`
`18-2422
`
`UNITED STATES COURT OF APPEALS FOR THE FEDERAL CIRCUIT
`
`VEVEO, INC.,
`
`Appellant
`
`v.
`
`COMCAST CABLE COMMUNICATIONS, LLC,
`
`Appellee
`
`Appeal from the United States Patent and Trademark Office, Patent Trial
`and Appeal Board in No. IPR2017-00715
`
`
`APPELLANT’S CORRECTED PRINCIPAL BRIEF
`
`Jason D. Eisenberg
`Byron L. Pickard
`Kristina Caggiano Kelly
`Sterne Kessler Goldstein & Fox P.L.L.C.
`1100 New York Avenue, NW
`Washington, DC 20005
`202.371.2600
`
`Counsel for Appellant,
`VEVEO, INC.
`
`Dated: February 1, 2019
`
`
`
`Comcast, Ex. 1234
`Comcast v. Rovi
`IPR2019-00239
`
`
`
`Case: 18-2422 Document: 13 Page: 1 Filed: 10/12/2018
`Case: 18-2422 Document: 26 Page: 2 Filed: 02/01/2019
`
`FORM 9. Certificate of Interest
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
` Form 9
` Rev. 10/17
`UNITED STATES COURT OF APPEALS FOR THE FEDERAL CIRCUIT
`
`Veveo, Inc.
`Comcast Cable Communications
`
`
` v.
`
`
`
`
`
`
`
`
`
`18-2422
`Case No.
`
`
`
`
`AMENDED
`CERTIFICATE OF INTEREST
`
`
`Counsel for the:
`(cid:133) (petitioner) (cid:133) (appellant) (cid:133) (respondent) (cid:133) (appellee) (cid:133) (amicus) (cid:133) (name of party)
`
`
`
`Veveo, Inc.
`
`
`
`
`
`
`
`
`
`
`
`
`certifies the following (use “None” if applicable; use extra sheets if necessary):
`
`
`1. Full Name of Party
`Represented by me
`
`Veveo, Inc.
`
`
`
`
`
`2. Name of Real Party in interest
`(Please only include any real party
`in interest NOT identified in
`Question 3) represented by me is:
`Rovi Guide
`
`
`
`
`
`
`
`
`
`3. Parent corporations and
`publicly held companies
` that own 10% or more of
`stock in the party
`Rovi Corp.
`TiVo Corp.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`4. The names of all law firms and the partners or associates that appeared for the party or amicus now
`represented by me in the trial court or agency or are expected to appear in this court (and who have not
`or will not enter an appearance in this case) are:
`Ropes & Gray LLP - Mark D. Rowland, Gabrielle E. Higgins, Scott A. McKeown, James R. Batchelder,
`Kevin J. Post and Josef B. Schenker.
`
`
`
`Case: 18-2422 Document: 13 Page: 2 Filed: 10/12/2018
`Case: 18-2422 Document: 26 Page: 3 Filed: 02/01/2019
`
`FORM 9. Certificate of Interest
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
` Form 9
` Rev. 10/17
`
`
`5. The title and number of any case known to counsel to be pending in this or any other court or agency
`that will directly affect or be directly affected by this court’s decision in the pending appeal. See Fed. Cir.
`R. 47. 4(a)(5) and 47.5(b). (The parties should attach continuation pages as necessary).
`
`None
`
`
`
`
`
`
`
`
`
`
`10/12/2018
`
`
`
`
` Date
`
`
`
`
`Please Note: All questions must be answered
`
`
`
`
`
`
`
`
`Counsel of Record
`cc:
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`/s/ Jason D. Eisenberg
`
`
`
`
`
`
`Signature of counsel
`Jason D. Eisenberg
`
`
`
`
`
`
`Printed name of counsel
`
`
`
`
`
`Reset Fields
`
`
`
`Case: 18-2422 Document: 26 Page: 4 Filed: 02/01/2019
`
`TABLE OF CONTENTS
`
`STATEMENT OF RELATED CASES ..................................................................... v
`JURISDICTIONAL STATEMENT .......................................................................... 1
`STATEMENT OF THE ISSUE ................................................................................. 2
`STATEMENT OF THE CASE .................................................................................. 3
`I.
`Background of the Technology ....................................................................... 3
`II.
`Background of the Present Dispute ............................................................... 12
`III. The Alleged Prior Art .................................................................................... 15
`A. U.S. Pat. App. Pub. No. 2004/0133564 “Gross” ................................ 15
`B.
`U.S. Pat. No. 6,529,903 “Smith” ......................................................... 19
`SUMMARY OF THE ARGUMENT ...................................................................... 21
`ARGUMENT ........................................................................................................... 26
`I.
`Standard of Review ........................................................................................ 26
`II.
`Relevant Legal Principles .............................................................................. 26
`III. The Challenged Claims are Not Obvious In View of Smith and Gross ........ 29
`A.
`The “associating” step requires directly mapping content items to
`ambiguous prefix strings ..................................................................... 30
`The combination of Smith and Gross does not disclose a method of
`directly mapping content items to ambiguous prefix strings .............. 34
`CONCLUSION AND RELIEF SOUGHT .............................................................. 38
`
`
`B.
`
`
`
`
`
`ii
`
`
`
`Case: 18-2422 Document: 26 Page: 5 Filed: 02/01/2019
`
`TABLE OF AUTHORITIES
`
` Page(s)
`
`Cases
`ActiveVideo Networks, Inc. v. Verizon Commc’ns, Inc.,
`694 F.3d 1312 (Fed. Cir. 2012) .......................................................................... 28
`
`Application of Irmscher, 262 F.2d 85 (CCPA 1958) ............................................... 28
`
`Belden Inc., v. Berk-tek LLC,
`805 F.3d 1064 (Fed. Cir. 2015) .......................................................................... 26
`
`In re Bigio,
`381 F.3d 1320 (Fed.Cir.2004) ............................................................................ 29
`
`In re Clay,
`966 F.2d 656 (Fed. Cir. 1992) ............................................................................ 29
`
`In re Cyclobenzaprine Hydrochloride Extended-Release Capsule
`Patent Litig.,
`676 F.3d 1063 (Fed. Cir. 2012) .......................................................................... 29
`
`DePuy Spine, Inc. v. Medtronic Sofamor Danek, Inc.,
`567 F.3d 1314 (Fed. Cir. 2009) .......................................................................... 28
`
`In re Gordon,
`733 F.2d 900 (Fed. Cir. 1984) ............................................................................ 28
`
`Graham v. John Deere Co.,
`383 U.S. 1 (1966) ................................................................................................ 28
`
`Innogenetics, N.V. v. Abbott Labs.,
`512 F.3d 1363 (Fed. Cir. 2008) .......................................................................... 27
`
`Innovention Toys, LLC, v. MGA Entm’t., Inc.,
`637 F.3d 1314 (Fed.Cir.2011) ............................................................................ 29
`
`Kinetic Concepts, Inc. v. Smith & Nephew, Inc.,
`688 F.3d 1342 (Fed. Cir. 2012) .................................................................... 26, 27
`
`KSR Int’l Co. v. Teleflex Inc.,
`550 U.S. 398 (2007) ............................................................................................ 27
`iii
`
`
`
`Case: 18-2422 Document: 26 Page: 6 Filed: 02/01/2019
`
`Los Angeles Biomedical Research Inst. v. Eli Lilly & Co.,
`849 F.3d 1049 (Fed. Cir. 2017) .......................................................................... 27
`
`Par Pharm., Inc. v. TWi Pharms., Inc.,
`773 F.3d 1186 (Fed. Cir. 2014) .......................................................................... 27
`
`Tandon Corp. v. U.S. Int’l Trade Comm’n,
`831 F.2d 1017 (Fed. Cir. 1987) .................................................................... 32, 33
`
`Trivascular, Inc., v. Samuels,
`812 F.3d 1056 (Fed. Cir. 2016) .......................................................................... 26
`
`Yamanouchi Pharm. Co. v. Danbury Pharmacal, Inc.,
`231 F.3d 1339 (Fed. Cir. 2000) .......................................................................... 29
`
`Statutes
`
`35 U.S. § 103 ............................................................................................................ 29
`
`
`
`
`
`
`
`iv
`
`
`
`Case: 18-2422 Document: 26 Page: 7 Filed: 02/01/2019
`
`STATEMENT OF RELATED CASES
`
`No related cases are currently pending before this Court. The patent at issue
`
`in this appeal is not currently asserted in any copending litigation.
`
`
`
`v
`
`
`
`Case: 18-2422 Document: 26 Page: 8 Filed: 02/01/2019
`
`JURISDICTIONAL STATEMENT
`
`The Court has jurisdiction under 28 U.S.C. § 1295(a)(4)(A) and 35 U.S.C. §
`
`141(c). The United States Patent and Trademark Office’s Patent Trial and Appeal
`
`Board issued the Final Written Decision on appeal pursuant to 35 U.S.C. § 318(a)
`
`and 37 C.F.R. § 42.73 on July 25, 2018. Veveo, Inc., timely appealed that decision
`
`on September 24, 2018.
`
`
`
`
`
`
`
`1
`
`
`
`Case: 18-2422 Document: 26 Page: 9 Filed: 02/01/2019
`
`STATEMENT OF THE ISSUE
`
`Whether claims 1-10, 12-24, and 26-31 are obvious in view of Gross and
`
`Smith, where the combination of Gross and Smith does not disclose “content items
`
`[] directly mapped to the corresponding strings of one or more overloaded keys by
`
`a direct mapping.”
`
`
`
`
`
`2
`
`
`
`Case: 18-2422 Document: 26 Page: 10 Filed: 02/01/2019
`
`STATEMENT OF THE CASE
`
`Veveo, Inc. is a member of the Rovi, Inc. family of companies. Following
`
`the 1953 launch of TV Guide Magazine, the Rovi family of companies (which has
`
`included, through strategic joint ventures, mergers and acquisitions, TV Guide
`
`Onscreen, StarSight Telecast, Prevue, TV Guide, Video Guide, Gemstar, Aptiv
`
`Digital, United Video, Macrovision, Veveo, and FanTV) has pioneered
`
`technologies that facilitate consumer access to television and other audiovisual
`
`media.
`
`Today, Rovi’s market-leading digital entertainment solutions are featured in
`
`interactive television program guides, program-search and channel-browse
`
`functions, video recording and retrieval, viewer-profile and favorites settings, and
`
`data analytics supporting program recommendation capability. Companies around
`
`the world use Rovi’s solutions in a wide variety of applications such as cable,
`
`satellite, and internet protocol television (“IPTV”) receivers (including digital
`
`television set-top boxes (“STBs”) and digital-video recorders (“DVRs”)).
`
`I.
`
`Background of the Technology
`
`In the past decade, Cable-television services and various subscription-based
`
`entertainment-content providers have exploded in popularity. These services allow
`
`a user to select content from large libraries or broadcast offerings for download,
`
`recording, queuing, or on-demand viewing in the context of a traditional
`
`3
`
`
`
`Case: 18-2422 Document: 26 Page: 11 Filed: 02/01/2019
`
`entertainment system, i.e., a television set with a set-top box for receiving signals
`
`from a provider and processing those signals for viewing on the television. Today,
`
`these services are standard in most American households.
`
`Finding desired content became more difficult for users as the amount of
`
`available media content massively increased in the 1990s and 2000s. In 1990, the
`
`average person had access to roughly 33 television channels. See Television
`
`Audience 2008, NIELSEN.COM, at 13, http://www.nielsen.com/content/dam/
`
`corporate/us/en/newswire/uploads/2009/07/tva_2008_071709.pdf (last visited Mar.
`
`16, 2016). By 2012, that number was nearly 800. See Cable’s Story, NCTA,
`
`https://www.ncta.com/who-we-are/our-story (last visited Mar. 16, 2016). The
`
`explosion in available television channels was accompanied by the growth of
`
`video-on-demand programming, which further increased the content available to
`
`consumers.
`
`The increasing search space created new problems with electronic program
`
`guides. One such problem was the challenge of giving users quick, easy-to-use,
`
`and accurate ways to find desired content using available input means and
`
`processing capabilities of existing entertainment-system devices. For example, not
`
`only was the user interface limited in a television environment, but the
`
`computational-load capacity of the hardware and networks through which
`
`television services are delivered was limited.
`
`4
`
`
`
`Case: 18-2422 Document: 26 Page: 12 Filed: 02/01/2019
`
`The defining user-interface tool in the television environment is the
`
`handheld remote control, typically the dominant, if not sole means for user input in
`
`normal operation of an entertainment system. As system functionality increased,
`
`remote controls needed more buttons to command those additional functions. At
`
`the same, there is a limit to how large and complicated a remote control can get,
`
`both due to physical size restrictions and user-friendliness. The size and
`
`complexity limitations of the handheld user interface were thus in tension with the
`
`ever-increasing functionality of the entertainment system.
`
`The patent at issue in this appeal, U.S. Pat. No. 8,433,696 (“the ’696
`
`patent”), describes and claims a system that searches a content database, such as a
`
`program schedule or video library, by inputting alphanumeric search terms using a
`
`handheld remote control. To limit the size and complexity of the remote, the
`
`alphanumeric symbols are presented on an overloaded keypad, which was a
`
`generally known apparatus at the time of the ’696 patent.
`
`
`
`5
`
`
`
`Case: 18-2422 Document: 26 Page: 13 Filed: 02/01/2019
`
`Because an overloaded key represents multiple alphanumeric characters, the
`
`use of overloaded keys presents the problem of ambiguity. For example, pressing
`
`the “2” key can mean any one of the alphanumeric characters “a,” “b,” “c,” or “2.”
`
`Thus, the use of overloaded keys requires some step to resolve this ambiguity
`
`(sometimes referred to as “disambiguation”). For example, as described in the
`
`Background section of the ’696 patent, it was known in the prior art that this
`
`ambiguity can be resolved by the user when inputting the keystrokes. In one
`
`method of disambiguation, the user presses the overloaded key multiple times,
`
`cycling through each character using repeated pushes until the desired letter is
`
`selected. Appx65, 1:50-55. For example, a user might push the “2” key once for
`
`the letter “A” and twice for the letter “B.” This method allowed the user to select
`
`the particular alphanumeric character desired, but the multiple presses were time-
`
`consuming, cumbersome, and prone to typographical error. Id. at 1:60-61.
`
`Another method of disambiguation by the user upon input was known as T9, which
`
`provided vocabulary based completion choices for each word entered. Id. at 1:55-
`
`57. The T9 method of disambiguation suffered from the drawback that it required
`
`that the user take the additional step of making a choice from a list of all of the
`
`possible word matches that were generated. Id. at 1:61-2:3.
`
`The ’696 patent went in an entirely different direction. Rather than require
`
`the user to disambiguate the overloaded keystroke, the system would simply accept
`
`6
`
`
`
`Case: 18-2422 Document: 26 Page: 14 Filed: 02/01/2019
`
`the ambiguous keystroke. Thus, for example, pressing the “2” key would
`
`intentionally correspond to any of the alphanumeric characters “A,” “B,” “C,” or
`
`“2.” The keystroke itself has no ability to specify among those characters. Entering
`
`multiple successive keystrokes in this fashion thus created ambiguous strings of
`
`numeric text, in which a particular numeric string could correspond to a number of
`
`different words or phrases. When ambiguous strings are used, more sophisticated
`
`algorithms are needed to process the input.
`
`The ’696 patent describes a unique algorithm for searching using ambiguous
`
`prefixes. The prefixes are pre-indexed with a direct mapping of each substring to
`
`one or more content items. The ’696 patent system allows a user to input search
`
`terms corresponding to various different descriptors associated with the content
`
`item, such as film genre, movie title, or actor name, using the overloaded keypad,
`
`and arrive at a desired content item. An example of the translation between
`
`alphanumeric search terms and ambiguous numeric prefix strings is shown in
`
`Figure 5, below.
`
`7
`
`
`
`Case: 18-2422 Document: 26 Page: 15 Filed: 02/01/2019
`
`
`As described in the specification, “[t]he exemplary terms ‘TOON’, ‘TOM’,
`
`‘TOMMY’, which can be search terms entered by a television viewer to identify
`
`television content, are mapped to the numeric equivalents of their prefix strings:
`
`‘T’(8), ‘TO’(86), ‘TOO’(866), ‘TOON’(8666), ‘TOMMY’(86669).” Appx67,
`
`5:50-65. The ’696 patent explains that “[t]his many-to-many mapping scheme
`
`enables incremental search processing by enabling even a single character entered
`
`by the user to retrieve relevant results. This many-to-many mapping is done during
`
`an indexing phase for all terms that can be used to discover a result.” Appx67,
`
`5:66-6:4.
`
`The exemplary search terms TOON, TOM, and TOMMY correspond to the
`
`descriptors “Toon: cartoon,” “Tom Hanks,” and “Tommy Boy,” respectively. Each
`
`8
`
`
`
`Case: 18-2422 Document: 26 Page: 16 Filed: 02/01/2019
`
`of these descriptors is associated with one or more content items. The index
`
`associates each numeric character corresponding to an ambiguous overloaded
`
`keystroke of the prefix string with all of the content items that correspond to the
`
`descriptors that the prefix might represent. During a search, a user enters an
`
`ambiguous prefix string by pressing a series of overloaded keys, one keystroke at a
`
`time. The search algorithm displays the subset of content items that are directly
`
`mapped to that prefix. The algorithm parses each keystroke incrementally,
`
`continuously refining the subset of content items as the user enters each subsequent
`
`character. For example, if a user inputs the overloaded key “8” (representing “T,”
`
`“U,” “V,” and “8”), the system would return at least the items “Toon: Cartoon,”
`
`“Tom Hanks,” and “Tommy Boy,” because each item has the letter “T” in the first
`
`position of a word in the title for each item. A graphic illustration of the
`
`association between the strings of ambiguous keystrokes and the content items that
`
`are stored internally and used for searching is shown in Figure 7, below.
`
`9
`
`
`
`Case: 18-2422 Document: 26 Page: 17 Filed: 02/01/2019
`
`
`Figure 7 shows the mapping algorithm for each successive ambiguous
`
`keystroke entered on an overloaded keypad. The direct mapping of each
`
`ambiguous prefix to the corresponding content items is represented by the dotted
`
`line from the numeric entry to “TOP M 704.” Figure 7 illustrates the “TOP M”
`
`search results for each of 8, 86, 866, and 866…9, as well as more complicated
`
`entries like 86_2. Each “TOP M” box represents a different subset of content
`
`items. As more keystrokes are entered, the subset of content items represented by
`
`“TOP M” becomes more refined.
`
`The incremental search feature of the algorithm saves time by allowing the
`
`user to make a final selection of a content item based on the shortest successful
`
`10
`
`
`
`Case: 18-2422 Document: 26 Page: 18 Filed: 02/01/2019
`
`search string, rather than entering an entire descriptor and then executing a search.
`
`In other words, the user can stop entering the characters in a descriptor and make a
`
`selection as soon as the desired result is displayed.
`
`Because matching the ambiguous input will result in an increased set of
`
`matching content items—when compared to matching to an unambiguous input—
`
`the algorithm of the ’696 patent includes additional processing to optimize the
`
`display of search results. Accordingly, the results are displayed in a particular
`
`order when presented to the user based on criteria, including, e.g., temporal
`
`relevance, location relevance, popularity, and personal preferences.
`
`Overall, the system of the ’696 patent accommodates ambiguous search
`
`strings in combination with incremental searching, allowing the user to receive the
`
`most relevant matching search results with as few keystrokes as possible. This
`
`system for mapping ambiguous search strings to content items produces the most
`
`accurate and comprehensive search results, from the broadest possible universe of
`
`descriptor search terms, with the shortest possible search string, created from the
`
`fewest possible keystrokes. The ’696 patent thus describes a unique search
`
`algorithm combined with the groundbreaking concept of implementing that
`
`algorithm in an interactive program guide for a television-entertainment system.
`
`The quality of Rovi’s innovative interactive program guide technologies,
`
`such as the technology described in the ’696 patent, have been recognized through
`
`11
`
`
`
`Case: 18-2422 Document: 26 Page: 19 Filed: 02/01/2019
`
`numerous industry awards and accolades. For example, in 2012, Rovi was awarded
`
`a Technology and Engineering Emmy® Award for its “Pioneering On-Screen
`
`Interactive Program Guides” that assist “viewer[s] in rapidly locating their desired
`
`program.” These Emmy® awards are designed to recognize “developments . . .
`
`involved in engineering technologies which either represent so extensive an
`
`improvement on existing methods or are so innovative in nature that they
`
`materially have affected the transmission, recording, or reception of television.”
`
`Technology & Engineering, The National Academy of Television Arts & Sciences,
`
`http://emmyonline.com/tech (last visited Mar. 28, 2016).
`
`II. Background of the Present Dispute
`Veveo has pioneered media technology innovation, including the technology
`
`used to facilitate consumer access to, and discovery of, television and other
`
`audiovisual media. Veveo’s technology is used daily in millions of American
`
`homes, and the ’696 patent is essential to any cable provider with an overloaded-
`
`keypad remote. Due to the importance and value of Rovi’s patented technologies,
`
`every major U.S. Pay-TV provider, including Defendant-Appellee Comcast Cable
`
`Communications, LLC, has taken a license to a portfolio of Rovi’s patents.
`
`Comcast first licensed Rovi’s technology around 15 years ago. Comcast
`
`spent the next decade building a cable empire on the back of that technology.
`
`Comcast still uses Veveo’s technology in various Xfinity products on the market
`
`12
`
`
`
`Case: 18-2422 Document: 26 Page: 20 Filed: 02/01/2019
`
`today. Comcast’s license expired on March 31, 2016. Rather than renew its license,
`
`Comcast decided to infringe Rovi’s patents, including Veveo’s patents, and file
`
`IPRs on all patents that Rovi and Veveo attempted to assert. The ’696 patent at
`
`issue in this appeal is one of those patents.
`
`The original provisional applications that led to the issuance of the ’696
`
`Patent were filed on August 26, 2005 and September 12, 2005. Comcast filed two
`
`petitions for inter partes review of all claims of the ’696 patent on January 19,
`
`2017, alleging that various combinations of prior art rendered the claims obvious.
`
`The Board declined to institute review for one of the two petitions. The Board
`
`instituted review for the other petition on July 28, 2017.
`
`The instituted petition challenges all claims of the ’696 patent as obvious.
`
`Independent claim 1 recites a method and independent claim 15 recites a
`
`corresponding system for performing that method. Claim 1 is representative of the
`
`challenged claims:
`
`1. A method, comprising:
`
`associating subsets of content items with corresponding strings of one or
`more overloaded keys of a keypad so that the subsets of content items are
`directly mapped to the corresponding strings of one or more overloaded keys
`by a direct mapping, wherein at least one overloaded key of the one or more
`overloaded keys is associated with a plurality of alphabetical and/or
`numerical symbols;
`
`ranking content items within at least one of the subsets of content items
`according to one or more ordering criteria;
`
`13
`
`
`
`Case: 18-2422 Document: 26 Page: 21 Filed: 02/01/2019
`
`subsequent to the associating and ranking, receiving entry of a first
`overloaded key;
`
`selecting and presenting a first of the subsets of content items that is
`associated with the first overloaded key based on the direct mapping;
`
`subsequent to receiving entry of the first overloaded key, receiving entry of a
`second overloaded key the same as or different than the first overloaded key,
`the second overloaded key forming a string with the first overloaded key;
`and
`
`selecting and presenting a second of the subsets of content items that is
`associated with the string of overloaded keys formed by the first overloaded
`key and the second overloaded key based on the direct mapping.
`
`Appx68, 7:62-8:20.
`
`
`
`The Board recognized that content items, as used in the patent, refer to
`
`movies, television shows, and similar video files that can be indexed in a video
`
`library or database using descriptors. Appx515-518. The Board also noted that
`
`descriptors are words that describe a content item, such as a title, director, or actor,
`
`and are not the content items themselves. Appx6 (“A person of ordinary skill in the
`
`art reading the claims in light of the specification would therefore understand that
`
`‘content items’ are not ‘descriptors.’”); Appx515-516. Descriptors are used as
`
`search queries to search for content items that are matched to those descriptors in
`
`an index. Subject to this understanding, the Board construed “content item” as “an
`
`item which contains information and is identifiable and selectable from a set of
`
`items through use of a search query.” Appx6.
`
`14
`
`
`
`Case: 18-2422 Document: 26 Page: 22 Filed: 02/01/2019
`
`The Board then construed “direct mapping” to mean “matching each
`
`alphanumeric character of a descriptor identifying a content item with its
`
`corresponding numeric key equivalent on an overloaded keypad.” Appx9. In other
`
`words, the Board found that content items are directly mapped to strings of
`
`overloaded keys by matching the descriptors corresponding to those content items
`
`with the numeric key equivalents of each letter in the descriptor. Appx8-9. The
`
`step of mapping strings of overloaded keystrokes to prefix strings of descriptors is
`
`more specifically recited in claims 3 and 17, which depend from claims 1 and 15,
`
`respectively, and recite a process of building the index that achieves the direct
`
`mapping recited in claim 1. Appx68, 8:25-31.
`
`The Board then found that “Gross teaches or suggests all the limitations of
`
`independent claims except for the recited ‘overloaded keys of a keypad,’” which it
`
`found in the Smith reference. Appx32. Based on these findings, the Board found
`
`that it would have been obvious for a person of ordinary skill in the art to have
`
`combined Gross with Smith to arrive at the claims of the ’696 patent.
`
`III. The Alleged Prior Art
`A. U.S. Pat. App. Pub. No. 2004/0133564 “Gross”
`Gross relates to a personal-computer program. It describes an incremental
`
`search tool for searching a variety of text-based files, such as word documents,
`
`emails, and web pages. Appx775 ¶ [0010]. Gross explains that incremental
`
`15
`
`
`
`Case: 18-2422 Document: 26 Page: 23 Filed: 02/01/2019
`
`searching displays and narrows search results continuously as each character of a
`
`search term is entered, thereby providing the user faster results. Id. Returning a
`
`search result before the complete search term is entered is referred to in the art (and
`
`in the record in this case) as searching based on “prefixes.” The example given in
`
`Gross is for the search term “dog,” delivering results based on the prefixes “d” and
`
`“do” as the user types each letter of the word.
`
`The search terms in Gross are not descriptors as described in the ’696 patent
`
`because they are words that appear within the content item—not words that
`
`describe the content item. Put differently, the search terms in Gross are words that
`
`appear in the body text of the documents being searched. A user enters the search
`
`terms based on the user’s familiarity with the content of the word document that he
`
`desires to locate. The search algorithm in Gross merely matches the letters in the
`
`search term to the letters in the words of the documents. No index or metadata is
`
`required to perform the search.
`
`While the Board also found that Gross teaches generating indices that
`
`contain content information and attribute information corresponding to the search
`
`targets, such as file name, date created, and author, these indices are not the indices
`
`discussed in the ’696 patent. Appx11. The index in Gross is designed to help a user
`
`determine which document among the search results is the document he is looking
`
`for, by providing information about the document rather than the mere fact that the
`
`16
`
`
`
`Case: 18-2422 Document: 26 Page: 24 Filed: 02/01/2019
`
`search term appears somewhere within it. Appx777 ¶ [0045]. The indexing
`
`functionality in Gross, therefore, is a different operation than the indexing
`
`described in the ’696 patent and serves an entirely different purpose. Appx778 ¶
`
`[0051].
`
`Contrastingly, the index in the ’696 patent is the tool with which the search
`
`algorithm matches each character in the ambiguous search prefixes to
`
`corresponding content items, such that the system is able to incrementally return
`
`results with each subsequent key entered. The many-to-many mapping used to
`
`create the index allows the system to return results without a disambiguation step.
`
`Gross does not contemplate ambiguous search terms at all, and its index does
`
`nothing to map content items to numeric prefix strings.
`
`For example, a user of Gross’s system may search based on a word in the
`
`body of the desired document. The search results, delivered incrementally, would
`
`show the files that contain the search term in the file name and would display
`
`information like the file type, author, or creation date in column format. Id. The
`
`indexing system in Gross also allows faster searching by confining searches to
`
`words that appear in certain attribute fields. Id. For example, a user can refine his
`
`search to a character string that appears in the “sender” field of an email, rather
`
`than anywhere in the file, to narrow the search task. Id.
`
`17
`
`
`
`Case: 18-2422 Document: 26 Page: 25 Filed: 02/01/2019
`
`The ’696 patent does not use indexing to limit the universe of a search.
`
`Indeed, the method described in the ’696 patent does the opposite. The ’696 patent
`
`uses indexing to compile all possible search fields into a single operation. Thus, a
`
`user entering the search string “866” would obtain search results for the actor
`
`“Tom Hanks,” the title “Tommy Boy,” and the genera “cartoon.” The index in the
`
`’696 patent is therefore designed to expand the universe of search results—not
`
`narrow it as in Gross—and does so using a completely different methodology than
`
`the system in Gross.
`
`Furthermore, there is no dispute that Gross does not contemplate use with an
`
`overloaded keypad, processing ambiguous search strings, or use in a television
`
`environment. Appx17. Indeed, the method in Gross is implemented in a personal
`
`computer environment, where a user has a full “qwerty” keyboard with which to
`
`enter unambiguous search terms. The user in Gross, who is searching for files on a
`
`computer, has different search priorities and objectives than the user looking for
`
`entertainment on TV. It is therefore unsurprising that Gross’s methodology is
`
`focused on precision and organization in searching, while the ’696 patent
`
`methodology is designed to maximize breadth and flexibility in processing results.
`
`Because Gross only accepts input of conventional, unambiguous
`
`alphanumeric keys (e.g., where pressing the “a” key alone has only one meaning—
`
`“a”), Gross has no need for, and therefore does not disclose “direct mapping,” as
`
`18
`
`
`
`Case: 18-2422 Document: 26 Page: 26 Filed: 02/01/2019
`
`required by the claims of the ’696 patent. For example, Gross discloses an index
`
`that simply contains prefix strings of words, e.g., “d” and “do” for the word “dog,”
`
`that appear in the documents themselves. Appx778 ¶ [0051]. Thus, the reference in
`
`Gross to the use of a “computer networkable wireless phone” (Appx779 [0060]),
`
`relied on by the Board,Appx14, Appx30, Appx32, Appx33, Appx41, does not
`
`suggest the use of the overloaded keypad of the wireless phone for the input of
`
`ambiguous overloaded keys, e.g., where the pressing of the “2” key has the
`
`ambiguous input of “a,” “b,” “c