Center for Advanced Study & Research on Innovation Policy


CASRIP Newsletter - Spring 2007, Volume 14, Issue 2

Supreme Court Disagrees with Federal Circuit in Two Controversial Patent Cases: KSR Int’l Co. v. Teleflex and Microsoft Corp. v. AT&T Corp.

By Toshiko Takenaka1


On April 30, the Supreme Court handed down two controversial decisions, KSR Int’l Co. v. Teleflex2 and Microsoft Corp. v. AT&T Corp., 3 in both cases, reversing the Federal Circuit judgments.  Between the two cases, the patent community has been most anxious for the outcome of KSR Int’l v. Teleflex because the Supreme Court might have completely changed the way the United States Patent and Trademark Office (“USPTO”) and district courts examine nonobviousness in patent cases, which is the most frequently cited grounds for rejecting claims during the examination period as well as common grounds for challenging patent validity during litigation.  Fortunately, the changes brought by KSR will be less than some groups in the patent community anticipated because the Supreme Court preserved the Federal Circuit’s teaching-suggestion-motivation (“TSM”) test, with an instruction to change the way to apply the test.


            A.        Case History

In KSR, the petitioner-accused infringer, KSR, challenged the validity of U.S. Patent No. 6,237,565 (“the ‘565 patent”), which is exclusively licensed to a competitor, Teleflex. The ‘565 patent relates to a mechanism for combining an electronic sensor with an adjustable automobile pedal so that the pedal’s position can be transmitted to a computer that controls the throttle in the vehicle’s engine.  The district court viewed the ‘565 patent as a mere combination of old elements, which are disclosed in separate prior art references and held the invention obvious, finding only “little difference” in the adjustable pedal mechanism in the asserted claim from that disclosed in a reference referred by Asano, which was not reviewed by the examiner during the prosecution.  The small gap was then filled with other references that suggested a combination of electronic sensors with adjustable pedals and a position of such sensors on the fixed structure of such pedals. After rejecting Teleflex’s argument for nonobviousness relying on commercial success, the district court granted summary judgment for KSR. 

The Court of Appeals for the Federal Circuit reversed the summary judgment, finding that the district court’s nonobviousness analysis was improper.4  According to the Federal Circuit, the district court erred in applying the TSM test because it failed to find a specific understanding or principle within the knowledge of one skilled in the art that would have motivated one to locate a sensor in the adjustable pedal in the Asano reference in the manner as disclosed in the ‘565 patent.  The district court also found to have erred in granting summary judgment because there were genuine issues of material fact.

            B.        Expansive and Flexible Interpretation

In a unanimous opinion authored by Justice Kennedy, the Supreme Court reversed the Federal Circuit’s judgment sustaining the validity of the ‘565 patent.  In interpreting its own precedent, the Supreme Court described its nonbviousness analysis framework from Graham5 as expansive and flexible.  Although the Court acknowledged the benefit of the TSM test as providing helpful insight, it criticized the Federal Circuit’s application of the test for being too rigid and thus inconsistent with its own approach.  In particular, the Court emphasized its instructions in early decisions with respect to the need for caution in granting a patent on a combination invention.  Such inventions are qualified for patents only with a showing of an unexpected result, if the invention consists of elements disclosed in the prior art and these known elements are combined in a manner known to one skilled in the art. 

The Court did not completely reject the use of the TSM test, however. The Court acknowledged the necessity to identify a reason that would prompt one skilled in the art to combine known elements in the way the patent discloses for making an assessment of nonobviousness.  Nonetheless, to be consistent with its own precedent, the Court urged lower courts to apply the test more flexibly acknowledging that it is not always necessary to find a specific teaching in published articles.  Due attention must be paid to the effects of demands in the marketplace and the background knowledge of one skilled in the art.  If applied properly, there is no inconsistency between the idea underlying the TSM test and the Graham analysis.

            C.        Problems in Federal Circuit TSM Application

The Supreme Court also listed a series of problems with respect to the Federal Circuit’s application of TSM test.  First, the Federal Circuit’s obviousness inquiry is too narrow because the TSM test focuses on the specific problem the patentee was trying to solve.  The Court urged analysis of the objective reach of the claim at issue by examining known problems at the time of the invention.  Keeping such problems in mind, one must evaluate whether a solution adopted by the claim was obvious.  The Supreme Court also criticized the Federal Circuit for looking into the elements in the prior art that are designed to solve the same problem as that of the claimed invention.  Instead, in combining elements from different references, the Court explained that one skilled in the art should know obvious uses of such elements beyond their primary uses and thus should be able to mosaic elements from multiple references to conceive the claimed invention.  

Furthermore, the Court criticized the Federal Circuit’s application of the “obvious-to-try” test.  According to the Court, a claimed invention is obvious if there is a design need or market pressure to solve a problem and there are a finite number of solutions to the problem.  Finally, the Court found problems with respect to the Federal Circuit’s citing a risk of falling prey to hindsight bias to support its nonobvioiusness finding and noted that such risk does not justify the Federal Circuit’s rigid rule for preventing examiners and district courts from taking recourse from common sense for a finding of obviousness.

            D.        Application of New Test

In applying the redefined flexible nonobviousness test to the facts in KSR, the Supreme Court agreed with the district court and found the claim obvious.  Teleflex was prevented from raising the argument that a sensor cannot be combined with the adjustable pedal in the Asano reference because the Court found that Teleflex did not raise that argument in the district court in a clear fashion.  After acknowledging the need for converting mechanical pedals to electronic pedals and the presence of multiple methods for such conversion in the market at the time of invention, the Court held the patent claim obvious because one skilled in the art would have conceived the improvement of the Asano pedal by combining a sensor.  Interrelated teachings of two more references, locating a sensor on a non-moving point and avoiding a wire-chafing problem, suggested locating the sensor on a fixed pivot point.  Teleflex’s teaching away argument was rejected because there was no evidence that the adjustable pedal was so flawed that no one skilled in the art was motivated to improve the pedal.  Finally, the Court agreed with the district court that Teleflex was unable to produce evidence on secondary considerations to refute the nonobviousness determination that resulted from the assessment of the technical factors.  In short, placing a sensor on a fixed pivot point of the adjustable pedal in the Asano reference was within the scope of the ordinary skill of a person skilled in the art of invention at the time of the invention.

The petitioner challenged the Federal Circuit’s ruling in finding error for the district court’s grant of summary judgment. The Supreme Court again disagreed with the Federal Circuit, denying the existence of a dispute over an issue of material facts. Unless an expert provides a conclusory affidavit of nonobviousness and thus the court misunderstood the role of experts, district courts can properly grant summary judgment if parties do not dispute factual findings.  This is because the ultimate judgment of obviousness is a question of law. Accordingly, the Federal Circuit’s judgment was reversed and the case was remanded to the Federal Circuit to affirm the district court’s summary judgment.

            E.         What the Future Holds

It is very difficult to predict the scale of changes brought by KSR.  The Court cited Federal Circuit decisions issued after the grant of certiorari and expressly endorsed the current effort for making the application of the TSM test more flexible, the Federal Circuit may find little need to change their practice except for providing more emphasis on a finding of implicit teaching, suggestion and motivation.  Instead, the USPTO may interpret the Court’s caution against patenting combination inventions to endorse a practice of finding prima facie obviousness without any showing of TSM if an invention consists from known elements.  Such practice will shift the focus of disputes between the USPTO and applicants from the absence of TSM to the presence of teaching away and secondary considerations.

KSR attracted attention worldwide, particularly in the Japanese patent community.  Japan’s new IP High Court has been criticized for its high invalidity rate for lack of an inventive step.  Under the current practice in the Japanese Patent Office (“JPO”) and Japanese courts, a combination of known elements is obvious as long as these elements are disclosed in references analogous to the art of the claimed invention.  Some argue that it is too difficult to persuade the JPO and Japanese courts to find an inventive step by showing evidence of teaching away.  Further, neither the JPO nor Japanese courts take account of secondary considerations such as commercial success in finding an inventive step.  Since most of the inventions in matured technology fields are combination inventions, some have challenged the current practice as being inappropriate by placing an extremely high hurdle for obtaining a patent. If KSR is interpreted to eliminate any necessity to show a reason for combining elements in multiple references, the U.S. nonobviousness standard may face the same challenge from the patent community unless the USPTO and U.S. courts reinforce the argument of teaching away and secondary considerations.  However, such an interpretation seems inconsistent with the Court’s instruction to require an explanation of combining known elements from multiple resources, such as the knowledge of one skilled in the art, and the demand of the market.


            A.        Extraterritoriality

The Microsoft dispute attracted the attention of the business community worldwide because of the scale of money involved in the litigation.  Microsoft involved a patent infringement dispute over Microsoft’s main product, Windows software.  Internet developments and technological advancements have made it nearly impossible to keep products within the border of one country.  This is particularly true with respect to software products, because software products are manufactured and used extraterritorially.6  Patent doctrines are developed around territoriality principles, however.  Therefore, patentees must obtain a patent in each country where they need protection and must sue for infringement in each country respectively. 

Despite the territoriality principle, AT&T tried to collect damages for Microsoft’s activities both within and outside of the United States, relying on the exception provided in 35 USC §271(f).  Section 271(f) provides for infringement of U.S. patents with respect to an act of supplying components of a patented invention from the United States for combination abroad.  However, this attempt failed because the Supreme Court restrictively interpreted the §271(f) exception and refused to find infringement of a U.S. patent with respect to computers made abroad when they are loaded with software copied from a master disk or electronic transmission sent from the United States.

            B.        Case History

In Microsoft, AT&T sued Microsoft for infringement of its patent directed to an apparatus for digitally encoding and compressing recorded speech.  Although Microsoft’s Windows software does not infringe the AT&T patent unless it is loaded with a computer, AT&T asserted infringement of U.S. patents with respect to not only domestically loaded computers, but also those in which software was copied and loaded outside the United States under 35 USC §271(f).  The district court adopted an expansive interpretation in including unincorporated software within the definition of “component” and dispatching unincorporated software to be copied and installed in a computer abroad within the definition of “supply from the United States.”  In a two-to-one decision, the Federal Circuit affirmed the expansive interpretation.

            C.        Supreme Court Limits §271(f) Expansion

In a 7-1 ruling, the Supreme Court rejected the expansive interpretation, emphasizing the nature of 35 USC §271(f) as an exception to the territoriality principle.  Instead, the Court adopted a restrictive interpretation in limiting the application of §271(f) only to components that are combined to form the patented invention.  Software in AT&T does not constitute such a component unless it is incorporated in a computer readable copy.  Software in the abstract cannot be combined to form the patented speech processing computer because it cannot be installed or executed in a computer.  The Court analogized unincorporated software to an idea or information, but distinguished physical embodiments covered in the §271(f) exception.  Since §271(f) is silent with respect to information, the Court concluded that the exception cannot be interpreted to cover unincorporated software.

The Court neither found §271(f) infringement with respect to Microsoft’s making copies nor installing them in computers in foreign countries. Under a conventional reading, §271(f) requires the very components, instead of their copies, to be supplied from the United States.  Without statutory language addressing copying, courts should refrain from expanding the §271(f) coverage through judicial interpretation. Accordingly, Microsoft’s dispatching unincorporated software does not fall within the scope of the §271(f) exception because copies of Windows software were made and installed outside the United States.

            D.        Impact of Microsoft

In concluding its analysis, the Supreme Court reemphasized the territoriality principle, stating: “[A]ny doubt that Microsoft’s conduct falls outside §271(f)’s compass would be resolved by the presumption against extraterritoriality.”  Rejecting AT&T’s argument that a restrictive application of §271(f) may create a loophole for software makers, the Court reasoned that closing the loophole is a task of Congress through legislation rather than judicial interpretation.
Economically, the impact of Microsoft is very significant.  United States courts have found large damages against Microsoft, including foreign software sales, under the §271(f) exception.  The Court’s restrictive interpretation will protect Microsoft from U.S. liability for foreign infringement.  It will require patentees to sue Microsoft and other software makers in the country where their foreign patents are violated.  This adherence to the territoriality principle is in line with the Federal Circuit’s interpretation of 28 USC §1367 with respect to the supplemental jurisdiction over foreign infringement in Voda v. Cordis.  However, U.S. manufacturers are subjected to U.S. patent infringement liability as long as physical components of a patented invention are exported.  As the Supreme Court suggests, until Congress addresses the §271(f) exception, the loophole is open only for software manufacturers.7 

1. Dr. Toshiko Takenaka is the Director of CASRIP and the Washington Research Foundation/ W. Hunter Simpson Professor of Law at the University of Washington School of Law.

2. 550 U.S.__(2007).

3. 550 U.S.__(2007).

4. For an analysis of the Federal Circuit decision, see Benjamin Keim, Gobbledygook or the Future of Obviousness in U.S. Patent Law, CASRIP Newsletter, Vol. 14, Issue 1 (Autumn/Winter 2007) available at:

5. Graham v. John Deere Co., 383 U.S. 1 (1966).

6. NTP v. Research In Motion, 418 F.3d 1282, 75 USPQ2d 1763 (Fed. Cir. 2005)  An element of a patented apparatus located outside the United State.  The Federal Circuit nevertheless found infringement of US patent.  Barry Bretschneider, NTP v. RIM – Pushing the Envelope of Territoriality, CASRIP Newsletter Vol. 13, Issue 2 (Spring/Summer 2006) available at

7. Unlike the bill introduced in Senate last year, the bill introduced to Senate and House April 18 does not include a proposal to repeal §271(f).  Thus, it is unlike to see the fate of the loophole under this administration.

Top of Page

Connect with us:

© Copyright 2015, All Rights Reserved University of Washington School of Law

4293 Memorial Way Northeast, Seattle, WA 98195