UNITED STATES PATENT AND TRADEMARK OFFICE
`
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
` ____________
`
`SEGA OF AMERICA, INC., UBISOFT, INC., KOFAX, INC.,
`CAMBIUM LEARNING GROUP, INC. AND PERFECT WORLD
`ENTERTAINMENT, INC.
`Petitioners
`
`v.
`
`UNILOC USA, INC. and UNILOC LUXEMBOURG S.A.,
`Patent Owner
`
`____________
`
`Case No. IPR2014-014531
`Patent 5,490,216
` ____________
`
`
`
`CONSOLIDATED FILING:
`
`PETITIONERS’ REPLY IN SUPPORT OF PETITION FOR INTER
`
`PARTES REVIEW OF U.S. PATENT NO. 5,490,216
`
`
`
`
`1 Case IPR2015-01026 has been joined with this proceeding.
`
`
`
`

`
`
`
`
`
`I. INTRODUCTION ............................................................................................ 1
`II. CLAIM CONSTRUCTION ........................................................................... 1
`III. THE ‘216 PATENT IS NOT ENTITLED TO THE EARLIER
`PRIORITY DATE OF THE AUSTRALIAN PROVISIONALS ....................... 2
`A. THE AUSTRALIAN PROVISIONALS DO NOT DISCLOSE A “LICENSEE UNIQUE ID
`GENERATING MEANS” AS CONSTRUED BY THE DISTRICT COURT ........................... 3
`B. THE AUSTRALIAN PROVISIONALS DO NOT DISCLOSE A “MODE SWITCHING
`MEANS” AS CONSTRUED BY THE DISTRICT COURT ................................................. 7
`IV. SCHULL ANTICIPATES CLAIMS 1-11 AND 17-20 ................................ 9
`A. CONCATENATION OF SCHULL IS A “SUMMATION ALGORITHM” ....................... 9
`B. CHECKSUM OF SCHULL IS A “SUMMATION ALGORITHM” ............................... 10
`V. LOGAN ANTICIPATES CLAIMS 12-14 ................................................... 14
`A. LOGAN TEACHES “GENERATING A SECURITY KEY FROM INFORMATION …
`WHICH UNIQUELY IDENTIFIES AN INTENDED REGISTERED USER” ........................ 14
`1. The Combination of a Software Serial Number and Hardware Serial
`Number Together Are Uniquely Associated with a Person Who Intends to
`Become a Licensee .......................................................................................... 14
`2. A Serial Number is Uniquely Associated with a Person Who Intends to
`Become a Licensee .......................................................................................... 14
`B. LOGAN TEACHES THE FINAL ELEMENT OF CLAIM 12 ..................................... 17
`VI. LOGAN AND GRUNDY RENDERS CLAIM 15 AND 16 OBVIOUS ... 19
`VII. HAINES AND MANDULEY RENDERS CLAIMS 12-14 OBVIOUS .. 21
`A. HAINES AND MANDULEY ARE ANALOGOUS ART ........................................... 21
`B. HAINES DISCLOSES A “REGISTRATION SYSTEM” ........................................... 22
`C. HAINES AND MANDULEY TEACH ALL LIMITATIONS OF CLAIMS 12-14 .......... 24
`D. HAINES AND MANDULEY RENDER CLAIMS 13 AND 14 OBVIOUS ................... 24
`VIII. CONCLUSION ......................................................................................... 25
`
`
`
`
`

`
`I.
`
`INTRODUCTION
`
`For nearly 10 years, in litigation and before the PTO, PO has argued that the
`
`claims of the now-expired ‘216 patent should be broadly interpreted. Now,
`
`confronted with new art, arguments, and challenges that render the claims of the
`
`‘216 patent invalid, PO advances the mirror opposite in a desperate attempt to re-
`
`write history. Not only is this legally improper under the principles of judicial
`
`estoppel2, but PO’s new-found arguments are technically and legally incorrect.
`
`Viewed through the proper lens, the law, evidence, and teachings of the references
`
`establish that the claims of the ‘216 Patent are invalid.
`
`II. CLAIM CONSTRUCTION
`
`The District Court and Federal Circuit have already construed the key claim
`
`terms of the ‘216 Patent, on which Petitioners rely. PO now urges this Panel to
`
`narrow the constructions of those same claim terms, despite its previous successful
`
`litigation argument against “improperly narrow[ing]” them. See e.g., Ex. 1010 at 18.
`
`This is particularly true with respect to the terms “security key” and “licensee
`
`unique ID,” which have been construed to mean “a unique identifier associated with
`
`
`2 See, e.g., Kilopass Tech. Inc. v. Sidense Corp., 738 F.3d 1302, 1307-08 (Fed. Cir.
`
`2013) (patent owner admonished for “gamesmanship” for making claim
`
`construction arguments to distinguish over a key piece of prior art in an inter partes
`
`reexamination that were directly contrary to those made in litigation).
`
`
`
`1
`
`

`
`a licensee.” Ex. 1008, CC Order at 9-21. The Federal Circuit held that “vendor-
`
`provided information, like Microsoft's Product Key, could be the basis for a
`
`‘licensee unique ID.’” Ex. 1009, Fed. Circ. 2008 Dec. at 11 (emphasis added).
`
`Throughout that litigation, PO agreed, arguing that the LUID can be generated from
`
`“the unique number on each copy of the software product being protected,” and that
`
`“[t]he Uniloc system simply does not care whether the information is vendor
`
`information, machine information, personal information, or any other type of
`
`information.” Ex. 1036, PO Opening CC Br. at 1; Ex. 1035, PO Reply CC Brief at
`
`12; see also Ex. 1037, PO Op. Br. 2nd Appeal at 10.
`
`Now PO submits for the first time that “vendor information alone” or a
`
`“product number alone” cannot be the basis of a security key. PO Resp. at 9-11. PO
`
`should be judicially estopped from advancing this opposite position. But most
`
`importantly, the key term “security key” should be construed and interpreted
`
`consistently with the district court and Federal Circuit orders.
`
`III. THE ‘216 PATENT IS NOT ENTITLED TO THE EARLIER
`PRIORITY DATE OF THE AUSTRALIAN PROVISIONALS
`
`The filing date of the ‘216 Patent – September 21, 1993 – is the applicable
`
`priority date for the ‘216 Patent claims, as the Board previously determined in its
`
`Institution Decision. Nevertheless, PO argues that the claims are entitled to claim
`
`priority to the earlier-filed Australian Provisionals by ignoring Lockwood, cited by
`
`the Board in its Institution Decision, which states in relevant part:
`
`
`
`2
`
`

`
`The question is not whether a claimed invention is an obvious variant
`of that which is disclosed in the specification. Rather, a prior
`application itself must describe an invention, and do so in sufficient
`detail that one skilled in the art can clearly conclude that the inventor
`invented the claimed invention as of the filing date sought.
`Lockwood v. Am. Airlines, Inc., 107 F.3d 1565, 1572 (Fed. Cir. 1997) (emphasis
`
`added). The Australian Provisionals do not “actually or inherently disclose the
`
`claim element[s]” at issue, and the ‘216 Patent does not get earlier priority.
`
`PowerOasis, Inc. v. T-Mobile USA, Inc., 522 F.3d 1299, 1306-07 (Fed. Cir. 2008).
`
`A. The Australian Provisionals Do Not Disclose a “Licensee Unique
`ID Generating Means” as Construed by the District Court
`
`The structure of the “licensee unique ID generating means” is “a summation
`
`algorithm or a summer and equivalents thereof.” Ex. 1008 at 25. The Australian
`
`Provisionals do not disclose a “summation algorithm” or “summer.”
`
`It is undisputed that the Australian Provisionals lack the only disclosure in
`
`the ‘216 Patent that has ever been identified by the PO, the district court, or the
`
`Federal Circuit for a “summation algorithm” or “summer”:
`
`Applying these teachings, and having scrutinized the ‘216 Patent in
`detail, the Court concludes that the only algorithm specified in the ‘216
`Patent for generating a licensee unique ID is found in the sixth
`embodiment, which states:
`The algorithm, in this embodiment, combines by addition the
`serial number 50 with the software product name 64 and
`
`
`
`3
`
`

`
`customer information 65 and previous user identification 22 to
`provide registration number 66.
`‘216 Patent, col. 11, ll. 53-56. Similarly, the only hardware component
`disclosed for performing the stated function is a “summer.” Id. at col.
`12, ll. 62-65.
`
`Ex. 1008, CC Order at 27 (emphasis added). PO has previously conceded that this
`
`is the only algorithm identified in the ‘216 Patent. Ex. 1038, D.R.I. JMOL Order at
`
`24-25 (“[t]here is no dispute that the generating means structure is fleshed out only
`
`in the sixth embodiment”)(citing 11:53-63, Fig. 9); Ex. 1010, Fed Cir. Dec. at 15
`
`(noting the algorithm at 11:53-56 and the construction is “undisputed on appeal").
`
`The undisputed lack of disclosure of this “only algorithm” in the Australian
`
`Provisionals should end the inquiry. In addition, PO does not even attempt to
`
`identify a hardware “summer” – the other construed structure for the LUID
`
`generating means – in the Australian Provisionals. Indeed, there is none. Pet. at 16-
`
`17; Ex. 1007, Madisetti Dec. at ¶¶23-25. Thus, PO has failed to show that it had
`
`possession of the claimed invention at the time of the Australian Provisionals,
`
`defeating its priority assertion. Lucent Techs., Inc. v. Gateway, Inc., 543 F.3d 710,
`
`719-20 (Fed. Cir. 2008) (means-plus-function claims not entitled to priority to
`
`earlier foreign application when corresponding structure was not disclosed in the
`
`foreign application); see also Lockwood, 107 F.3d at 1572.
`
`Instead, PO grasps to identify any new disclosure of a purported “summation
`
`
`
`4
`
`

`
`algorithm” in the Australian Provisionals, and comes up empty. The best that PO
`
`can do is point to prose that references “combin[ing] information,” and a figure that
`
`states “user details added to serial number.” PO Resp. at 15-17; Ex. 2008, DiEuliis
`
`Dec. at ¶¶49, 53. But PO fails to mention that both of these disclosures are included
`
`in the ‘216 Patent itself, yet have never been identified as a “summation algorithm”
`
`despite the district court having “scrutinized the ‘216 Patent in detail.” Compare
`
`Ex. 1025, PL4842 at 4, Fig. 2B with Ex. 1001, ‘216 Patent at 4:6-11, Fig. 2B.
`
`This is not surprising, as the vague words “combin[ing] information” or “user
`
`details added to serial number” do not lead one of skill in the art to “clearly
`
`conclude” that the inventor was in possession of a summation algorithm. First,
`
`“user details added to serial number” is not a mathematical operation. As the
`
`Australian Provisionals make clear, the “user details” are not necessarily numbers.
`
`Ex. 1025, PL4842 at Fig. 2B (“user details” can be “name, company, address, state,
`
`[or] contact number”). Moreover, as Dr. Madisetti explained, the Australian
`
`Provisionals fail to describe any steps (in prose, figures, or otherwise) by which
`
`user details and a serial number are combined to generate a registration number. Ex.
`
`2009, Madisetti Tr. at 127:2-4 (“I don’t see how you could add user details to a
`
`number unless you provided a procedure for it.”); 62:19-23 (“It refers to a very
`
`vague description with no detail … as to how this combination takes place because
`
`some of the user data and information is in different forms and formats.”); see also
`
`
`
`5
`
`

`
`id. 79:19-80:2, 125:7-14, 82:2-7, 87:13-18, 88:7-15, 132:24-133:14.
`
`Perhaps recognizing that “user details added to serial number” does not lead
`
`one of skill in the art to “clearly conclude” that the inventor was in possession of a
`
`mathematical “summation algorithm,” PO and its expert next attempt to re-write it.
`
`Specifically, Dr. DiEuliis opines that “registration number generated from user
`
`details added to serial number” in Figure 2B really means “registration number
`
`generated from user details added to serial number,” and then concludes that “the
`
`registration number and the serial number are numerical data (that is, numbers), and
`
`that Fig. 2B’s disclosure of ‘adding’ them means the addition of two numbers.” PO
`
`Resp. at 19; see also id. at 20, 22; Ex. 2008, DiEuliis Dec. at ¶¶51,53,57.
`
`This is not what Figure 2B says. As explained in the Australian Provisional
`
`PL4842: “Preferably said registration number algorithm combines information
`
`entered by a prospective registered user unique to that user with a serial number. . .
`
`.” Ex. 1025 at 3. The “registration number” is the output – not an input – of the
`
`registration number algorithm. See Ex. 1025 at Fig. 2B (“Registration number
`
`generated …” is followed by “User given registration number”); 6 (“registration
`
`number algorithm . . . generates a registration number from the information unique
`
`to the user together with the serial number previously generated”)(emphasis added).
`
`Finally, given the obvious lack of disclosure in the Australian Provisionals,
`
`PO attempts to deflect from the problems with its own priority arguments by
`
`
`
`6
`
`

`
`accusing Dr. Madisetti of “self-contradiction.” PO Resp. at 23-25, Ex. 2008 at ¶60.
`
`But again, this is not true. Neither PO nor Dr. DiEuliis refute Dr. Madisetti’s
`
`opinion that “there were multitudes of ways to ‘combine’ information in 1992” that
`
`do not involve a summation algorithm. PO Resp. at 24-25; Ex. 2008 at ¶60. Dr.
`
`DiEuliis confirms the same. Ex. 2008 at ¶78. And Dr. Madisetti provided many
`
`examples of combining letters and numbers without mathematical addition.3 Ex.
`
`2009, Madisetti Tr. at 121:13-124:1, 128:5-16, 128:25-129:12, 132:11-20, 154:22-
`
`155:5. The Australian Provisionals do not disclose a summation algorithm or a
`
`summer and the ‘216 Patent is not entitled to earlier priority.
`
`B.
`
`The Australian Provisionals Do Not Disclose a “Mode Switching
`Means” as Construed by the District Court
`
`The structure of the “mode switching means” is “program code which
`
`performs a comparison of two numbers or a comparator and equivalents thereof.”
`
`Ex. 1008 at 41-44. PO does not dispute that the Australian Provisionals contain no
`
`disclosure of a hardware “comparator,” and has failed to show that it had possession
`
`
`3 PO appears to take issue with just one of Dr. Madisetti’s examples, but his
`
`testimony that “appending ‘4’ to ‘T,’ ‘O’ to ‘4,’ and so forth, to arrive at
`
`‘T4O0M4’” (i.e., letters and numbers) is not a mathematical “summation
`
`algorithm” is entirely consistent with his opinion that Schull’s disclosure of integer
`
`(i.e., whole number) concatenation is a “summation algorithm.” PO Resp. at 24-25.
`
`
`
`7
`
`

`
`of the claimed invention at the time of the Australian Provisionals. See supra at p. 4
`
`(citing Lucent and Lockwood).
`
`Regarding the “program code which performs a comparison of two
`
`numbers,” PO does not dispute that the ‘216 patent algorithm (Ex. 1001 at 6:12-14)
`
`is not present in the Australian Provisionals. Instead, PO attempts to paint purely
`
`functional language in the Australian Provisionals as “structure”: “the security
`
`routine checks to see whether the entered registration number matches the
`
`calculated registration number” (Ex. 1025, PL4842 at 6-7), “… there is provided a
`
`security routine attachable to software to be protected” (id. at 2), and “Application
`
`uses unlocking algorithm to check validity” (id. at Fig. 2B). PO Resp. at 27-28. This
`
`is no different from the functional claim language itself, and is not sufficient
`
`structure to achieve the specified function. In fact, PO previously argued that the
`
`“modifying recitation in each claim merely provides the functional criteria for the
`
`mode switching means; it does not provide structural limitations for achieving
`
`modes switching.” Ex. 1023, 11/20/2012 OA Resp. at 25-26. PO’s new citations to
`
`the Australian Provisionals similarly fail.
`
`PO also argues that because Schull discloses structure corresponding to a
`
`“mode switching means,” so too do the Australian Provisionals. PO Resp. at 30-32.
`
`Not so. Schull expressly discloses program pseudocode that compares two numbers
`
`to determine whether advanced features should be unlocked. Ex. 1002, Schull at
`
`
`
`8
`
`

`
`7:22-27, 13:3-17; Pet. at 27-28; Ex. 1039, Madisetti Dec. at ¶¶12-13. The
`
`Australian Provisionals do not contain a similar disclosure.
`
`IV. SCHULL ANTICIPATES CLAIMS 1-11 AND 17-20
`
`Because the ‘216 Patent is not entitled to a priority before September 21,
`
`1993, Schull is prior art under 35 USC §102(e). As set forth below and in the
`
`petition, Schull teaches a summation algorithm, summer, or equivalent and
`
`therefore anticipates claims 1-11 and 17-20.
`
`A. Concatenation of Schull is a “Summation Algorithm”
`
`PO does not dispute that the password-generating algorithm (i.e., LUID
`
`generating means) of Schull concatenates a Program ID, Feature-ID, and Target-ID.
`
`Pet. at 21-22; Ex. 1007, Madisetti Dec. at ¶¶40-47. PO also does not dispute that
`
`concatenation of three integers described in Schull, when performed through the
`
`two basic approaches explained by Dr. Madisetti, is a summation algorithm or
`
`equivalent thereof. Pet. at 21-22; Ex. 1007, Madisetti Dec. at ¶¶40-47. Thus, there is
`
`no genuine dispute that Schull satisfies the LUID generating means limitations.
`
`Instead, PO and its expert merely posit that there is hypothetically a third
`
`approach to concatenation that does not involve summation, claiming that
`
`“concatenation is normally accomplished by copying the data to a contiguous
`
`section of memory so that the result is stored as a continuous array.” Ex. 2008,
`
`DiEuliis Dec. at ¶78; see also id. at ¶¶78-90, PO Resp. at 35. But this makes no
`
`
`
`9
`
`

`
`sense in view of the concatenated Passwordable ID disclosed in Schull. Ex. 1039,
`
`Madisetti Dec. at ¶¶7-14. In particular, Schull expressly discloses that the generated
`
`passwordable ID in the Pascal programming language is a “longint,” which is a
`
`single integer or whole number ranging from -2147483648 to +2147483648. Id.;
`
`see also Ex. 1002, Schull at 13:3-10. The result of simply moving bytes into
`
`contiguous sections of memory, under Dr. DiEuliis’s approach to concatenation, is
`
`three separate integers/numbers (e.g., [1234], [56], [789]) that are stored in separate
`
`places in memory, not a single whole number or “longint” integer, as required by
`
`Schull. Ex. 1039, Madisetti Dec. at ¶¶15-20. Thus, one of ordinary skill in the art
`
`could not use Dr. DiEuliis’s approach when implementing the password-generating
`
`algorithm of Schull, much less consider such an approach to be “normal.” Id. at
`
`¶¶15-23. Instead, even under Dr. DiEuliis’ approach, a programmer would have to
`
`additionally actually concatenate the three separately stored numbers using the
`
`summation approaches explained by Dr. Madisetti in order to obtain a single whole
`
`number. Id. at ¶20.
`
`B. Checksum of Schull is a “Summation Algorithm”
`
`In addition to the integer concatenation “summation algorithm,” Schull also
`
`discloses a summation algorithm through the use of a “checksum” in generating a
`
`Passwordable ID. Pet. at 22, 25. PO’s principal response to this additional
`
`disclosure is to suggest that the checksum is not necessarily a permanent part of the
`
`
`
`10
`
`

`
`Passwordable ID. PO Resp. at 38; Ex. 2008, DiEuliis Dec. at ¶96. To the contrary,
`
`Schull discloses “append[ing] two or more digits to the ID which would constitute a
`
`checksum.” Ex. 1002, Schull at 7:28-36 (emphasis added); Ex. 1007, Madisetti Dec.
`
`at ¶48 (noting that appending a checksum to the concatenated number would result
`
`in a Passwordable ID that is N+M+T+2 digits long). “Append” means “to add as a
`
`supplement” or “to fix to; attach,” such that it would become a part of the ID itself.
`
`Ex. 1040, American Heritage Dictionary at 120.
`
`Moreover, “the ID” received by the Licensing Computer is “checked for
`
`internal consistency based upon the methods of Step 25” – which is the same “Step
`
`25” process described at 7:10-42 of Schull that is used by the local computer to
`
`generate “the ID,” i.e., concatenating a Program ID, Feature ID, and Target ID and
`
`appending a two or more digit checksum. Ex. 1002, Schull at 10:35-39, 7:10-42.
`
`Schull also discloses that “[o]ther information could usefully be encrypted into the
`
`ID as well,” clearly indicating that the checksum and possibly “other information”
`
`become a permanent part of the ID. Id. at 7:37-42.
`
`PO also argues, for the first time, that “summation is not inherent in the
`
`generation of checksums.” PO Resp. at 39 (emphasis original). In support of this
`
`position, PO’s expert cites to three publications that have nothing to do with Schull,
`
`but purportedly describe table lookup type methods for performing checksums. Ex.
`
`2008, DiEuliis Dec. at ¶¶97-104. Putting aside the fact that these do not concern
`
`
`
`11
`
`

`
`Schull and are therefore of no relevance, each article confirms that table lookup
`
`methods compute checksums using summation. Ex. 1039, Madisetti Dec. at ¶¶24-
`
`36; see also Ex. 2009, Madisetti Tr. at 144:16-20. Moreover, Dr. DiEuliis admitted
`
`in his deposition that “most checksums” use addition, and that he had never created
`
`a checksum that did not use summation. Ex. 1041, DiEuliis Tr. 63:10-17, 88:8-10.
`
`Even worse, PO has unequivocally argued in at least 3 other proceedings
`
`concerning the ‘216 Patent that a checksum is a summation algorithm – before two
`
`different district courts, the Federal Circuit, and the PTO. First, in the Microsoft
`
`trial, PO relied on expert testimony from Dr. Klausner that “various types of hashes
`
`ranging from simple checksums to very complicated hashes” were all “known
`
`forever to be summaries, to be summations,” even going so far to say that “[t]here is
`
`no hashing algorithm that I know of that fails to do summarizing and addition, and
`
`that’s been true for a long time.” Ex. 1032, Day 2 Trial Tr. at 177:7-21. PO took the
`
`same position in JMOL briefing (Ex. 1042, PO Op. Br. JMOL at 20-21), and the
`
`Federal Circuit determined that the jury’s finding of infringement was supported by
`
`Dr. Klausner’s testimony, as well as a definition submitted by PO that a
`
`“checksum” is “a value that is the summation of a byte stream.” Ex. 1010, Fed. Cir.
`
`2011 Dec. at 17, 24; see also Ex. 1037, PO App. Br. at 30, 41-42.
`
`Second, PO relied on a second expert, William Rosenblatt, in the first
`
`reexamination to argue that one of ordinary skill in the art in the 1992-1994 time
`
`
`
`12
`
`

`
`frame would understand the term “checksum” as being calculated from data that is
`
`“added up (e.g., broken up into C-byte chunks, where C is a small number such as
`
`1, 2, 4, or 8, and summed).” Ex. 1033, Rosenblatt Dec. at ¶52, ¶¶53-55 (citing
`
`1993/94 dictionary definitions of “checksum” as “adding the binary value” and “the
`
`sum of a group of data”); Ex. 1016, 11/29/2010 OA Response at 27. Dr. Rosenblatt
`
`also provided specific examples of checksums, all of which include summation. Id.
`
`at ¶¶57-59; see also e.g., ¶58 (“All checksum algorithms in practical use are derived
`
`from this basic one,” where “all the digits … are added”).
`
`Third, in recent litigation in the Eastern District of Texas, PO successfully
`
`defeated the defendants’ allegation that it disclaimed the use of checksums in
`
`generating an LUID. In doing so, PO again relied on Dr. Rosenblatt’s declaration
`
`that a checksum includes summation. Ex. 1043, PO Opening CC Br. at 18 (“Uniloc
`
`did not, however, disclaim use of checksums or algorithms used to test data
`
`integrity to generate licensee unique IDs.”), 20 (characterizing Dr. Rosenblatt’s
`
`declaration as “a primer of checksums”), 6-7.
`
`PO now relies on the testimony of a third expert, Dr. DiEuliis, for the
`
`opposite position: “Summation is not inherent in the generation of checksums.” Ex.
`
`2008, DiEuliis Dec. ¶97. When questioned on this unique opinion, Dr. DiEuliis did
`
`not disagree with the checksum definition cited by the Federal Circuit. Ex. 1041,
`
`DiEuliis Tr. at 61:21-62:20. He also admitted that his opinion conflicts with PO’s
`
`
`
`13
`
`

`
`previous positions. Id. at 70:8-72:22, 80:10-81:21. PO is bound by its previous
`
`admissions that checksums are summation.
`
`V. LOGAN ANTICIPATES CLAIMS 12-14
`A.
`Logan Teaches “Generating a Security Key from Information …
`Which Uniquely Identifies an Intended Registered User”
`
`The term “information . . . which uniquely identifies an intended registered
`
`user” means “information that is uniquely associated with a person who intends to
`
`become a licensee so as to access full functionality of the digital data.” Ex. 1008,
`
`CC Order at 22-24.
`
`1.
`
`The Combination of a Software Serial Number and Hardware
`Serial Number Together Are Uniquely Associated with a
`Person Who Intends to Become a Licensee
`
`As set forth in the petition and recognized by the Board, “[t]he combination
`
`of the software serial number and the hardware serial number together being input
`
`by the user to generate the first intermediate code also ‘uniquely identifies an
`
`intended registered user’ as recited in claim 12 compared to the software serial
`
`number alone.” Inst. Dec. at 16; see also Pet. at 38-39. PO does not argue to the
`
`contrary and claims 12-14 should be canceled as anticipated by Logan for this
`
`reason alone.
`
`2.
`
`A Serial Number is Uniquely Associated with a Person Who
`Intends to Become a Licensee
`
`PO argues that Logan’s software serial number is not “uniquely associated
`
`with a person” because “multiple users input the same software serial number.” PO
`
`
`
`14
`
`

`
`Resp. at 42. PO’s position is directly contrary to the claim construction. At PO’s
`
`request during the original claim construction process, the Court explicitly rejected
`
`confining “uniquely associated with a person” to “one-of-a-kind information that
`
`describes/identifies a person.” Ex. 1008 at 22, 11-12; see also Ex. 1044, Klausner
`
`Dec. at 10 (‘216 Patent “expressly contemplates information that is not one-of-a-
`
`kind but that is nonetheless unique”); id. at 3-4; Ex. 1035, PO CC Reply Br. at 7-8.
`
`PO also alleges that neither the Federal Circuit nor the District Court
`
`determined, “with respect to claims 12-14, whether a vendor’s product number
`
`suffices as a unique input associated with a user.” PO Resp. at 42-43. This, too, is
`
`not true. The Federal Circuit expressly rejected Microsoft’s argument that a Product
`
`Key did not constitute “information uniquely associated with the user.” Ex. 1045,
`
`Microsoft 1st App. Br. at 44 (arguing that “the Product Key is information provided
`
`by the software vendor” and is not “information uniquely associated with the
`
`user”); Ex. 1009, Fed. Cir. 2008 Dec. at 8 (“We have considered these
`
`[Microsoft’s] arguments and conclude they are without merit.”).4
`
`4 In PO’s own words: “Unique user information can include the product’s serial
`
`number; information about the user (e.g., name, address); or a vendor-provided
`
`identifier contained, for example, on an installation disk.” Ex. 1037, PO 2nd Appeal
`
`Op. Br. at 10. The inventor also testified that the software serial number “identifies
`
`the owner of the software.” Ex. 1046, EA Trial Dec. 2 AM Tr. at 27:1-3.
`
`
`
`15
`
`

`
`Even further, PO has previously taken the exact opposite position, arguing
`
`that a Product ID (e.g., a serial number or Product Key) by itself is uniquely
`
`associated with a person. Ex. 1044, Klausner Dec. at 10 (“the specification
`
`discloses other types of information that can be uniquely descriptive of an intending
`
`licensee such as Product IDs”); Ex. 1016, 11/29/2010 OA Resp. at 39 (“The licensee
`
`unique ID is generated … based on a unique serial number, or product key, that is
`
`assigned to each copy of the software when it is shopped. Each unique identifying
`
`number is associated with the purchase of each copy of the software ”).
`
`Like the Microsoft Product Key, Logan’s software serial number is unique to
`
`each original copy of the software, provided by the vendor, and is generally
`
`imprinted on a label of the software program. Compare Ex. 1003, Logan at 4:19-31
`
`with Ex. 1045, Microsoft 1st App. Br. at 10, 44. The Microsoft Product Key and
`
`Logan’s software serial number also provide “a way to help [the software supplier]
`
`identify its customers.” Compare Ex. 1003, Logan at 6:10-29 with Ex. 1047, PO 1st
`
`App. Reply Br. at 14. Further, both Logan and Microsoft’s product activation
`
`process try to prevent casual copying of software among a small group of people.
`
`Compare Ex. 1003, Logan at 1:27-40 and 1:41-56 with Ex. 1045, Microsoft 1st App.
`
`Br. at 11. Logan fits squarely within the bounds of the construed claim scope. Just
`
`as Microsoft’s Product Key is information uniquely associated with a person, so is
`
`Logan’s vendor-assigned serial number.
`
`
`
`16
`
`

`
`Finally, PO hollowly argues that “Petitioners stretch the imagination”
`
`because “[t]he word instantiation appears nowhere in the District Court claim
`
`construction.” PO Resp. at 42 (emphasis original). But PO has previously approved
`
`the word “instantiation,” endorsing PTO Examiner Heneghan’s characterization of
`
`the claimed invention: “The Reexam Order correctly quoted the Reasons for
`
`Confirmation in the First Reexam, observing that ‘the licensee unique ID . . . must
`
`be derived from at least piece of information that is specific to the user, such as
`
`name, billing information, or product information unique to the instantiation entered
`
`by the user.’” Ex. 1023, 11/20/2012 OA Resp. at 14.
`
`B.
`
`Logan Teaches the Final Element of Claim 12
`
`In the face of Logan’s clear disclosure of claim 12, PO once again resorts to
`
`arguments that are contrary to its previous positions and the District Court’s claim
`
`construction order. Specifically, PO asserts that Logan does not disclose “checking
`
`by the registration authority” because “the software supplier does not check or
`
`verify anything” and Logan “describes actions that take place on the user’s
`
`computer, not actions taken by the software supplier.” PO Resp. at 44-45; Ex. 2008,
`
`DiEuliis Dec. at ¶¶130-131; Ex. 1041, DiEuliis Tr. at 100:17-101:9. This is not a
`
`viable argument. In fact, PO told the District Court that “this claim term is silent
`
`regarding the location of the checking.” Ex. 1035, PO CC Reply Br. at 32. And the
`
`District Court agreed: “Because claim 12 simply states that checking is done by the
`
`
`
`17
`
`

`
`registration authority and not at the registration authority … this Court declines to
`
`include the limitation that the checking is necessarily done on the remote side.” Ex.
`
`1008, CC Order at 53 (underline in original).
`
`When confronted, Dr. DiEuliis admitted that his opinion – and PO’s positions
`
`– are contrary to the claim construction order. Ex. 1041, DiEuliis Tr. at 103:23-
`
`105:13 (“according to this judge’s Claim Construction Order, especially, and given
`
`in light, yeah, I think my statement in my – my paragraph [131] would be incorrect.
`
`It's not consistent with the judge’s order) (Emphasis added).
`
`The same is true regarding PO’s attempt to now import a “temporal aspect of
`
`the checking (i.e., ‘at the time the security key is generated’)” into claim 12. PO
`
`Resp. at 45; Ex. 2008, DiEuliis Dec. at ¶131. Again, PO argued the opposite to the
`
`District Court: “There is no requirement . . . that the remote side check ‘at the point
`
`in time when the security key is generated on the local side.” Ex. 1044, Klausner
`
`Dec. at 10; see also id. at 11. The District Court did not import a temporal
`
`requirement either. Ex 1008, CC Order at 53. Dr. DiEuliis also admits that his
`
`opinion in this regard was wrong (Ex. 1041, DiEuliis Tr. at 103:23-105:13).
`
`When viewed consistently with the ‘216 Patent and the District Court’s claim
`
`construction, Logan discloses the “checking” limitation of the ‘216 Patent. The ‘216
`
`patent discloses that the registration authority generates a security key/licensee
`
`unique ID, communicates the key to the intended licensee and the user’s PC (i.e.,
`
`
`
`18
`
`

`
`the local side, not the registration authority), and subsequently performs a
`
`comparison of the two generated security keys. See, e.g., Ex. 1001 at 7:21-50. The
`
`registration authority is not provided with the results of the comparison or even the
`
`locally generated security key, and could not perform any comparison or
`
`verification itself. See, e.g., id. at 7:19-20, 7:36-50. The ‘216 Patent discloses only
`
`that the user, not the registration authority, performs the check. Ex. 1001 at 7:39-50.
`
`Logan also discloses that the local side, not the remote side, performs the
`
`comparison of the local and remote algorithm outputs. Ex. 1001 at 7:39-50 and Ex.
`
`1003, Logan at 5:68-6:10; see also Pet. at 40. Claim 12 is anticipated by Logan.
`
`VI. LOGAN AND GRUNDY RENDERS CLAIMS 15 AND 16 OBVIOUS
`
`Petitioner’s mapping of Logan in view of Grundy to claim 15 is undisputed –
`
`PO does not take issue with any aspect of it. Instead, PO merely concludes that
`
`“neither Logan nor Grundy d