US company has no patents and have this foreign company producing the patented components for its export markets…
    basically patent no longer provide any protection against illegal export of patented technology

    if this is true, then it seems to me extremely dangerous for two reasons :
    *the trivial one being that as world evolve US domestic market will become less and less important as larger countries (China, India and so many more) start entering the competition. this will mean that there will be no way of preventing US company from being defrauded of their export market by rival US company cooperating with foreign third party company…. this will unavoidably lead to push US company toward trade secret instead of patents…
    *the second threat is more serious, US defense contractor in their legitimate quest for profit might be tempted to foster such “foreign third company” deal to extract revenue from export of products that have been developed by their US competition…

    I guess anyone will see how dangerous such a behavior could be for the future of the nation… at least in my opinion much more threatening than growing line of unemployed programmers but that’s just my two pences…

  3. 20

    Funny how Rader gets kudos from the majority for his dissent in AT&T, but Rader wrote the opinion in Eolas v. Microsoft that found that 271(f) DOES apply to “golden master” sales overseas.
    See link to

  4. 19

    This pretty well kills protection for Internet-based services. Now you can put a server in any territory in which the patentee lacks patent protection, upload software components from the U.S., and fire away.

  5. 18

    Here’s what I don’t understand about Justice Steven’s dissent. He states that section 271(f)(2) provides the best support for AT&T’s position. 271(f)(2) requires that the component is not “a staple article or commodity of commerce suitable for substantial noninfringing use.” Justice Stevens conclusorily states that Windows is “surely not” such a commodity.

    Is he really saying that there are no other possible uses of Windows other than to infringe AT&T’s patent? That makes no sense to me. Windows can do a ton of things that doesn’t involve the encoding and decoding of speech. For example: Word, Excel, PowerPoint, Solitaire, Calculator, and Outlook. In fact, there are many things that Windows can do that does not involve the processing of speech.

    How can he so easily say that there are no noninfringing uses of Windows

  6. 17

    Since the jury in the damages phase could not take into account any damages for the foreign non-infringement, the royalty value for the infringement on the master disk is zero. Microsoft, knowing it is not liable for the foreign infringement, thumbs its nose any anyone trying to get royalties on master disks. Remember, it is paying for all US copies.

  7. 16

    Let’s be clear here… for something to be a component IT DOES NOT NEED TO BE PATENTABLE. For that matter, it doesn’t need to be protectible in and of itself at all.

    That’s the whole point of the section in the patent law. It recognized that you could make and export unpatented components here, assemble them abroad into something that may be patented, and presto you’ve avoided infringement.

    People have mentioned things like recipes on paper, etc., as if the nature of the component somehow mattered. However, if a patent covered a machine used to organize a specific type of paper recipe, and a party was sending the special paper (not itself patentable) abroad to assemble the machine, we have this situation. Granted, even in that example, simply sending a list of recipes to be copied onto the special paper abroad (something closer to the AT&T case) probably wouldn’t infringe. That point is really, though, that this case is not arguing whether software-as-software in this case was patentable (indeed, the opposite came up several times in oral arguments by both sides).

    The issue was really whether software not in the form it was ultimately used (e.g. the stupid blueprint analogy) could become a component.

  8. 15

    It seems to me that this may, for practical purposes, be a tempest in a teapot. Assuming that the jury got it correct that Microsoft infringed when it shipped the master disk overseas, then Microsoft is liable for infringement — whether it is just that one act of infringement when it shipped that one disk, or millions of acts infringement when it used that master disk to create the millions of copies of Windows that were sold overseas.

    Once liable, the question becomes what are the damages from the act or acts of infringement. Assuming a reasonable royalty, the question is what royalty would Microsoft agree to pay in a hypothetical negotiation. From an economic standpoint, whether it’s just one act of infringement, or millions of acts, Microsoft’s profits are the same — even if the only act of infringement were the shipping of the master disk, Microsoft would have lost all of its foreign sales if it had not shipped the master disk overseas to make the copies. So, Microsoft should be willing to pay the same royalty — so that it can enjoy the profit stream — regardless whether it was one act or millions of acts of infringement. And, it’s reasonable to think that a reasonable royalty for the right to ship an infringing master disk overseas so that you could generate the profits associated with selling millions of copies of Windows by making copies from that master disk would be some small percentage of each of those millions of sales. So, in the end, you get to the same place.

  9. 13

    This case reflects the shifting internationalization of trade and requisite IPR and changing relative influence of the US PTO. So many companies today develop both hardware and software products in multiple design centers using many patented ‘technologies’. The bulk of software I deal with gets embedded into products that are then put in the field or sold to end customers.. not this specific case. Several articles and some analysis, however, has already appeared that equates the Microsoft vs. ATT decision to a very broad interpretation of a ruling on patent infringement venue. This is a very important case but it is, as has been recognized on this blog at least, limited.

    The other recent ruling by the U.S. Supreme Court, KSR International v. Teleflex, is probably much more important to the field that I am most interested, evolving WiMAX, LTE 4G wireless systems and to patents in general. KSR rules to the core edict that a patent should not be granted if it is obvious to those versed in the art or an obvious combination of prior art.

    It can be argued that many patents used in complex products such as communications systems are built ‘on the shoulders of giants’ that have come before and are derivative rather than truly inventive. Many, imo, are simply describing an alternative method that is codified into the particular adopted standard – ‘essential(ly)’ boils down to patent counting. And many could be argued to have been derived through a process of group participation and revelation in addition to building on substantial prior art. My own opinion about U.S. and foreign patents in fields I have studied is that very few rise to being ‘fundamental’ eureka type inventions that become show stoppers similar to Qualcomm’s tested CDMA IPR.

    The U.S. PTO has come under criticism for being under staffed, under budgeted, under equipped for external literature references, and overly easy about granting patents. The KSR decision comes down on the side of cutting down some of the growing ‘patent thicket’.

  10. 12

    “What is so special about a master disk that it only carries intangible information, while a regular software CD contains patentable software. Am I missing something?”

    I believe so. What is special about the master disk is that it was shipped from the US and the foreign copies were not. However, if abstract software were a 271-component, the CD, master or otherwise, would be irrelevant. Export of the abstract software with intent that it be loaded in millions of foreign computers could be infringement.

    I read the main opinion as implying that there would have been infringement if the master CD had itself been loaded into millions of computers for purposes of copying the software. There was no infringement because a copy was used instead and the copy did not originate in the US.

    If software in the abstract was a component for purposes of 271, then it would be hard to argue that that exported component was not part of each finished loaded computer. So the majority decision turns on this point.

    According to the concurring opinion, there would be no infringement even if the master copy was directly loaded into millions of computers to copy the software. That is because the CD is removed after the software is copied and before the computer is sold. The CD is merely a tool used to create the finished computer with the loaded software. However, if abstract software is a 271 component, then even there I would think there could be infringement by placing this abstract software in each foreign computer. The CD — master or otherwise — becomes irrelevant.

  11. 11

    “What is so special about a master disk that it only carries intangible information, while a regular software CD contains patentable software. Am I missing something?”

    Nope. You are just confused by the inherent contradictions in the law which led to the patentability of information and/or instructions designed to be read by computers. Reasonable people should be confused.

  12. 10

    This has been observed many times by now, but I’m still fascinated by Microsoft’s efforts in this case to undermine the value of software patents. Of course, if there is one software company that does NOT need patents to succeed, it is Microsoft. They would surely prefer there to be no software patents at all. So in this case they were able to align their interests in both winning the issue at hand and undermining software patents. Nicely played.

  13. 9

    I agree with you that abstract information isn’t and shouldn’t be patentable, but what is software if not information? I know it becomes part of the machine in a sense, but isn’t like saying a recipe become part of the baker’s brain when he follows the instructions and bakes a cake? Don’t get me wrong, I’m in favor of some kind of protection for software (and I don’t believe that copyright is sufficient). And I’m not equating the human brain with a computer. It just seems to me that the nature of software is causing a lot of confusion in the patent realm, and leading to bad opinions. What is so special about a master disk that it only carries intangible information, while a regular software CD contains patentable software. Am I missing something?

  14. 8

    “Software is a blueprint no matter how you look at it. The confusing part is that we’ve developed machines that can actually “read” blueprints.”

    Computers have been able to “read” printed information for quite a while now. But information is not patentable and — guess what? — it never will be.

    Not tomorrow. Not in fifty years.

    “Microsoft didn’t export “abstract information” it exported a tangible copy, i.e. the golden master disk.”

    Of course, finding Microsoft guilty of that one easily circumnavigated act of infringement would be a hollow victory for AT&T. So AT&T pushed harder and went over the cliff. Buh-bye!

  15. 7

    The At&t decision seems flawed to me. The disctinction the court makes between abstract software information and a tangible copy of software seems irrelevant. Microsoft didn’t export “abstract information” it exported a tangible copy, i.e. the golden master disk. The court’s distinction seems to be that if Microsoft calls the disk a “master disk” it’s information, but if Microsoft calls the disk a copy of the Windows operating system then it’s a tangible copy. Or maybe the court is saying that the information that was actually transfered from the master to the computer was abstract and therefore not a component, but if that’s true then Microsoft would not have been guilty of violating 271(f) even if they exported each copy of Windows individually, because the intangible information on the CD would still have to be copied from the CD to the computer. In other words, the CD would not become a part of the computer, rather the intangible information on the CD would be transferred to the tangible form on the computer. I find it hard to believe that the issue of whether software is a component of a patented invention depends on whether one makes several copies of the software from one disk, or whether several disks are used to make one copy each.
    Software is a blueprint no matter how you look at it. The confusing part is that we’ve developed machines that can actually “read” blueprints.
    For an interesting case note on At&t v. Microsoft see 59 ME. L. REV. 191 (2007). (I should know I wrote it.)

  16. 6

    “[I]f a disk with software inscribed on it is a “component,” I find it difficult to understand why the most important ingredient of that component is not also a component.”

    Nice try, Justice Stevens.

    If a piece of paper with a recipe printed on it is a “component”, I find it difficult to understand why the most important aspect of that component is not also a component.

    The answer, Justice Stevens, is that US patent law does not protect information.

  17. 5

    “Can we infer from this decision that the Supreme court has recognized software to be patentable?”


    This has been another edition of simple answers to simple questions.

  18. 4

    “This case certainly implies that software in the abstract cannot be patentable.”

    Who cares if disembodied software is not patentable? Just draft claims that recite the same software embodied on a physical computer-readable medium.

  19. 3

    This case certainly implies that software in the abstract cannot be patentable. “Abstract software code is an idea without physical embodiment.”

  20. 2

    In the oral arguments, one of the Justices said that the Supreme Court has never held software to be patentable (i.e. only lower courts).

    Can we infer from this decision that the Supreme court has recognized software to be patentable?

  21. 1

    Justice Stevens may have the more intellectually satisfying approach to the question but the practical consequence of liability for Microsoft would have been to drive software development offshore.

    You can’t eat intellectual satisfaction and being right is cold comfort while standing in the unemployment line.

Comments are closed.