1308 767 federal reporter, 3d series - berkeley law · 1308 767 federal reporter, 3d series...

1308 767 FEDERAL REPORTER, 3d SERIES

incomplete that they could not be usedwithout undue difficulty.

[10] Contrary to Mukand’s argument,the deficiencies in its responses were notlimited to a discrete category of informa-tion. As Commerce noted, Mukand as-signed the ‘‘same amount of conversioncosts per kilogram of bar produced, irre-spective of the final size of the productproduced.’’ J.A. 1604. Mukand thuspremised all of its production cost data onthe assumption that product size is not asignificant cost factor—an assumption itfailed to support. In general, use of par-tial facts available is not appropriate whenthe missing information is core to the anti-dumping analysis and leaves little room forthe substitution of partial facts withoutundue difficulty.12 Without cost data bro-ken down by product size, Commerce wasunable to differentiate between differenttypes of steel bar products and could notcalculate an accurate constructed value forany of Mukand’s products. We thereforehold that Commerce’s reliance on totalAFA is supported by substantial evidence.

III

For the reasons set forth above, weaffirm the decision of the Trade Court.

AFFIRMED

,

VIRNETX, INC., Plaintiff–Appellee,

and

Science Applications InternationalCorporation, Plaintiff–

Appellee,

v.

CISCO SYSTEMS, INC., Defendant,

and

Apple Inc., Defendant–Appellant.

No. 2013–1489.

United States Court of Appeals,Federal Circuit.

Sept. 16, 2014.

Background: Patentee brought actionagainst mobile phone manufacturer, alleg-ing infringement of patents describingmethod of transparently creating virtualprivate network (VPN) between clientcomputer and target computer and patentsdisclosing secure domain name service.The United States District Court for theEastern District of Texas, Leonard Davis,Chief Judge, 2009 WL 2370727 and 2012WL 3135639, construed the claims, andthen denied manufacturer’s post-trial mo-tions after jury returned verdict in paten-tee’s favor 925 F.Supp.2d 816. Manufactur-er appealed.

Holdings: The Court of Appeals, Prost,Chief Judge, held that:

(1) term ‘‘domain name’’ meant name cor-responding to Internet Protocol (IP)address;

(2) term ‘‘secure communication link’’meant direct communication link thatprovided data security and anonymity;

12. See Shanghai Taoen Int’l Co. v. UnitedStates, 360 F.Supp.2d 1339, 1348 n. 13 (Ct.

Int’l Trade 2005).

1309VIRNETX, INC. v. CISCO SYSTEMS, INC.Cite as 767 F.3d 1308 (Fed. Cir. 2014)

(3) substantial evidence supported juryverdict that accused product met ‘‘di-rect communication’’ limitation;

(4) accused feature in mobile phone manu-facturer’s product met ‘‘determiningwhether’’ limitation;

(5) substantial evidence supported jury’sfinding that feature in accused productcreated ‘‘VPN’’ or ‘‘secure channel’’that extended from client to targetcomputer;

(6) security provided by accused systemthat included encryption on insecurepaths but otherwise relied on securityprovided by private networks was notequivalent to ‘‘encrypted channel’’;

(7) patentee could not rely on entire mar-ket value of multi-component productcontaining several non-infringing fea-tures with no relation to patented fea-ture to approximate reasonable royaltybase; and

(8) evidence relying on 50-50 startingpoint based on bargaining solutiontheorem was not admissible.

Affirmed in part, reversed in part, vacatedin part, and remanded.

1. Patents O324.5

Patent claim construction is a questionof law that is reviewed de novo.

2. Patents O165(3), 167(1)

The process of construing a patentclaim term begins with the words of theclaims themselves; however, the claimsmust be read in view of the specification,of which they are a part.

3. Patents O165(5)

The patent claim differentiation doc-trine disfavors reading a limitation from adependent claim into an independentclaim.

4. Patents O159Although courts are permitted to con-

sider extrinsic evidence like expert testi-mony, dictionaries, and treatises when con-struing a patent claim term, such evidenceis generally of less significance than theintrinsic record.

5. Patents O101(2)Term ‘‘domain name,’’ in patents that

disclosed domain name service (DNS) sys-tem that resolved domain names and facili-tates establishing secure communicationlinks, meant name corresponding to Inter-net Protocol (IP) address.

6. Patents O101(2)Term ‘‘secure communication link,’’ in

patents that disclosed domain name ser-vice (DNS) system that resolved domainnames and facilitates establishing securecommunication links, meant direct commu-nication link that provided data securityand anonymity.

7. Courts O96(7)In a patent case, the Court of Appeals

for the Federal Circuit reviews the denialof a motion for judgment as a matter oflaw (JMOL) or a new trial under the lawof the regional circuit.

8. Federal Courts O3602The Fifth Circuit requires that a

jury’s determination must be upheld if it issupported by substantial evidence.

9. Patents O312(6)Substantial evidence supported jury

verdict that accused product met ‘‘directcommunication’’ limitation in patents dis-closing secure domain name service; eachclaim required some indication that domainname service system supported establish-ing secure communication link, relay ser-ver created two separate communications,network address translators (NAT) usedby accused products did not impede direct


communication but operated like routersor firewalls, and NAT router still allowedfor ‘‘end-to-end communication betweenthe two devices’’ because it merely trans-lated addresses from public address spaceto private address space, but did not ter-minate the connection.

10. Patents O235(2)

Accused feature in mobile phone man-ufacturer’s product met ‘‘determiningwhether’’ limitation of patents generallydescribing method of transparently creat-ing virtual private network (VPN) betweenclient computer and target computer, sincemanufacturer’s accused feature could beconfigured to infringe based on manufac-turer’s internal documents and source codeand ‘‘determining whether’’ step could beperformed by comparing requested domainname against list of domain names; whilefeature could initiate VPN connectionswith unsecure websites, feature was notintended to be used in that manner andthere was no requirement in claims forverification of security of requested web-site or server.

11. Patents O312(6)

Substantial evidence supported jury’sfinding that feature in accused productcreated ‘‘VPN’’ or ‘‘secure channel’’ thatextended from client to target computerand thus literally infringed claim of patentthat required creating ‘‘secure channel’’‘‘between’’ client and secure server; pathextending from virtual private network(VPN) server to target computer, i.e.,within private network, would be secureand anonymous owing to protection provid-ed by private network, paths beyond VPNserver could be rendered secure and anon-ymous by means of ‘‘physical security’’present in private corporate networks, andpatentee did not have to prove that ac-cused product did not have any non-in-fringing modes of operation.

12. Patents O237Security provided by accused system

that included encryption on insecure pathsbut otherwise relied on security providedby private networks was not equivalent to‘‘encrypted channel’’ required by claim inpatent that generally described method oftransparently creating virtual private net-work (VPN) between client computer andtarget computer, and thus accused productdid not infringe under equivalents doc-trine; ‘‘security’’ of private network couldnot be equated with ‘‘encryption’’ providedby VPN server because encryption wasnarrower, more specific requirement thansecurity, according to patent.

13. Patents O237To find infringement under the equiv-

alents doctrine, any differences betweenthe claimed invention and the accusedproduct must be insubstantial; insubstan-tiality may be determined by whether theaccused device performs substantially thesame function in substantially the sameway to obtain substantially the same resultas the patent claim limitation.

14. Patents O314(5)Under the equivalents doctrine,

whether an accused device performs sub-stantially the same function in substantial-ly the same way to obtain substantially thesame result as the patent claim limitationis a question of fact.

15. Patents O237‘‘Vitiation’’ is not an exception to the

doctrine of equivalents, but instead is alegal determination that the evidence in apatent suit is such that no reasonable jurycould determine two elements to be equiv-alent.

16. Patents O112.5A party challenging the validity of a

patent must establish invalidity by clearand convincing evidence.


17. Patents O324.55(4)

Anticipation in a patent case is a fac-tual question that is reviewed for substan-tial evidence.

18. Patents O72(1)

A patent claim is anticipated only ifeach and every element is found within asingle prior art reference, arranged asclaimed.

19. Patents O62(1)

Substantial evidence supported find-ing that patents generally describingmethod of transparently creating a virtualprivate network (VPN) between a clientcomputer and a target computer and pat-ents disclosing secure domain name ser-vice were not invalid as anticipated; patentowner presented evidence and testimonythat prior publication failed to disclose sev-eral claim limitations including, ‘‘securecommunication link,’’ ‘‘virtual private net-work,’’ ‘‘a DNS proxy server,’’ ‘‘an encrypt-ed channel,’’ and ‘‘secure channel.’’ 35U.S.C.A. § 102(a).

20. Evidence O146

District court would not have abusedits discretion in finding that probative val-ue of evidence that mobile phone manufac-turer initiated re-examinations of patentsgenerally describing method of transpar-ently creating a virtual private network(VPN) between client computer and targetcomputer and patents disclosing secure do-main name service was substantially out-weighed by risk of unfair prejudice to pat-entee, confusion with invalidity on themerits, or misleading jury, thereby justify-ing exclusion in patent infringement ac-tion. Fed.Rules Evid.Rule 403, 28U.S.C.A.

21. Courts O96(7)

In a patent case, regional circuit law isapplied to evidentiary issues.

22. Federal Courts O3598(5)

The Fifth Circuit reviews a districtcourt’s exclusion of relevant evidence onthe basis that its probative value is sub-stantially outweighed by danger of unfairprejudice for clear abuse of discretion re-sulting in substantial prejudice. Fed.Rules Evid.Rule 403, 28 U.S.C.A.

23. Patents O318(4.1)

Patentee could not rely on entire mar-ket value of multi-component product con-taining several non-infringing featureswith no relation to patented feature toapproximate base for reasonable royaltythat manufacturer of accused productswould have been willing to offer to pay topatentee during hypothetical negotiation,without attempting to apportion value at-tributable to patented features. 35U.S.C.A. § 284.

24. Patents O319(1)

The most common method for deter-mining a reasonable royalty is the hypo-thetical negotiation approach, which at-tempts to ascertain the royalty upon whichthe parties would have agreed had theysuccessfully negotiated an agreement justbefore patent infringement began. 35U.S.C.A. § 284.

25. Patents O319(1)

In a patent case, a reasonable royaltymay be a lump-sum payment not calculat-ed on a per unit basis, but it may also be,and often is, a running payment that varieswith the number of infringing units; in thatevent, it generally has two prongs: a roy-alty base and a royalty rate. 35 U.S.C.A.§ 284.

26. Patents O319(1)

Whatever the form of the royalty, apatentee must take care to seek only thosedamages attributable to the infringing fea-tures. 35 U.S.C.A. § 284.


27. Patents O318(4.1)

When patent claims are drawn to anindividual component of a multi-componentproduct, it is the exception, not the rule,that damages may be based upon the valueof the multi-component product; in the ab-sence of a showing that the patented fea-ture creates the basis for customer de-mand or substantially creates the value ofthe component parts, principles of appor-tionment apply. 35 U.S.C.A. § 284.

28. Patents O319(1)

When determining a reasonable royal-ty in a patent case, absolute precision isnot required when assigning value to afeature that may not have ever been indi-vidually sold; this process may involvesome degree of approximation and uncer-tainty. 35 U.S.C.A. § 284.

29. Evidence O555.9

Testimony of patentee’s damages ex-pert that relied on entire value of accuseddevices and computers as ‘‘smallest salableunits’’ without attempting to apportion val-ue attributable to patented features wasnot admissible to approximate base forreasonable royalty that manufacturer ofaccused products would have been willingto offer to pay to patentee during hypo-thetical negotiation, for failure to comportwith settled principles of apportionment;where smallest salable unit was multi-com-ponent product containing non-infringingfeatures with no relation to patented fea-ture, patentee had to demonstrate thatpatented features drove demand for ac-cused products. 35 U.S.C.A. § 284; Fed.Rules Evid.Rule 702, 28 U.S.C.A.

30. Evidence O555.2

The admissibility of expert testimonyis governed by the Federal Rules of Evi-dence and the principles laid out in Dau-bert v. Merrell Dow Pharmaceuticals.Fed.Rules Evid.Rule 702, 28 U.S.C.A.

31. Evidence O555.2A district court’s ‘‘gatekeeping obli-

gation’’ applies to all types of expert testi-mony. Fed.Rules Evid.Rule 702, 28U.S.C.A.

32. Evidence O555.9While questions regarding which facts

are most relevant for calculating a reason-able royalty in a patent case are properlyleft to the jury, a critical prerequisite isthat the underlying methodology be sound.35 U.S.C.A. § 284; Fed.Rules Evid.Rule702, 28 U.S.C.A.

33. Patents O319(1)When calculating a reasonable royal-

ty, a patentee’s obligation to apportiondamages only to the patented featuresdoes not end with the identification of thesmallest salable unit if that unit still con-tains significant unpatented features. 35U.S.C.A. § 284.

34. Patents O318(4.1)When calculating a reasonable royalty

in a patent case, there is no necessity-based exception to the entire market valuerule; a patentee must be reasonable,though may be approximate, when seekingto identify a patent-practicing unit, tangi-ble or intangible, with a close relation tothe patented feature. 35 U.S.C.A. § 284.

35. Patents O319(1)Patentee must apportion the royalty

down to a reasonable estimate of the valueof its claimed technology, or else establishthat its patented technology drove demandfor the entire product. 35 U.S.C.A. § 284.

36. Evidence O555.9District court did not abuse its discre-

tion in patent case by permitting testimo-ny from damages expert regarding properroyalty rate based on allegedly comparablelicenses; four licenses related to actual pat-ents-in-suit and others were drawn to re-


lated technology, and differences werepresented to jury to allow jury to fullyevaluate relevance of licenses. 35U.S.C.A. § 284.

37. Patents O319(1)When relying on licenses to prove a

reasonable royalty, alleging a loose orvague comparability between differenttechnologies or licenses does not suffice.35 U.S.C.A. § 284.

38. Evidence O555.9Evidence relying on 50-50 starting

point based on bargaining solution theo-rem was not admissible under Daubert toapproximate reasonable royalty rate thatmanufacturer of accused product wouldhave been willing to offer to pay to paten-tee during hypothetical negotiation, sincepatentee did not sufficiently show howpremises of theorem actually applied tospecific facts of case at hand. 35 U.S.C.A.§ 284; Fed.Rules Evid.Rule 702, 28U.S.C.A.

39. Patents O319(1)A patentee may not balance out an

unreasonably high royalty base simply byasserting a low enough royalty rate.

Patents O328(2)6,502,135, 7,418,504, 7,490,151, 7,921,-

211. Valid and Infringed.

William F. Lee, Wilmer Cutler Picker-ing Hale and Dorr, LLP, of Boston, MA,argued for defendant-appellant. With himon the brief were Mark C. Fleming, Lau-ren B. Fletcher, and Rebecca Bact, of Bos-ton, MA, and Jonathan G. Cedarbaum,Brittany Blueitt Amadi, and Leah Litman,

of Washington, DC. Of counsel on the briefwas Danny L. Williams, Williams, Morgan& Amerson, P.C., of Houston, TX.

J. Michael Jakes, Finnegan, Henderson,Farabow, Garrett & Dunner, LLP, ofWashington, DC, argued for plaintiffs-ap-pellees. With him on the brief for Virnetx,Inc. were Kara F. Stoll and Srikala Atluri,of Washington, DC, and Benjamin R.Schlesinger, of Atlanta, GA. Of counsel onthe brief were Bradley W. Caldwell, JasonD. Cassady, and John Austin Curry, Cald-well, Cassady & Curry, of Dallas, TX. Onthe brief for Science Applications Interna-tional Corporation were Donald Urrabazo,Arturo Padilla, and Ronald Wielkopolski,Urrabazo Law, P.C., of Los Angeles, CA;and Andy Tindel, Mann, Tindel & Thomp-son, of Tyler, TX.

Before PROST,* Chief Judge andCHEN, Circuit Judge.**

PROST, Chief Judge.

Apple Inc. appeals from a final judg-ment of the U.S. District Court for theEastern District of Texas, in which a juryfound that Apple infringed U.S. PatentNos. 6,502,135 (‘‘8135 patent’’), 7,418,504(‘‘8504 patent’’), 7,490,151 (‘‘8151 patent’’),and 7,921,211 (‘‘8211 patent’’). The juryfurther found that none of the infringedclaims were invalid and awarded damagesto plaintiffs-appellees VirnetX, Inc. andScience Applications International Corpo-ration (‘‘SAIC’’) in the amount of$368,160,000.

For the reasons that follow, we affirmthe jury’s findings that none of the assert-ed claims are invalid and that many of theasserted claims of the 8135 and 8151 pat-ents are infringed by Apple’s VPN On

* Sharon Prost assumed the position of ChiefJudge on May 31, 2014.

** Randall R. Rader, who retired from the posi-tion of Circuit Judge on June 30, 2014, didnot participate in this decision.


Demand product. We also affirm the dis-trict court’s exclusion of evidence relatingto the reexamination of the patents-in-suit.However, we reverse the jury’s findingthat the VPN On Demand product in-fringes claim 1 of the 8151 patent underthe doctrine of equivalents. We also re-verse the district court’s construction ofthe claim term ‘‘secure communicationlink’’ in the 8504 and 8211 patents andremand for further proceedings to deter-mine whether the FaceTime feature in-fringes those patents under the correctclaim construction. Finally, we vacate thejury’s damages award and remand for fur-ther proceedings consistent with this opin-ion.

BACKGROUND

The patents at issue claim technologyfor providing security over networks suchas the Internet. The patents assert pri-ority to applications filed in the 1990s,originally assigned to SAIC. VirnetX, aNevada-based software development andlicensing enterprise, acquired the patentsfrom SAIC in 2006.

I. The 8504 and 8211 Patents and Face-Time

The 8504 and 8211 patents share a com-mon specification disclosing a domainname service (‘‘DNS’’) system that re-solves domain names and facilitates estab-lishing ‘‘secure communication links.’’ 8504patent col. 55 ll. 49–50. In one embodi-ment, an application on the client comput-er sends a query including the domainname to a ‘‘secure domain name service,’’which contains a database of secure do-main names and corresponding secure net-work addresses. Id. at col. 50 ll. 54–57,col. 51 ll. 11–19, col. 51 ll. 29–32. Thisallows a user to establish a secure commu-nication link between a client computer

and a secure target network address. Id.at col. 51 ll. 34–40.

Representative claim 1 of the 8504 pat-ent recites:

1. A system for providing a domainname service for establishing a securecommunication link, the system compris-ing:a domain name service system config-ured to be connected to a communicationnetwork, to store a plurality of domainnames and corresponding network ad-dresses, to receive a query for a networkaddress, and to comprise an indicationthat the domain name service systemsupports establishing a secure communi-cation link.

Id. at col. 55 ll. 49–56.

Before the district court, VirnetX ac-cused Apple of infringement based on its‘‘FaceTime’’ feature. Specifically, VirnetXaccused Apple’s servers that run Face-Time on Apple’s iPhone, iPod, iPad (collec-tively, ‘‘iOS devices’’), and Mac computersof infringing claims 1, 2, 5, 16, 21, and 27 ofthe 8504 patent as well as claims 36, 37, 47,and 51 of the 8211 patent. In operation,FaceTime allows secure video calling be-tween select Apple devices. J.A. 1443. Touse FaceTime, a caller enters an intendedrecipient’s e-mail address or telephonenumber into the caller’s device (e.g.,iPhone). J.A. 1451–52. An invitation isthen sent to Apple’s FaceTime server,which forwards the invitation to a networkaddress translator (‘‘NAT’’) which, in turn,readdresses the invitation and sends it onto the receiving device. J.A. 1821, 1824–25. The recipient may then accept or de-cline the call. J.A. 1453. If accepted,FaceTime servers establish a secure Face-Time call. J.A. 1453. Once connected, thedevices transmit audio/video data as pack-ets across the secure communication pathwithout passing through the FaceTimeserver. J.A. 1820, 1825.


II. The 8135 and 8151 Patents and VPNOn Demand

A conventional DNS resolves domainnames (e.g., ‘‘Yahoo.com’’) into InternetProtocol (‘‘IP’’) addresses. See 8135 patentcol. 37 ll. 22–27. A user’s web browserthen utilizes the IP address to request awebsite. Id. at col. 37 ll. 24–29.

The 8135 and 8151 patents share a com-mon specification disclosing a system inwhich, instead of a conventional DNS re-ceiving the request, a DNS proxy inter-cepts it and determines whether the re-quest is for a secure site. Id. at col. 38 ll.23–25. If the proxy determines that arequest is for a secure site, the systemautomatically initiates a virtual privatenetwork (‘‘VPN’’) between the proxy andthe secure site. Id. at col. 38 ll. 30–33. Ifthe browser determines that the requestwas for a non-secure website, then theDNS proxy forwards the request to a con-ventional DNS for resolution. Id. at col.38 ll. 43–47.

Representative claim 1 of the 8135 pat-ent recites:

1. A method of transparently creatinga virtual private network (VPN) be-tween a client computer and a targetcomputer, comprising the steps of:

(1) generating from the client computera Domain Name Service (DNS) requestthat requests an IP address correspond-ing to a domain name associated withthe target computer;

(2) determining whether the DNS re-quest transmitted in step (1) is request-ing access to a secure web site; and

(3) in response to determining that theDNS request in step (2) is requestingaccess to a secure target web site, auto-matically initiating the VPN between the

client computer and the target comput-er.

Id. at col. 47 ll. 20–32.

Claims 1 and 13 of the 8151 patent aresimilar to claim 1 of the 8135 patent exceptthat they recite initiating an ‘‘encryptedchannel’’ and creating a ‘‘secure channel,’’respectively, instead of creating a ‘‘VPN.’’8151 patent col. 46 ll. 55–67, col. 48 ll. 18–29.

Before the district court, VirnetX ac-cused Apple’s iPhone, iPad, and iPodTouch of infringing claims 1, 3, 7, and 8 ofthe 8135 patent and claims 1 and 13 of the8151 patent because they include a featurecalled ‘‘VPN On Demand.’’ When a userenters a domain name into the browser ofan iOS device, a DNS request is generat-ed. J.A. 1393–94. VPN On Demand re-ceives the request and checks a list ofdomain names for which a VPN connectionshould be established, known as a ‘‘config-uration file.’’ J.A. 1377. If the entereddomain name matches a domain name inthe configuration file, VPN On Demandcontacts a VPN server to authenticate theuser and, if successful, automatically estab-lishes a VPN between the user’s browserand the target computer with which therequested domain name is associated.J.A. 1377–78, 1396–98.

III. Five–Day Jury Trial and Post–Tri-al Motions

On August 11, 2010, VirnetX filed thisinfringement action, alleging that Apple’sFaceTime servers infringe certain claimsof the 8504 and 8211 patents, and thatApple’s VPN On Demand feature infringescertain claims of the 8135 and 8151 patents.Apple responded that FaceTime and VPNOn Demand do not infringe, and that theasserted claims were invalid as anticipatedby a 1996 publication by Takahiro Kiuchiet al. (‘‘Kiuchi’’).


On April 25, 2012, the district court con-strued disputed claim terms, and a jurytrial commenced on October 31, 2012. Af-ter a five-day trial, the jury returned itsverdict, finding all of the asserted claimsvalid and infringed. The jury awardedVirnetX $368,160,000 in reasonable royaltydamages. Apple moved for judgment as amatter of law (‘‘JMOL’’) or, alternatively,for a new trial or remittitur. On February26, 2013, the district court denied Apple’smotions. VirnetX, Inc. v. Apple Inc., 925F.Supp.2d 816 (E.D.Tex.2013).

Apple now appeals the denial of its post-trial motion for JMOL or a new trial.This court has jurisdiction under 28 U.S.C.§ 1295(a)(1).

DISCUSSION

I. Claim Construction

On appeal, Apple argues that the districtcourt erred in construing the terms ‘‘do-main name’’ and ‘‘secure communicationlink,’’ both recited in the 8504 and 8211patents. For the reasons that follow, weaffirm the construction of ‘‘domain name’’and reverse the construction of ‘‘securecommunication link.’’

[1–4] Claim construction is a questionof law that we review de novo. LightingBallast Control LLC v. Philips Elecs. N.Am. Corp., 744 F.3d 1272, 1276–77 (Fed.Cir.2014) (en banc); Cybor Corp. v. FASTechs., Inc., 138 F.3d 1448, 1456 (Fed.Cir.1998) (en banc). The process of constru-ing a claim term begins with the words ofthe claims themselves. See VitronicsCorp. v. Conceptronic, Inc., 90 F.3d 1576,1582 (Fed.Cir.1996); Phillips v. AWHCorp., 415 F.3d 1303, 1312–14 (Fed.Cir.2005) (en banc). However, the claims‘‘must be read in view of the specification,of which they are a part.’’ Phillips, 415F.3d at 1315 (quoting Markman v. West-view Instruments, Inc., 52 F.3d 967, 979

(Fed.Cir.1995) (en banc), aff’d 517 U.S.370, 116 S.Ct. 1384, 134 L.Ed.2d 577(1996)). Additionally, the doctrine of claimdifferentiation disfavors reading a limita-tion from a dependent claim into an inde-pendent claim. See InterDigitalCommc’ns, LLC v. Int’l Trade Comm’n,690 F.3d 1318, 1324 (Fed.Cir.2012). Al-though courts are permitted to considerextrinsic evidence like expert testimony,dictionaries, and treatises, such evidence isgenerally of less significance than the in-trinsic record. Phillips, 415 F.3d at 1317(citing C.R. Bard, Inc. v. U.S. SurgicalCorp., 388 F.3d 858, 862 (Fed.Cir.2004)).

A. ‘‘Domain Name’’

[5] The district court construed ‘‘do-main name’’ as ‘‘a name corresponding toan IP address.’’ Memorandum Opinion &Order at 16, VirnetX, Inc. v. Cisco Sys.Inc., No. 6:10–cv–416 (E.D.Tex. Apr. 25,2012), ECF No. 266 (‘‘Claim ConstructionOrder ’’). Apple argues, as it did below,that the proper construction is ‘‘a hierar-chical sequence of words in decreasing or-der of specificity that corresponds to anumerical IP address.’’ Apple insists thatits construction represents the plain andordinary meaning of the term, relying pri-marily on a technical dictionary definitionand several examples in the specification(e.g., ‘‘Yahoo.com’’). We disagree. Intrin-sic evidence supports the district court’sconstruction of ‘‘domain name.’’ The spec-ification of the 8504 and 8211 patents sug-gests the use of the invention for securecommunications between application pro-grams like ‘‘video conferencing, e-mail,word processing programs, telephony, andthe like.’’ 8504 patent col. 21 ll. 27–29.The disclosure of such applications demon-strates that the inventors did not intend tolimit ‘‘domain name’’ to the particular for-matting limitations of websites sought byApple, i.e., a top-level domain, second-leveldomain, and host name.


Additionally, fundamental principles ofclaim differentiation disfavor reading Ap-ple’s hierarchical limitation into the inde-pendent claims. Dependent claims in bothpatents require that ‘‘at least one’’ of thedomain names stored by the system com-prise a top-level domain name. See, e.g.,8504 patent col. 55 ll. 57–59 (‘‘The systemof claim 1, wherein at least one of theplurality of domain names comprises a top-level domain name.’’); 8211 patent col. 57ll. 47–50 (‘‘The non-transitory machine-readable medium of claim 36, wherein theinstructions comprise code for storing theplurality of domain names and correspond-ing network addresses including at leastone top-level domain name.’’). The specificlimitation of hierarchical formatting in thedependent claims strongly suggests thatthe independent claims contemplate do-main names both with and without thehierarchical format exemplified by ‘‘Ya-hoo.com.’’ See InterDigital, 690 F.3d at1324 (‘‘The doctrine of claim differentiationis at its strongest TTT ‘where the limitationthat is sought to be ‘‘read into’’ an inde-pendent claim already appears in a depen-dent claim.’ ’’ (quoting Liebel–FlarsheimCo. v. Medrad, Inc., 358 F.3d 898, 910(Fed.Cir.2004))).

Such intrinsic evidence is not out-weighed by the extrinsic evidence of onedictionary definition. This is particularlytrue here, where the dictionary definitionseems to contemplate web addresses onthe Internet, while the specification makesclear that the claim term in question is notso limited. See J.A. 6139–40. Thus, weaffirm the district court’s construction ofthe term ‘‘domain name’’ as ‘‘a name corre-sponding to an IP address.’’

B. ‘‘Secure Communication Link’’

[6] The district court construed ‘‘se-cure communication link’’ as ‘‘a direct com-munication link that provides data securi-ty.’’ Claim Construction Order at 13.Apple argues that this term should beconstrued consistent with ‘‘VPN,’’ whichthe district court construed to require notonly data security but also anonymity.1

As an initial matter, we note that thereis no dispute that the word ‘‘secure’’ doesnot have a plain and ordinary meaning inthis context, and so must be defined byreference to the specification. See OralArg. 31:50–32:40, available at http://www.cafc.uscourts.gov/oral-argument-recordings/13-1489/all (acknowledgementby VirnetX’s counsel that construction of‘‘secure’’ requires consideration of thespecification).

Moreover, we agree with Apple that,when read in light of the entire specifica-tion, the term ‘‘secure communication link’’requires anonymity. Indeed, the additionof anonymity is presented as one of theprimary inventive contributions of the pat-ent. For example, the Background of theInvention states that ‘‘[a] tremendous vari-ety of methods have been proposed andimplemented to provide security and ano-nymity for communications over the Inter-net.’’ 8504 patent col. 1 ll. 32–35 (emphasisadded). It goes on to define these twoconcepts as counterpart safeguards againsteavesdropping that could occur while twocomputer terminals communicate over theInternet. Id. at col. 1 ll. 38–54. Securityin this context refers to protection of dataitself, to preserve the secrecy of its con-tents, while anonymity refers to prevent-ing an eavesdropper from discovering the

1. The district court construed VPN to mean‘‘a network of computers which privately anddirectly communicate with each other by en-crypting traffic on insecure paths between the

computers where the communication is bothsecure and anonymous.’’ Claim ConstructionOrder at 8.


identity of a participating terminal. Id. atcol. 1 ll. 40–54.

Having thus framed the problem, thepatent (as expected) proposes a solution.Specifically, the Summary of the Inventionbegins by explaining how the inventionimproves security by using a ‘‘two-layerencryption format’’ known as the TunneledAgile Routing Protocol, or TARP. Id. atcol. 3 ll. 14–17. First, an ‘‘inner layer’’secures the data itself, id. at col. 4 ll. 5–7,and then a second ‘‘outer layer’’ concealsthe data’s ‘‘true destination,’’ id. at col. 3 ll.34–35. The fact that the Summary of theInvention gives primacy to these attributesstrongly indicates that the invention re-quires more than just data security. See,e.g., C.R. Bard, 388 F.3d at 864 (givingparticular weight to statements in theSummary of the Invention because‘‘[s]tatements that describe the inventionas a whole, rather than statements thatdescribe only preferred embodiments, aremore likely to support a limiting definitionof a claim term’’).

Consistent with this emphasis, the De-tailed Description states that ‘‘the messagepayload is embedded behind an inner layerof encryption’’ and ‘‘[e]ach TARP packet’strue destination is concealed behind anouter layer of encryption.’’ 8504 patentcol. 9 ll. 60–61, col. 11 ll. 2–4. The conceal-ment requirement appears throughout thespecification and is implicated in every em-bodiment associated with the ‘‘two-layerencryption’’ or TARP VPN. The fact thatanonymity is ‘‘repeatedly and consistently’’used to characterize the invention stronglysuggests that it should be read as part ofthe claim. See Eon–Net LP v. FlagstarBancorp., 653 F.3d 1314, 1321–23 (Fed.Cir.2011).

VirnetX attempts to rebut this sugges-tion by pointing to a single place in thespecification where a ‘‘secure communica-tion path’’ is referred to as providing only

security, without anonymity. See 8504 pat-ent col. 39 ll. 24–35. But that disclosurerelates to the ‘‘conventional architecture’’of the prior art that suffers precisely be-cause it ‘‘hamper[s] anonymous communi-cations on the Internet.’’ Id. at col. 39 ll.24, 32–33. And indeed, the specificationgoes on to explain how the invention solvesthat very problem by setting up a VPN,which requires anonymity. Id. at col. 39 ll.46–62.

VirnetX also argues that the specifica-tion teaches that different users have ‘‘dif-ferent needs’’ such that some users needdata security while, in other cases, ‘‘itmay be desired’’ to also have anonymity.Appellee’s Br. 48 (citing 8504 patent col. 1ll. 33–52). Thus, VirnetX insists, theTARP protocol (with its requirement ofanonymity) is but one type of ‘‘securecommunication link,’’ and does not limitthe construction of that term. To be sure,the specification mechanically prefacesmost passages with the phrase ‘‘accordingto one aspect of the present invention.’’See, e.g., 8504 patent col. 6 l. 36. But theBackground and Summary of the Inven-tion clearly identify the TARP protocol asa key part of the novel solution to the spe-cific problem identified in the prior art.Unsurprisingly, therefore, VirnetX has notidentified even a single embodiment thatprovides data security but not anonymity.

Moreover, in several instances the speci-fication appears to use the terms ‘‘securecommunication link’’ and ‘‘VPN’’ inter-changeably, suggesting that the inventorsintended the disputed term to encompassthe anonymity provided by a VPN. SeeNystrom v. Trex Co., 424 F.3d 1136, 1143(Fed.Cir.2005) (‘‘Different terms or phras-es in separate claims may be construed tocover the same subject matter where thewritten description and prosecution historyindicate that such a reading of the termsor phrases is proper.’’). For example, it


states that ‘‘[w]hen software module 3309is being installed or when the user is off-line, the user can optionally specify that allcommunication links established over com-puter network 3302 are secure communi-cation links. Thus, anytime that a com-munication link is established, the link is aVPN link.’’ 8504 patent col. 52 ll. 15–19(emphases added). Similarly, in the verynext paragraph the specification statesthat ‘‘a user at computer 3301 can optional-ly select a secure communication linkthrough proxy computer 3315. According-ly, computer 3301 can establish a VPNcommunication link 3323 with secure ser-ver computer 3320 through proxy comput-er 3315.’’ Id. at col. 52 ll. 25–29 (emphasesadded). In both of these instances, thespecification equates the term ‘‘securecommunication link’’ with a ‘‘VPN.’’ Theonly counter-example VirnetX can point tois an instance where the specificationstates, in relation to one aspect of theinvention, that ‘‘[t]he secure communica-tion link is a virtual private network com-munication link over the computer net-work.’’ Id. at col. 6 ll. 61–63. But equatingthe two terms with respect to one aspect ofthe present invention is a far cry fromexpressly divorcing those terms elsewhere,particularly in the absence of any embodi-ment or disclosure that does so.

Thus, we reverse the district court’sclaim construction and conclude that theterm ‘‘secure communication link’’ as usedin the 8504 and 8211 patents requires ano-nymity. Accordingly, the term should beconstrued as ‘‘a direct communication linkthat provides data security and anonymi-ty.’’

II. Infringement

[7, 8] We review the denial of a motionfor JMOL or a new trial under the law ofthe regional circuit. Verizon Servs. Corp.v. Cox Fibernet Va., Inc., 602 F.3d 1325,

1331 (Fed.Cir.2010). The Fifth Circuit re-quires that a jury’s determination must beupheld if it is supported by substantialevidence. ClearValue, Inc. v. Pearl RiverPolymers, Inc., 668 F.3d 1340, 1343 (Fed.Cir.2012) (citing Med. Care Am., Inc. v.Nat’l Union Fire Ins. Co., 341 F.3d 415,420 (5th Cir.2003)).

A. 8504 and 8211 Patents

[9] Apple argues that there was notsubstantial evidence to support the jury’sverdict that its FaceTime servers infringethe asserted claims of the 8504 and 8211patents. Apple insists that FaceTime doesnot infringe the ‘‘secure communicationlink’’ claim term for two reasons: first,because when properly construed it re-quires anonymity, which the FaceTimeservers do not provide, and second, be-cause they do not provide ‘‘direct’’ commu-nication, as required by the district court’sclaim construction.

With respect to the first argument, wehave now construed the disputed claimterm so as to require anonymity. Seesupra at 1319. However, the jury was notpresented with the question of whetherFaceTime infringes the asserted claimsunder a construction requiring anonymity.Thus, we remand for further proceedingsto determine whether Apple’s FaceTimeservers provide anonymity.

With respect to the second argument,Apple argues that FaceTime servers donot provide ‘‘direct’’ communication be-cause the communications are addressed toa NAT, rather than to the receiving device.Appellant’s Br. 43. The district court con-cluded that there was substantial evidenceto support the jury’s finding that the NATrouters used by FaceTime do not impededirect communication, VirnetX, 925F.Supp.2d at 831, and we agree. As thedistrict court noted, VirnetX’s expert testi-fied that the NAT routers still allow for


‘‘end-to-end communication between thetwo devices,’’ J.A. 1565, because theymerely translate addresses from the publicaddress space to the private address space,but do not terminate the connection. J.A.1465, 1536–37. Even Apple’s expert ad-mitted that the connection does not stop atthe NAT routers. J.A.1984.

Apple argues that this testimony cannotsupport a finding of infringement becauseit is inconsistent with the court’s claimconstruction that required ‘‘direct address-ability.’’ Appellant’s Br. 43–45. But thedistrict court considered this argumentand disagreed, noting that its claim con-struction expressly provided that ‘‘routers,firewalls, and similar servers TTT do notimpede ‘direct’ communication,’’ and Vir-netX presented evidence that NATs oper-ate like routers or firewalls. VirnetX, 925F.Supp.2d at 831.

Thus, we do not think the district courterred in finding that there was substantialevidence on which the jury could haverelied to reach its finding of infringementon this element.

B. 8135 and 8151 Patents

Apple also argues that there was notsubstantial evidence to support the jury’sverdict that its VPN On Demand productinfringed the asserted claims of the 8135and 8151 patents for several reasons, dis-cussed in turn below.

1. ‘‘Determining Whether’’

[10] Apple argues that its VPN OnDemand feature does not infringe the as-serted claims of the 8135 and 8151 patentsbecause it does not ‘‘determine whether’’ arequested domain name is a secure web-site or server. Instead, Apple insists thatit merely determines whether the request-ed website is listed in the user-created‘‘configuration file’’ and initiates a VPNconnection for any domain name on that

list, regardless of whether or not it issecure. In response, VirnetX argues thatthere was substantial evidence demon-strating that the VPN On Demand systemis designed and intended to be used onlyfor accessing secure private networks.We agree with VirnetX.

Here, the evidence presented at trialsupports the conclusion that Apple’s VPNOn Demand product infringes the assertedclaim limitation in its normal configuration.In particular, VirnetX’s expert testifiedthat Apple’s technical design documentsand internal technical presentations relat-ing to the VPN On Demand system (manyof which are confidential and cannot bequoted here) make clear that a VPN con-nection should only be established for pri-vate web addresses. Thus, regardless ofwhether a user could misconfigure the listby entering public domain names, Apple’splanning documents, internal emails, andpresentations all explained that VPN OnDemand’s primary use is to connect usersto secure sites using a VPN. That is allthat is required. See Hilgraeve Corp. v.Symantec Corp., 265 F.3d 1336, 1343 (Fed.Cir.2001).

Moreover, this description of the VPNOn Demand feature is consistent with howthe claimed functionality is described inthe specification. For example, in one em-bodiment, the DNS proxy determineswhether a request is for a secure site bychecking the domain name against a tableor list of domain names. 8135 patent col.38 ll. 23–30. In other words, the proxyidentifies a request for ‘‘access to a securesite TTT by reference to an internal table ofsuch sites.’’ Id. That is precisely how theVPN On Demand feature operates.

We therefore conclude that the jury’sfinding that the VPN On Demand productinfringes the ‘‘determining whether’’ limi-


tation was supported by substantial evi-dence.

2. ‘‘Between’’

a. Literal Infringement of Claim 1of the 8135 Patent and Claim

13 of the 8151 Patent

[11] Claim 1 of the 8135 patent re-quires creating a ‘‘VPN’’ ‘‘between’’ theclient and a target computer. 8135 patentcol. 47 ll. 20–22. Similarly, claim 13 of the8151 patent requires creating a ‘‘securechannel’’ ‘‘between’’ the client and the se-cure server. 8151 patent col. 48 ll. 28–29.For both claims, the district court con-strued ‘‘between’’ to mean ‘‘extendingfrom’’ the client to the target computer.Claim Construction Order at 26.

Apple argues that its VPN On Demandproduct fails to meet this limitation be-cause it only secures communications be-tween the iPhone and the VPN server, butnot between the VPN server and the tar-get computer. VirnetX responds that Ap-ple’s product is intended to be used withprivate networks, which are generally con-figured to be both secure and anonymous.In other words, VirnetX argues that thesecure channel between the VPN serverand the target computer is provided by thetarget computer itself. After consideringthe record as a whole, we conclude thatthere was substantial evidence to supportthe jury’s verdict of infringement on thislimitation.

At trial, VirnetX presented evidence andtestimony to the jury that ‘‘the virtualprivate network extend[s] from the clientcomputer to the target computer TTT be-cause it’s encrypted on the insecure paths,and it’s secure within the corporate net-work.’’ J.A. 1400–01. VirnetX’s experttestified that one of ordinary skill wouldunderstand that the path extending fromthe VPN server to the target computer,i.e., within the private network, would be

secure and anonymous owing to protectionprovided by the private network. J.A.1080 (‘‘That network is secure, because it’sbeen physically secured; and it also haswhat’s called a firewall between its net-work and the public network. So it keepsthe bad guys out.’’); J.A. 1379 (‘‘If that’s aprivate network of the company thatthey’ve set up behind a VPN server, thecompany would have configured that to besecure.’’); J.A. 1396 (‘‘[T]hese are TTT pri-vate networks that are not to be accessedby others. They require authorization foraccess.’’). The jury also heard testimonythat while in some situations traffic couldbe unsecured behind the VPN server, J.A.1997–98, this scenario would be ‘‘atypical.’’J.A.1992–93. For example, VirnetX pre-sented evidence to the jury that Appleitself advertised that VPN On Demand isdesigned to connect with ‘‘private corpo-rate networks’’ and ‘‘works with a varietyof authentication methods.’’ J.A. 20001.And, more to the point, the jury heard thatthe ‘‘private corporate networks’’ to whichVPN On Demand is intended to connectemploy security measures including VPNservers, VPN authentication servers,proxy servers, and firewalls which regulateaccess to private resources and preventunauthorized users from breaching. J.A.1080, 1379, 1401.

Apple argues that this finding of in-fringement necessarily rests on a series of‘‘assumptions’’ about how all private net-works operate in order to conclude thatVPN On Demand is ‘‘typically’’ configuredto operate in the manner accused of in-fringement. Appellant’s Br. 30–31. How-ever, VirnetX’s expert relied on Apple’sown internal technical documentation,product specifications, and marketing pres-entations, several of which describe specif-ic security measures used by the privatenetworks to which VPN On Demand isintended to connect. This evidence dem-


onstrates not only that VPN On Demandmay be configured to interact with privatenetworks, but that this was apparently Ap-ple’s primary objective. Apple would haveVirnetX prove that VPN On Demand hasno non-infringing modes of operation.But, as noted above, VirnetX bears nosuch burden. See supra at 1320–21; seealso z4 Techs., Inc. v. Microsoft Corp., 507F.3d 1340, 1350 (Fed.Cir.2007) (‘‘[I]n-fringement is not avoided merely becausea non-infringing mode of operation is pos-sible.’’). We cannot agree that the jury’sfinding lacks substantial evidence becauseVirnetX did not specifically disprove thatVPN On Demand can, in atypical situa-tions, establish a VPN with insecure net-works.

Apple also responds that this evidence isinsufficient because VirnetX’s expert testi-fied that VPN On Demand only encryptscommunications between the iPhone andthe VPN server—by implication leavingthe path from the VPN server to the tar-get unencrypted. Appellant’s Br. 29(quoting J.A. 1392). However, the districtcourt’s construction of ‘‘VPN’’ does notrequire that traffic on a secure path beencrypted. Rather, the construction onlyrequires encryption of traffic ‘‘on insecurepaths.’’ Claim Construction Order at 8.Moreover, as indicated by the 8135 patent,encryption is just one possible way to ad-dress data security. 8135 patent col. 1 ll.38–39 (‘‘Data security is usually tackledusing some form of data encryption.’’ (em-phasis added)). And VirnetX providedsubstantial evidence for the jury to con-clude that paths beyond the VPN servermay be rendered secure and anonymousby means of ‘‘physical security’’ present inthe private corporate networks connectedto by VPN On Demand. See, e.g., J.A.1401.

Accordingly, we conclude that the jury’sfinding that the VPN On Demand feature

creates a ‘‘VPN’’ or a ‘‘secure channel’’that extends from the client to the targetcomputer was supported by substantial ev-idence. We therefore affirm the districtcourt’s denial of JMOL as to claim 1 of the8135 patent and claim 13 of the 8151 patent.

b. Infringement of Claim 1 of the 8151Patent Under the Doctrine of

Equivalents

[12] Claim 1 of the 8151 patent is simi-lar to claim 13 except that it requiresinitiating an ‘‘encrypted channel’’—ratherthan a ‘‘secure channel’’—‘‘between’’ theclient and the secure server. 8151 patentcol. 46 ll. 66–67. With respect to infringe-ment, VirnetX conceded that VPN On De-mand does not literally practice this limita-tion because the private network betweenthe VPN server and the target is ‘‘notnecessarily encrypted’’ from end to end.J.A. 1420–21. Rather, VirnetX assertedthat VPN On Demand infringes under thedoctrine of equivalents because the differ-ence between secure communication viaencryption and secure communication ingeneral is insubstantial. J.A. 1421–24.Apple argues that VirnetX’s theory ofequivalents is legally insufficient because itvitiates the ‘‘encrypted channel’’ element.Appellant’s Br. 32–33.

[13–15] To find infringement under thedoctrine of equivalents, any differences be-tween the claimed invention and the ac-cused product must be insubstantial. SeeGraver Tank & Mfg. Co. v. Linde AirProds. Co., 339 U.S. 605, 608, 70 S.Ct. 854,94 L.Ed. 1097 (1950). Insubstantialitymay be determined by whether the ac-cused device performs substantially thesame function in substantially the sameway to obtain substantially the same resultas the claim limitation. Crown PackagingTech., Inc. v. Rexam Beverage Can Co.,559 F.3d 1308, 1312 (Fed.Cir.2009). Thisis a question of fact. Anchor Wall Sys.,


Inc. v. Rockwood Retaining Walls, Inc.,340 F.3d 1298, 1313 (Fed.Cir.2003). Viti-ation is not an exception to the doctrine ofequivalents. Deere & Co. v. Bush Hog,LLC, 703 F.3d 1349, 1356 (Fed.Cir.2012).Rather, it is a legal determination that‘‘the evidence is such that no reasonablejury could determine two elements to beequivalent.’’ Id. (citation omitted).

After considering the record as a whole,we conclude that the evidence presented attrial does not support the jury’s finding ofinfringement under the doctrine of equiva-lents. VirnetX’s expert testified that VPNOn Demand (a) performs substantially thesame function because it secures the com-munication between the client and the se-cure server, (b) does so in substantially thesame way by protecting data through en-cryption on insecure paths that are vulner-able to eavesdroppers, and (c) achievessubstantially the same result of successful-ly protecting the entire communicationpath from potential eavesdroppers. SeeJ.A. 1424.

Notably, in explaining the ‘‘way’’ thatVPN On Demand secures communications,the expert did not testify that VPN OnDemand provides encryption on the alleg-edly secure pathway between the VPNserver and the private network, but onlyon the insecure portion of the pathway.Thus, his testimony effectively equates the‘‘security’’ of the private network with the‘‘encryption’’ provided by the VPN server.But the patent consistently differentiatesbetween ‘‘security’’ and ‘‘encryption.’’Both the claims and the specification of the8151 patent make clear that encryption is anarrower, more specific requirement thansecurity. For example, the specificationstates that encryption is just one possibleway to address data security. 8151 patentcol. 1 ll. 49–50 (‘‘Data security is usuallytackled using some form of data encryp-tion.’’ (emphasis added)). Additionally,

one of the primary differences between thesteps performed in claim 1 of the 8151patent and the steps performed in claim 13is that claim 13 requires creating a ‘‘se-cure’’ channel, while claim 1 specificallyrequires that the channel be ‘‘encrypted.’’

In light of these distinctions in the pat-ent itself, the jury’s implicit finding thatVPN On Demand achieves the result ofprotecting communications from eaves-dropping in ‘‘substantially the same way’’as contemplated by the ‘‘encrypted chan-nel’’ claim limitation was not supported byVirnetX’s expert’s testimony. See CrownPackaging, 559 F.3d at 1312. No reason-able jury could have determined that thesecurity provided by the VPN On Demandsystem—which includes encryption on theinsecure paths but otherwise relies on thesecurity provided by private networks—isequivalent to the ‘‘encrypted channel’’ re-quired by claim 1 of the 8151 patent. Thedistrict court’s denial of JMOL as to thatclaim must therefore be reversed.

III. Invalidity

[16–18] A party challenging the validi-ty of a patent must establish invalidity byclear and convincing evidence. See Micro-soft Corp. v. i4i Ltd. P’ship, ––– U.S. ––––,131 S.Ct. 2238, 2242, 180 L.Ed.2d 131(2011). Anticipation is a factual questionthat we review for substantial evidence.SynQor, Inc. v. Artesyn Techs., Inc., 709F.3d 1365, 1373 (Fed.Cir.2013). A claim isanticipated only if each and every elementis found within a single prior art reference,arranged as claimed. See Net MoneyIN,Inc. v. VeriSign, Inc., 545 F.3d 1359, 1369(Fed.Cir.2008).

[19] Apple argues that the assertedclaims are anticipated by the Kiuchi refer-ence. However, we conclude that the juryheard substantial evidence that at leastone element of each asserted claim wasmissing from that reference. With respect


to the 8135, 8504, and 8211 patents, the juryheard evidence that Kiuchi’s proxy serversat least do not teach ‘‘direct communica-tion’’ between a client and target comput-er, which is sufficient to defeat a claim ofanticipation. J.A. 2343–44. Specifically,the jury heard expert testimony that Kiu-chi’s client-side and server-side proxiesterminate the connection, process informa-tion, and create a new connection—actionsthat are not ‘‘direct’’ within the meaning ofthe asserted claims. J.A. 2334–35. Vir-netX distinguished such proxy activitiesfrom the operation of NAT routerswhich—unlike proxy servers in the priorart-do not terminate the connection.

Additionally, with respect to the 8151patent, there was substantial evidence tosupport VirnetX’s argument that Kiuchifails to disclose the requirement that theDNS request be ‘‘sent by a client.’’ 8151patent col. 46 l. 57. Apple argued that the‘‘client-side proxy’’ of Kiuchi meets the‘‘client’’ limitation, but there was evidencethat the ‘‘client’’ of Kiuchi is actually a webbrowser, a component that is distinguish-able from the client-side proxy. See J.A.2341. Thus, the district court did not errin denying Apple’s JMOL motion with re-spect to invalidity.

IV. Exclusion of Evidence

[20] At trial, to prove induced infringe-ment, VirnetX attempted to show that Ap-ple knew or was willfully blind to the factthat its customers’ use of its productswould infringe valid patent claims. In de-fense, Apple sought to inform the jurythat, after learning of VirnetX’s allega-tions, Apple initiated reexaminationsagainst the asserted patents. Apple’s re-quests for reexamination resulted in initialrejections of the asserted claims at theUnited States Patent and Trademark Of-fice (‘‘PTO’’). Apple offered these rejec-tions as evidence of Apple’s reasonably-

held belief that the patents were invalid.The district court, however, excluded thisproffer, concluding that such non-final ac-tions in pending reexaminations would be‘‘highly prejudicial evidence that risks mis-leading the jury.’’ VirnetX, 925 F.Supp.2dat 842.

[21, 22] We apply regional circuit lawto evidentiary issues. The Fifth Circuitreviews a district court’s exclusion of evi-dence under Federal Rule of Evidence 403for ‘‘clear abuse of discretion’’ resulting insubstantial prejudice. Wellogix, Inc. v.Accenture, L.L.P., 716 F.3d 867, 882 (5thCir.2013). In this case, we cannot say thatthe district court abused its discretion inexcluding this evidence.

Apple asserts that the rejections arerelevant because they establish its goodfaith belief that the asserted claims areinvalid, thereby negating the requisite in-tent for inducement. Appellant’s Br. 50(citing Commil USA, LLC v. Cisco Sys.,Inc., 720 F.3d 1361, 1368–69 (Fed.Cir.2013)). As an initial matter, we note thatthis court’s precedent has often warned ofthe limited value of actions by the PTOwhen used for such purposes. See, e.g.,Hoechst Celanese Corp. v. BP Chems. Ltd.,78 F.3d 1575, 1584 (Fed.Cir.1996) (‘‘[G]rantby the examiner of a request for reexami-nation is not probative of unpatentabili-ty.’’); Acoustical Design, Inc. v. ControlElecs. Co., 932 F.2d 939, 942 (Fed.Cir.1991) (‘‘[I]nitial rejection by the [PTO] TTT

hardly justifies a good faith belief in theinvalidity of the claims.’’). However, inthis case we need not decide whether ouropinion in Commil justifies reliance on re-examination evidence to establish a goodfaith belief of invalidity. Instead, we con-clude that, regardless of the evidence’srelevance to a fact at issue at trial, thedistrict court would still not have abusedits discretion in finding that the probativevalue was substantially outweighed by the


risk of unfair prejudice to the patentee,confusion with invalidity (on the merits), ormisleading the jury, thereby justifying ex-clusion under Federal Rule of Evidence403. See, e.g., SynQor, 709 F.3d at 1380(finding no abuse of discretion for exclud-ing non-final reexamination evidence as be-ing ‘‘confusing and more prejudicial thanprobative’’); Callaway Golf Co. v. Acush-net Co., 576 F.3d 1331, 1342–43 (Fed.Cir.2009) (finding the probative value of acopending reexamination marginal and theeffect likely to be highly prejudicial).Thus, we affirm the district court’s exclu-sion of this evidence.

V. Damages

[23] At trial, VirnetX’s damages ex-pert, Mr. Roy Weinstein, provided threereasonable royalty theories, which the dis-trict court admitted over Apple’s chal-lenges under Daubert v. Merrell DowPharmaceuticals, Inc., 509 U.S. 579, 113S.Ct. 2786, 125 L.Ed.2d 469 (1993).

Weinstein’s first approach began withthe lowest sale price of each model of theaccused iOS devices containing the accusedfeatures. J.A. 1616–23. Weinstein thenapplied a 1% royalty rate to the base,derived from a VirnetX policy of seekingto license its patents for at least 1–2% ofthe entire value of products sold and sev-eral allegedly comparable licenses. J.A.1595, 1613–14. This theory yielded a $708million demand, consisting of $566 millionfor products including both FaceTime andVPN On Demand, and $142 million forthose including only VPN On Demand.J.A. 1622–24, 1644.

Weinstein also offered a second damagestheory, regarding FaceTime alone, relyingon a mathematical theorem proved byJohn Nash, a mathematician who proved anumber of results in game theory thathave become important in economics andother fields. J.A. 1628–29. Nash was a

co-winner of the 1994 Nobel Prize in Eco-nomics for some of this work, though notthe theorem at issue here—published as‘‘The Bargaining Problem’’ in 18 Econome-trica 155–62 (Apr.1950). Like other math-ematical theorems, this theorem states anumber of premises and establishes a con-clusion that follows from those premises.In particular, under the conditions statedin the premises, where two persons bar-gain over a matter, there is a ‘‘solution’’ tothe negotiation problem satisfying statedconditions on a desirable result (bargain).That solution—in which ‘‘each bargainerget[s] the same money profit,’’ id. at 162—has come to be called the Nash BargainingSolution.

Weinstein, invoking the Nash Bargain-ing Solution, testified that ‘‘the parties[would have] split between themselves theincremental or additional profits that areassociated with the use of the patentedtechnology.’’ J.A. 1630. Weinstein de-rived the profits associated with FaceTimefrom the revenue generated by the addi-tion of a ‘‘front-facing’’ camera on Apple’smobile devices. Without examining theapplicability to this case of all the precon-ditions for the Nash Bargaining Solution,he invoked the Solution as suggesting a50/50 split of those profits, and then modi-fied that result by 10%, explaining thatVirnetX would have received only 45% ofthe profit because of its weaker bargainingposition, leaving 55% for Apple. J.A. 1633,1709. This calculation amounted to $588million in damages for infringement byFaceTime. J.A. 1633–38.

Finally, Weinstein offered yet anothertheory for FaceTime, again relying on theNash Bargaining Solution. This time, heclaimed that FaceTime ‘‘drove sales’’ forApple iOS products. J.A. 1639. Wein-stein extrapolated from a customer surveyto assert that 18% of all iOS device saleswould not have occurred without the addi-


tion of FaceTime. J.A. 1641. From thatfigure, he determined the amount of Ap-ple’s profits that he believed were attrib-utable to the FaceTime feature, and ap-portioned 45% of the profits to VirnetX,consistent with his previous application ofthe Nash theory. Using this approach,Weinstein arrived at damages of $5.13 perunit, totaling $606 million in damages forFaceTime. J.A. 1643.

Ultimately, the jury awarded VirnetX$368 million in damages. Apple now chal-lenges each of Weinstein’s damages theo-ries, as well as the district court’s juryinstruction on damages. For the reasonsstated below, we vacate the jury’s damagesaward and remand for further proceedingsconsistent with this opinion.

A. Jury Instruction

[24, 25] Upon a finding of infringe-ment, ‘‘the court shall award the claimantdamages adequate to compensate for theinfringement, but in no event less than areasonable royalty for the use made of theinvention by the infringer.’’ 35 U.S.C.§ 284. The most common method for de-termining a reasonable royalty is the hypo-thetical negotiation approach, which ‘‘at-tempts to ascertain the royalty upon whichthe parties would have agreed had theysuccessfully negotiated an agreement justbefore infringement began.’’ LucentTechs., Inc. v. Gateway, Inc., 580 F.3d1301, 1324 (Fed.Cir.2009). A reasonableroyalty may be a lump-sum payment notcalculated on a per unit basis, but it mayalso be, and often is, a running paymentthat varies with the number of infringingunits. In that event, it generally has twoprongs: a royalty base and a royalty rate.

[26] No matter what the form of theroyalty, a patentee must take care to seekonly those damages attributable to the in-fringing features. Indeed, the SupremeCourt long ago observed that a patentee

must in every case give evidence tendingto separate or apportion the defendant’sprofits and the patentee’s damages be-tween the patented feature and the un-patented features, and such evidencemust be reliable and tangible, and notconjectural or speculative; or he mustshow, by equally reliable and satisfacto-ry evidence, that the profits and dam-ages are to be calculated on the wholemachine, for the reason that the entirevalue of the whole machine, as a market-able article, is properly and legally at-tributable to the patented feature.

Garretson v. Clark, 111 U.S. 120, 121, 4S.Ct. 291, 28 L.Ed. 371 (1884).

[27] Thus, when claims are drawn toan individual component of a multi-compo-nent product, it is the exception, not therule, that damages may be based upon thevalue of the multi-component product.LaserDynamics, Inc. v. Quanta Computer,Inc., 694 F.3d 51, 67–68 (Fed.Cir.2012).Indeed, we recently reaffirmed that ‘‘[a]patentee may assess damages based on theentire market value of the accused productonly where the patented feature createsthe basis for customer demand or substan-tially creates the value of the componentparts.’’ Versata Software, Inc. v. SAPAm., Inc., 717 F.3d 1255, 1268 (Fed.Cir.2013) (emphasis added) (quoting SynQor,709 F.3d at 1383). In the absence of sucha showing, principles of apportionment ap-ply.

These strict requirements limiting theentire market value exception ensure thata reasonable royalty ‘‘does not overreachand encompass components not covered bythe patent.’’ LaserDynamics, 694 F.3d at70; see also Garretson, 111 U.S. at 121, 4S.Ct. 291 (‘‘[T]he patentee must show inwhat particulars his improvement has add-ed to the usefulness of the machine orcontrivance.’’). Thus, ‘‘[i]t is not enough to


merely show that the [patented feature] isviewed as valuable, important, or even es-sential to the use of the [overall product].’’LaserDynamics, 694 F.3d at 68. Instead,this court has consistently held that ‘‘areasonable royalty analysis requires acourt to TTT carefully tie proof of damagesto the claimed invention’s footprint in themarket place.’’ ResQNet.com, Inc. v. Lan-sa, Inc., 594 F.3d 860, 869 (Fed.Cir.2010);see also Cornell Univ. v. Hewlett–PackardCo., 609 F.Supp.2d 279, 285 (N.D.N.Y.2009) (‘‘The entire market value rule in-deed permits damages on technology be-yond the scope of the claimed invention,but only upon proof that damages on theunpatented components or technology isnecessary to fully compensate for infringe-ment of the patented invention.’’). Addi-tionally, we have also cautioned againstreliance on the entire market value of theaccused products because it ‘‘cannot helpbut skew the damages horizon for the jury,regardless of the contribution of the pat-ented component to this revenue.’’ UnilocUSA, Inc. v. Microsoft Corp., 632 F.3d1292, 1320 (Fed.Cir.2011).

Apple argues that the district court mis-stated this law on the entire market valuerule in its jury instruction. The districtcourt instructed the jury as follows:

In determining a royalty base, youshould not use the value of the entireapparatus or product unless either: (1)the patented feature creates the basisfor the customers’ demand for the prod-uct, or the patented feature substantiallycreates the value of the other componentparts of the product; or (2) the productin question constitutes the smallest sale-able unit containing the patented fea-ture.

J.A. 2515–16. Apple argues that this in-struction inappropriately created a secondexception that would allow a patentee torely on the entire market value of a multi-

component product so long as that productis the smallest salable unit containing thepatented feature.

We agree with Apple that the districtcourt’s instruction misstates our law. Tobe sure, we have previously permitted pat-entees to base royalties on the ‘‘smallestsalable patent-practicing unit.’’ LaserDy-namics, 694 F.3d at 67. However, theinstruction mistakenly suggests that whenthe smallest salable unit is used as theroyalty base, there is necessarily no fur-ther constraint on the selection of thebase. That is wrong. For one thing, thefundamental concern about skewing thedamages horizon—of using a base thatmisleadingly suggests an inappropriaterange—does not disappear simply becausethe smallest salable unit is used.

Moreover, the smallest salable unit ap-proach was intended to produce a royaltybase much more closely tied to theclaimed invention than the entire marketvalue of the accused products. Indeed,that language first arose in the Cornellcase, where the district court noted that,rather than pursuing a ‘‘royalty base claimencompassing a product with significantnon-infringing components,’’ the patenteeshould have based its damages on ‘‘thesmallest salable infringing unit with closerelation to the claimed invention.’’ 609F.Supp.2d at 287–88 (emphasis added).In other words, the requirement that apatentee identify damages associated withthe smallest salable patent-practicing unitis simply a step toward meeting the re-quirement of apportionment. Where thesmallest salable unit is, in fact, a multi-component product containing several non-infringing features with no relation to thepatented feature (as VirnetX claims it washere), the patentee must do more to esti-mate what portion of the value of thatproduct is attributable to the patentedtechnology. To hold otherwise would per-


mit the entire market value exception toswallow the rule of apportionment.2

[28] In reaching this conclusion, weare cognizant of the difficulty that paten-tees may face in assigning value to a fea-ture that may not have ever been individu-ally sold. However, we note that we havenever required absolute precision in thistask; on the contrary, it is well-understoodthat this process may involve some degreeof approximation and uncertainty. Seegenerally Unisplay, S.A. v. Am. Elec. SignCo., 69 F.3d 512, 517 (Fed.Cir.1995).

We conclude that the district court’sjury instruction regarding the entire mar-ket value rule was legally erroneous.Moreover, that error cannot be consideredharmless, as VirnetX’s expert relied on theentire value of the iOS devices as the‘‘smallest salable units,’’ without attempt-ing to apportion the value attributable tothe VPN On Demand and FaceTime fea-tures. Thus, it is clear that the jury’sverdict was tainted by the erroneous juryinstruction.

B. Weinstein’s First Approach:Royalty Base

[29] In addition to the erroneous juryinstruction, Apple argues that the testimo-ny of VirnetX’s expert on the proper royal-ty base should have been excluded becauseit relied on the entire market value ofApple’s products without demonstratingthat the patented features drove the de-mand for those products. For similar rea-sons to those stated above, we agree.

[30–32] The admissibility of expert tes-timony is governed by the Federal Rulesof Evidence and the principles laid out inDaubert v. Merrell Dow Pharmaceuticals,Inc., 509 U.S. 579, 113 S.Ct. 2786, 125

L.Ed.2d 469 (1993). The district court’s‘‘gatekeeping obligation’’ applies to alltypes of expert testimony. Kumho TireCo. v. Carmichael, 526 U.S. 137, 147, 119S.Ct. 1167, 143 L.Ed.2d 238 (1999). Whilequestions regarding which facts are mostrelevant for calculating a reasonable royal-ty are properly left to the jury, a criticalprerequisite is that the underlying meth-odology be sound. Here, it was not, andthe district court should have exercised itsgatekeeping authority to ensure that onlytheories comporting with settled principlesof apportionment were allowed to reachthe jury.

Under Weinstein’s first damages theory,he undisputedly based his calculations onthe entire cost of the iOS devices, rangingin value from $199 for the iPod Touch to$649 for the iPhone 4S. Weinstein used thebase price at which each product was sold,excluding only charges for additional mem-ory sold separately. He called this thesmallest salable unit. However, whenasked whether this ‘‘remove[d] featuresthat aren’t accused in this case,’’ Weinsteinanswered as follows:

To the extent that the products thatwe’re talking about here contain addi-tional features, like additional memory,for instance, that Apple was chargingfor, by using the lowest saleable unit,I’m doing as much as I can to removepayments for those featuresTTTT

J.A. 1620 (emphasis added). This testimo-ny confirms that Weinstein did not evenattempt to subtract any other unpatentedelements from the base, which thereforeincluded various features indisputably notclaimed by VirnetX, e.g., touchscreen,camera, processor, speaker, and micro-phone, to name but a few. J.A. 1143–44.

2. As, indeed, it did in this case, where VirnetXeffectively relied on the entire market value ofthe iOS devices without showing that the pat-

ented features drove demand for those de-vices, simply by asserting that they were thesmallest salable units.


VirnetX defends Weinstein’s approachby insisting that ‘‘software creates thelargest share of the product’s value’’ forthese popular iOS products. Appellee’sBr. 60. But this misses the point. Wheth-er ‘‘viewed as valuable, important, or evenessential,’’ the patented feature must beseparated. LaserDynamics, 694 F.3d at68. Weinstein made no attempt to sepa-rate software from hardware, much less toseparate the FaceTime software from oth-er valuable software components.

[33] Indeed, the record supports Ap-ple’s contention that Weinstein could haveapportioned a smaller per unit figure forFaceTime; namely, for the use of Face-Time on Mac computers he used a royaltybase of $29—the cost of the software up-grade. J.A. 1619. And he used an evenlower estimate to represent the patentablecontributions to iOS devices in his applica-tion of the Nash Bargaining Solution, cal-culating incremental revenues due to Face-Time at $15 per iOS device. J.A. 1634–36.The only reason Weinstein gave for notusing the $29 as the base for other iOSproducts was that Apple does not actuallycharge separately for FaceTime on thosedevices. J.A. 1673–74. But, as explainedabove, a patentee’s obligation to apportiondamages only to the patented featuresdoes not end with the identification of thesmallest salable unit if that unit still con-tains significant unpatented features.3

[34] Thus, VirnetX cannot simply hidebehind Apple’s sales model to avoid thetask of apportionment. This court rejectsthe excuse that ‘‘practical and economicnecessity compelled [the patentee] to baseits royalty on the price of an entire [de-vice].’’ LaserDynamics, 694 F.3d at 69.There is no ‘‘necessity-based exception tothe entire market value rule.’’ Id. at 70.

On the contrary, a patentee must be rea-sonable (though may be approximate)when seeking to identify a patent-practic-ing unit, tangible or intangible, with aclose relation to the patented feature.

[35] In the end, VirnetX should haveidentified a patent-practicing feature witha sufficiently close relation to the claimedfunctionality. The law requires patenteesto apportion the royalty down to a reason-able estimate of the value of its claimedtechnology, or else establish that its pat-ented technology drove demand for theentire product. VirnetX did neither. Aswe noted in LaserDynamics:

Whether called ‘‘product value appor-tionment’’ or anything else, the fact re-mains that the royalty was expresslycalculated as a percentage of the entiremarket value of a [multi-componentproduct] rather than a patent-practicing[component] alone. This, by definition,is an application of the entire marketvalue rule.

Id. at 68. In calculating the royalty base,Weinstein did not even try to link demandfor the accused device to the patentedfeature, and failed to apportion value be-tween the patented features and the vastnumber of non-patented features containedin the accused products. Because Wein-stein did not ‘‘carefully tie proof of dam-ages to the claimed invention’s footprint inthe market place,’’ Uniloc, 632 F.3d at1317 (quoting ResQNet, 594 F.3d at 869),his testimony on the royalty base underthis approach was inadmissible and shouldhave been excluded.

C. Weinstein’s First Approach:Royalty Rate

[36] In addition to challenging Wein-stein’s testimony with respect to the royal-

3. Because Apple has not challenged it, weoffer no opinion on whether the $29 softwareupgrade is itself so closely related to the pat-

ented feature that VirnetX may rely on itsentire value in determining the proper royaltybase for the FaceTime feature.


ty base, Apple argues that his testimonywith respect to the royalty rate should alsohave been excluded.

After determining the royalty base,Weinstein applied a 1% royalty rate, basedon six allegedly comparable licenses, aswell as his understanding that VirnetX hada ‘‘policy’’ of licensing its patents for 1–2%.Apple argues that the licenses on whichWeinstein relied were not sufficiently com-parable to the license that would haveresulted from the hypothetical negotiation.In particular, Apple points out that two ofthe licenses predated the patents-in-suit.Both of those agreements related to tech-nology leading to the claimed invention,and one contained a software license inaddition to a license for various patentapplications. Apple further complains thatthree of the other licenses were enteredinto in 2012, a full three years after thedate of the ‘‘hypothetical negotiation,’’ setin June 2009. Apple argues that at thetime those licenses were entered into, Vir-netX was in a much better financial posi-tion (and therefore a better bargainingposition) than it was in 2009. Finally,Apple notes that the sixth license coveredsixty-eight VirnetX patents, and wastherefore much broader than the license tofour patents Apple would be seeking in thehypothetical negotiation. It also equatedto a 0.24% royalty rate, which is signifi-cantly lower than the 1–2% rate Weinsteintestified VirnetX would accept.

[37] We have held that in attemptingto establish a reasonable royalty, the ‘‘li-censes relied on by the patentee in provingdamages [must be] sufficiently comparableto the hypothetical license at issue in suit.’’Lucent, 580 F.3d at 1325. ‘‘When relyingon licenses to prove a reasonable royalty,alleging a loose or vague comparabilitybetween different technologies or licensesdoes not suffice.’’ LaserDynamics, 694F.3d at 79. However, we have never re-

quired identity of circumstances; on thecontrary, we have long acknowledged that‘‘any reasonable royalty analysis ‘necessar-ily involves an element of approximationand uncertainty.’ ’’ Lucent, 580 F.3d at1325 (quoting Unisplay, 69 F.3d at 517).Thus, we have cautioned that ‘‘districtcourts performing reasonable royalty cal-culations [must] exercise vigilance whenconsidering past licenses to technologiesother than the patent in suit,’’ ResQNet,594 F.3d at 869, and ‘‘must account fordifferences in the technologies and eco-nomic circumstances of the contractingparties,’’ Finjan, Inc. v. Secure ComputingCorp., 626 F.3d 1197, 1211 (Fed.Cir.2010).

With those principles in mind, we con-clude that the district court here did notabuse its discretion in permitting Wein-stein to rely on the six challenged licenses.To begin with, four of those licenses didindeed relate to the actual patents-in-suit,while the others were drawn to relatedtechnology. Moreover, all of the otherdifferences that Apple complains of werepresented to the jury, allowing the jury tofully evaluate the relevance of the licenses.See J.A. 1600, 1650, 1678–82. No more isrequired in these circumstances.

Our case law does not compel a contraryresult. In ResQNet, we faulted the dis-trict court for relying on licenses with ‘‘norelationship to the claimed invention,’’ noreven a ‘‘discernible link to the claimedtechnology.’’ 594 F.3d at 870. And inLucent, we rejected reliance on licensesfrom ‘‘vastly different situation[s]’’ orwhere the subject matter of certain agree-ments was not even ascertainable from theevidence presented at trial. 580 F.3d at1327–28. The licenses in this case—though not immune from challenge—beara closer relationship to the hypotheticalnegotiation that would have occurred.

This case is therefore much more akin tothe circumstances in Finjan and ActiveVi-


deo Networks, Inc. v. Verizon Communi-cations, Inc., 694 F.3d 1312 (Fed.Cir.2012).In Finjan, there were several differencesbetween the single license relied upon andthe hypothetical negotiation, most notablythat Finjan did not compete with the licen-see as it did with the defendant in thecase, and that the license involved a lumpsum rather than a running royalty. 626F.3d at 1212. Nevertheless, we affirmedthe damages award based on that licensebecause ‘‘[those] differences permitted thejury to properly discount the TTT license.’’Id. And in ActiveVideo, the damages ex-pert relied on two agreements, one ofwhich post-dated the hypothetical negotia-tions by two years, did not involve thepatents-in-suit, and did not cover the tech-nologies in the case, while the other agree-ment covered both patents and softwareservices. 694 F.3d at 1333. Nevertheless,we concluded that the ‘‘degree of compara-bility’’ of the license agreements was ‘‘[a]factual issue[ ] best addressed by crossexamination and not by exclusion.’’ Id.Similarly, here, though there were un-doubtedly differences between the licensesat issue and the circumstances of the hypo-thetical negotiation, ‘‘[t]he jury was enti-tled to hear the expert testimony and de-cide for itself what to accept or reject.’’ i4iLtd. P’ship v. Microsoft Corp., 598 F.3d831, 856 (Fed.Cir.2010), aff’d ––– U.S.––––, 131 S.Ct. 2238, 180 L.Ed.2d 131(2011).

Thus, we do not believe the districtcourt abused its discretion by permittingWeinstein’s testimony regarding the prop-er royalty rate based on these allegedlycomparable licenses.

D. Weinstein’s Second andThird Approaches: Nash

Bargaining Solution

[38] Weinstein also offered two otherestimates of the damages attributable to

the FaceTime feature. Both of these esti-mates relied on the Nash Bargaining Solu-tion. Weinstein began by determining ‘‘in-cremental or additional profits that areassociated with the use of the patentedtechnology.’’ J.A. 1630. Weinstein usedtwo different methods to estimate the in-cremental profits associated with the Face-Time feature. First, he used the front-facing camera as a proxy for the FaceTimefeature, and calculated the profits that hebelieved were attributable to the additionof the front-facing camera to certain Appleproducts. And second, he relied on cus-tomer surveys to assert that 18% of iOSdevice sales would not have occurred butfor the inclusion of FaceTime, and deter-mined the profits attributable to thosesales.

Having thus purported to determinethose profits, Weinstein then testifiedabout how the parties would split thoseincremental profits. To do this, he beganwith the assumption that each party wouldtake 50% of the incremental profits, invok-ing the Nash Bargaining Solution, andthen adjusted that split based on ‘‘the rela-tive bargaining power of the two entities.’’J.A. 1632.

Apple challenges both steps of Wein-stein’s analysis. First, Apple insists thatWeinstein did not adequately isolate theincremental profits attributable to the pat-ented technology under either approach.And second, Apple argues that the invoca-tion of a 50/50 starting point based on theNash Bargaining Solution is akin to the‘‘25 percent rule of thumb’’ that we reject-ed in Uniloc as being insufficientlygrounded in the specific facts of the case.Because we agree with Apple on the sec-ond point, we need not reach the first.

In recent years, numerous districtcourts have confronted experts’ invocationsof the Nash Bargaining Solution as a mod-el for reasonable royalty damages, with


varying results. Compare Robocast, Inc.v. Microsoft Corp., No. 10–1055, 2014 WL350062 (D.Del. Jan. 29, 2014) (excludingexpert testimony based on Nash Bargain-ing Solution because it was not sufficientlytied to the facts of the case); DynetixDesign Solutions, Inc. v. Synopsys, Inc.,No. 11–5973, 2013 WL 4538210, at *4–5(N.D.Cal. Aug. 22, 2013) (excluding experttestimony on royalty rate that began froma starting point of a 50/50 split because theexpert’s methodology was ‘‘indistinguish-able from 25% rule’’); Oracle Am., Inc. v.Google Inc., 798 F.Supp.2d 1111, 1119–21(N.D.Cal.2011) (excluding testimony basedon Nash Bargaining Solution because it‘‘would invite a miscarriage of justice byclothing a fifty-percent assumption in animpenetrable facade of mathematics’’) withMformation Techs., Inc. v. Research inMotion Ltd., No. 08–4990, 2012 WL1142537, at *3 n. 19 (N.D.Cal. Mar. 29,2012) (declining to exclude Weinstein’s tes-timony based on Nash Bargaining Solutionbecause he used it only ‘‘as a check’’ inaddition to the Georgia–Pacific analysis,rather than in lieu of it); Gen–Probe Inc.v. Becton Dickinson & Co., No. 09–2319,2012 WL 9335913, at *3 (S.D.Cal. Nov. 26,2012) (permitting testimony based on NashBargaining Solution because calculationwas sufficiently tied to the facts of thecase, ‘‘including the competitive environ-ment and Gen–Probe’s policy of exploitingits own patents’’); Sanofi–AventisDeutschland Gmbh v. Glenmark Pharms.Inc., USA, No. 07–5855, 2011 WL 383861,at *12–13 (D.N.J. Feb. 3, 2011) (determin-ing that expert’s testimony asserting a50/50 profit split was based on the specificfacts of the case); Amakua Dev. LLC v.Warner, No. 05–3082, 2007 WL 2028186, at*20 (N.D.Ill. July 10, 2007) (permittingreliance on Nash because the ‘‘[d]efen-dants ha[d] not challenged the reliability ofNash’s theories, and the assessment ofwhether the theory persuasively can be

applied in the context of this case is for thejury’’).

For the reasons that follow, we agreewith the courts that have rejected invoca-tions of the Nash theorem without suffi-ciently establishing that the premises ofthe theorem actually apply to the facts ofthe case at hand. The use here was justsuch an inappropriate ‘‘rule of thumb.’’

Previously, damages experts often reliedon the ‘‘25 percent rule of thumb’’ in deter-mining a reasonable royalty rate in ahypothetical negotiation. That rule hy-pothesized that 25% of the value of the in-fringing product would remain with thepatentee, while the remaining 75% wouldgo to the licensee. In Uniloc, however, weheld the ‘‘25 percent rule of thumb’’ to beinadmissible ‘‘because it fails to tie a rea-sonable royalty base to the facts of thecase at issue.’’ 632 F.3d at 1315. In sodoing, we noted that the rule did not dif-ferentiate between different industries,technologies, or parties. Id. at 1317.Rather, it assumed the same 25/75 royaltysplit regardless of the size of the patentportfolio in question or the value of thepatented technology. Id. The problemwas that the 25% rule made too crude ageneralization about a vastly more compli-cated world.

The problem with Weinstein’s use of theNash Bargaining Solution, though some-what different, is related, and just as fatalto the soundness of the testimony. TheNash theorem arrives at a result that fol-lows from a certain set of premises. Ititself asserts nothing about what situationsin the real world fit those premises. Any-one seeking to invoke the theorem as ap-plicable to a particular situation must es-tablish that fit, because the 50/50 profit-split result is proven by the theorem onlyon those premises. Weinstein did not doso. This was an essential failing in invok-ing the Solution. Moreover, we do not


believe that the reliability of this method-ology is saved by Weinstein’s attempts toaccount for the unique facts of the case indeviating from the 50/50 starting point.As we noted in Uniloc:

It is of no moment that the 25 percentrule of thumb is offered merely as astarting point to which the Georgia–Pa-cific [Corp. v. U.S. Plywood Corp., 318F.Supp. 1116 (S.D.N.Y.1970) ] factorsare then applied to bring the rate up ordown. Beginning from a fundamentallyflawed premise and adjusting it based onlegitimate considerations specific to thefacts of the case nevertheless results ina fundamentally flawed conclusion.

632 F.3d at 1317. Indeed, Weinstein’s thinattempts to explain his 10% deviation fromthe 50/50 baseline in this case demon-strates how this methodology is subject toabuse. His only testimony on the matterwas that although he ‘‘considered othersplits,’’ he ultimately determined that a10% deviation—resulting in a 45/55 split—was appropriate ‘‘to reflect the fact thatApple would have additional bargainingpower over VirnetX back in TTT 2009.’’JA. 1708–09. Such conclusory assertionscannot form the basis of a jury’s verdict.See Gen. Elec. Co. v. Joiner, 522 U.S. 136,146, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997)(noting that where an expert considersrelevant material but fails to provide anopinion explaining how that material leadsto his conclusion, ‘‘[a] court may concludethat there is simply too great an analyticalgap between the data and the opinion prof-fered’’).

[39] More importantly, even if an ex-pert could identify all of the factors thatwould cause negotiating parties to deviatefrom the 50/50 baseline in a particularcase, the use of this methodology wouldnevertheless run the significant risk of in-appropriately skewing the jury’s verdict.This same concern underlies our rule that

a patentee may not balance out an unrea-sonably high royalty base simply by as-serting a low enough royalty rate. SeeUniloc, 632 F.3d at 1320. Although theresult of that equation would be mathe-matically sound if properly applied by thejury, there is concern that the high royaltybase would cause the jury to deviate up-ward from the proper outcome. Id. Thus,in Uniloc, we noted that ‘‘[t]he disclosurethat a company has made $19 billion dol-lars in revenue from an infringing productcannot help but skew the damages horizonfor the jury, regardless of the contributionof the patented component to this reve-nue.’’ Id. Similarly, here, the use of a50/50 starting point—itself unjustified byevidence about the particular facts—pro-vides a baseline from which juries mighthesitate to stray, even if the evidence sup-ported a radically different split.

Even the 25% rule had its share ofsupport in the literature, which had ob-served that, at least as an anecdotal mat-ter, a 25% royalty rate was a commonstarting point—and not far off from a com-mon end point—of licensing negotiationsacross numerous industries. See Uniloc,632 F.3d at 1313 (citing Robert Gold-scheider, John Jarosz and Carla Mulhern,Use of the 25 Per Cent Rule in ValuingIP, 37 les Nouvelles 123, 132–33 (Dec.2002); Stephen A. Degnan & Corwin Hor-ton, A Survey of Licensed Royalties, 32 lesNouvelles 91, 95 (June 1997)). Neverthe-less, we rejected it, insisting on testimonytied to the particular facts. The sameinsistence is vital here.

We note that the Nash Bargaining Solu-tion does offer at least one noticeable im-provement over the 25% rule: where the25% rule was applied to the entire profitsassociated with the allegedly infringingproduct, the Nash theory focuses only onthe incremental profits earned by the in-fringer from the use of the asserted pat-


ents. But while we commend parties forusing a theory that more appropriately(and narrowly) defines the universe ofprofits to be split, the suggestion thatthose profits be split on a 50/50 basis—even when adjusted to account for certainindividual circumstances—is insufficientlytied to the facts of the case, and cannot besupported.

For each of the reasons stated above, wevacate the damages award and remand forfurther proceedings consistent with thisopinion.

AFFIRMED–IN–PART, REVERSED–IN–PART, VACATED–IN–PART andREMANDED

COSTS

Each party shall bear its own costs.

,

G. David JANG, M.D., Plaintiff–Respondent,

v.

BOSTON SCIENTIFIC CORPORA-TION and Scimed Life Systems,

Inc., Defendants–Petitioners.

No. 2014–134.

United States Court of Appeals,Federal Circuit.

Sept. 16, 2014.

Background: Inventor brought actionagainst manufacturers, as the assignees ofa patent, alleging breach of a contract thatgoverned the assignment of the patents.The United States District Court for theCentral District of California, VirginiaAnne Phillips, J., entered consent judg-

ment. The inventor appealed. The Court ofAppeals, Dyk, Circuit Judge, 532 F.3d1330, vacated and remanded. The DistrictCourt, Phillips, J., entered revised consentjudgment and revised stipulated summaryjudgment order. The inventor appealed.The Court of Appeals, Linn, Circuit Judge,493 Fed.Appx. 70, reversed in part, vacat-ed in part, and remanded. On remand, theDistrict Court, Phillips, J., denied the man-ufacturers’ motion for summary judgment,and certified questions for interlocutoryappeal.

Holdings: The Court of Appeals, Linn,Circuit Judge, held that:

(1) the Court had subject matter jurisdic-tion over the appeal, and

(2) the Court would decline to exercisejurisdiction over the interlocutory ap-peal.

Petition denied.

1. Federal Courts O3906A state-law breach of contract action

concerning royalties owed for the assign-ment of patents involved a patent issuethat was necessarily raised, actually dis-puted, substantial, and capable of resolu-tion in federal court without disrupting thefederal-state balance approved by Con-gress, as required for the Court of Appealsfor the Federal Circuit to exercise exclu-sive jurisdiction over an interlocutory ap-peal from the Central District of Califor-nia, even though the patents had beenfound to be invalid by the United StatesPatent and Trademark Office (PTO) on exparte reexamination, where the reexamina-tion occurred after the breach of contractclaim was filed. 28 U.S.C.A. §§ 1292(b),1295(a)(1), 1338(a).

2. Federal Courts O2076Subject matter jurisdiction in diversi-

ty cases is determined on the facts as theyexisted at the time the claim was filed.

1308 767 federal reporter, 3d series - berkeley law · 1308 767 federal reporter, 3d series...

Documents