`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`
`
`
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`
`
`
`MICROSOFT CORPORATION,
`
`Petitioner,
`
`v.
`
`BRADIUM TECHNOLOGIES LLC,
`
`Patent Owner.
`
`
`
`
`
`
`
`Case IPR2016-00448
`Patent 7,908,343 B2
`
`
`
`
`
`
`
`
`
`
`
`
`Paper 34
`
`PETITIONER’S REPLY TO PATENT OWNER’S RESPONSE
`
`February 6, 2017
`_____________
`
`

`

`TABLE OF CONTENTS
`
`UPDATED EXHIBIT LIST .................................................................................... II
`I.
`CLAIM CONSTRUCTION ........................................................................... 1
`A.
`“Limited Bandwidth Communications Channel” ................................ 1
`B.
`“Limited Communication Bandwidth Computer Device” ................... 2
`II. ALL CLAIM ELEMENTS ARE TAUGHT OR SUGGESTED BY
`REDDY IN VIEW OF HORNBACKER ....................................................... 3
`A.
`“Limited Bandwidth” Elements ........................................................... 3
`B.
`“Said processor operative to select said defined data parcel…
`and render said defined data parcel… for a progressive
`resolution enhancement” ...................................................................... 6
`Prioritization ......................................................................................... 8
`C.
`D. Data Structure Arguments .................................................................. 10
`III. A POSITA WOULD COMBINE REDDY AND HORNBACKER ............ 16
`A.
`Reddy and Hornbacker are Analogous art ......................................... 16
`B.
`Reddy Does not Teach Away from the Claimed Solutions ............... 17
`C.
`Reddy Encourages Use of its Teachings on a Device with
`Limited Bandwidth ............................................................................. 18
`Bradium Fails to Show that Reddy and Hornbacker are
`Incompatible ....................................................................................... 19
`IV. NO SECONDARY INDICIA OF NON-OBVIOUSNESS .......................... 21
`A. No Long-Felt Need ............................................................................. 21
`B. No Nexus and No Evidence of Praise, Licenses, or Commercial
`Success ............................................................................................... 22
`1.
`Purported awards: .................................................................... 24
`2.
`Purported Licenses and Funding Agreements: ........................ 24
`3.
`Navi2Go: .................................................................................. 25
`4.
`3DVU / Microsoft (Ex. 2013 at 2, Ex. 2015) preliminary
`discussions: .............................................................................. 26
`Conclusion regarding Secondary Considerations .................... 26
`5.
`LEVEL OF ORDINARY SKILL IN THE ART .......................................... 27
`V.
`VI. CONCLUSION ............................................................................................. 27
`
`D.
`
`- i -
`
`

`

`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`UPDATED EXHIBIT LIST
`
`Ex. 1001 U.S. Patent No. 7,908,343 B2 to Levanon et al. (“the 343 Patent”)
`
`Ex. 1002 U.S. Patent No. 8,924,506 B2 to Levanon et al. (“the 506 Patent”)
`
`Ex. 1003
`
`PCT Publication No. WO 99/41675 to Cecil V. Hornbacker, III
`(“Hornbacker”)
`
`Ex. 1004 Reddy et al., “TerraVision II: Visualizing Massive Terrain Databases
`in VRML,” IEEE Computer Graphics and Applications March/April
`1999, pp. 30-38 (“Reddy”) (with added paragraph numbers by
`Petitioner for ease of reference in the Petition)
`
`Ex. 1005 Declaration of Prof. William R. Michalson (“Michalson Decl.”) with
`Appendices
`
`Ex. 1006
`
`EP1070290 to Cecil V. Hornbacker, III
`
`Ex. 1007
`
`Printout of IEEE Explore citations to Reddy et al. (Ex. 1004)
`
`Ex. 1008 Printout of Google Scholar citations to Reddy et al. (Ex. 1004)
`
`Ex. 1009 Cover page and authenticating declaration of Reddy et al. (Ex. 1004)
`from British Library
`
`Ex. 1010 Cover page of Reddy et al. (Ex. 1004) from Linda Hall Library
`
`Ex. 1011
`
`Proof of Service dated Jan. 12, 2015 in Case No. 15-cv-00031-RGA,
`Bradium Technologies, LLC v. Microsoft Corp.
`
`Ex. 1012 Affidavit of Matthew C. Bernstein in Support of Petitioner's Motion
`for Pro Hac Vice Admission
`
`Ex. 1013 Affidavit of Evan S. Day in Support of Petitioner's Motion for Pro
`Hac Vice Admission
`
`LinkedIn Page for Isaac Levanon
`
`Ex. 1014
`
`
`
`- ii -
`
`

`

`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`
`
`
`
`
`IEEE 802.11a1999, Wikipedia, available at
`https://en.wikipedia.org/w/index.php?title=IEEE_802.11a-
`1999&oldid=740398108
`
`Ex. 1015
`
`
`
`
`
`
`Ex. 1016 Declaration of Prof. William R. Michalson in Support of Petitioner's
`Reply to Patent Owner's Response
`
`Ex. 1017 Declaration of Yonatan Lavi, with Exhibits A-F
`
`Ex. 1018 Deposition Transcript of Peggy Agouris, dated January 13, 2017
`
`Ex. 1019 Deposition Transcript of Isaac Levanon, dated January 18, 2017
`
`Ex. 1020
`
`Fujitsu Technical Reference Guide, Stylistic 2300, Copyright 1998
`
`Ex. 1021 Not Used in This Proceeding
`
`Ex. 1022
`
`The Universal Grid System, NGA Office of GEOINT Sciences,
`March 2007
`
`Ex. 1023 Wolford, B., FXT1: 3dfx Texture Compression, Last Updated
`September 14, 1999, available at http://web.archive.org/web/
`20000114134331/http://www.combatsim.com/htm/sept99/3dfx-
`tc1.htm
`
`Ex. 1024 U.S. Patent Publication No. 2008/0294332 A1 to Levanon et al.
`
`Ex. 1025 U.S. Patent No. 7,561,156 B2 to Levanon et al.
`
`Ex. 1026
`
`Excerpt of Initial Claim Chart, Bradium Techs. LLC v. Microsoft
`Corporation, No. 15-cv-31-RGA (D. Del)
`
`Ex. 1027 Bing Maps Preview App for Windows in the Window Store, available
`at http://web.archive.org/web/20140101045325/http:/apps.microsoft.
`com/windows/en-us/app/bing-maps-preview/75ce2a6a-8a25-4916-
`83d0-19b8e7b60787
`
`Ex. 1028 Windows 8 and 8.1 System Requirements, available at
`https://support.microsoft.com/en-us/help/12660/windows-8-system-
`requirements
`
`iii
`
`

`

`
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`Ex. 1029 Home and Home Office Computing, Compaq, available at
`http://web.archive.org/web/20000304135335/http://athome.compaq.co
`m/store/default.asp?cpqsid=0TGMCU2GPSSH2GAH00AKHBU1A1
`EDBHWB.
`
`Ex. 1030 Barclay, T. et al., Microsoft TerraServer: A Spatial Data Warehouse,
`Microsoft Research, June 1999.
`
`Ex. 1031
`
`
`
`
`Intel Microprocessor Quick Reference Guide - Product Family,
`available at http://www.intel.com/pressroom/kits/quickreffam.htm
`
`
`
`iv
`
`

`

`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`Microsoft’s Petition established by a preponderance of evidence that the
`
`claims would be obvious to a person of ordinary skill in the art (POSITA). For the
`
`reasons set forth below, Bradium does not rebut Microsoft’s showing.
`
`I.
`
`CLAIM CONSTRUCTION
`
`Microsoft proposes that “bandwidth” means just that—bandwidth. It
`
`doesn’t mean the physical type of channel (wired or wireless), the processing
`
`power of the processor, the form factor, or the purpose of the device utilizing that
`
`channel. Accordingly, “limited bandwidth communications channel” and “limited
`
`communication bandwidth computer device” require no further construction. Ex.
`
`1016, ¶¶ 14-30.
`
`“Limited Bandwidth Communications Channel”
`
`A.
`Bradium’s construction of “channel” is vague and unhelpful. Both parties’
`
`experts agree that “narrowband” is simply another way to say “limited bandwidth.”
`
`Ex. 1016, ¶ 18; Ex. 1018 at 36:13-39:8. The specification does not say that limited
`
`bandwidth communications channels must be wireless, just that wireless conditions
`
`may result in limited bandwidth. Ex. 1001, 3:6-9; Ex. 1016, ¶¶ 19-20.
`
`Bradium disregards a statement in the ‘343 Patent’s specification that high
`
`concurrent user load can cause an otherwise high bandwidth communications
`
`channel to be “limited.” Ex. 1001, 3:9-14. Dr. Michalson explains that Bradium’s
`
`distinction between limited bandwidth “channels” and “conditions” (Response at
`
`- 1 -
`
`

`

`9-10, 13-14) is irrelevant from the perspective of the client device, and a POSITA
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`would recognize the same design considerations apply. Ex. 1016, ¶¶ 21-23.
`
`Inventor Levanon admits that limited bandwidth channels “can be limited by the
`
`amount of users.” Ex. 1019 at 40:18-41:10. Therefore, a channel may also be
`
`limited bandwidth due to concurrent user load. Ex. 1016, ¶¶ 17-25.
`
`“Limited Communication Bandwidth Computer Device”
`
`B.
`Neither the claim language nor the specification supports Bradium’s attempt
`
`to redefine this term to processing power or size of the device. The ‘343 Patent
`
`describes alleged problems with the computational requirements of prior art image
`
`transmission methods (Ex. 1001, 2:38-3:6) as a separate problem from limited
`
`network bandwidths. Id., 3:7-31. Likewise, the specification identifies a need for
`
`a system that can do two distinct things: (1) “support small client systems… and
`
`[(2)] efficiently utilize low to very low bandwidth network connections.” Id., 3:32-
`
`36. A POSITA would therefore not conflate the processing power or size of the
`
`device with bandwidth. Ex. 1016, ¶¶ 26-30. Bradium’s “small client” redefinition
`
`is unsupported, does not clarify the claims, and has nothing to do with the claim
`
`language. Id.
`
`The co-inventor, Yonatan Lavi, further testifies that the early prototype
`
`alleged by Levanon to embody the claimed invention (Ex. 2004, ¶¶ 25-31) ran on a
`
`standard desktop computer utilizing dial-up internet connection and would not
`
`- 2 -
`
`

`

`have satisfied Bradium’s proposed construction. Ex. 1017, ¶ 16; Ex. 2020 at 1.
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`Meanwhile, in concurrent litigation, Bradium contradicts its current “small client”
`
`construction by accusing software which requires more advanced capabilities by
`
`any measure than a conventional desktop PC at the time of the alleged invention.
`
`Ex. 1016, ¶ 30; Ex. 1026, Ex. 1027, Ex. 1028.
`
`Because Bradium’s proposed construction is not the BRI and is contradicted
`
`by the evidence, the Board should construe this term according to its plain and
`
`ordinary meaning.
`
`II. ALL CLAIM ELEMENTS ARE TAUGHT OR SUGGESTED BY
`REDDY IN VIEW OF HORNBACKER
`A.
`As the Petition explained, the “bandwidth” elements of the challenged
`
`“Limited Bandwidth” Elements
`
`claims (“limited bandwidth communications channel” in claim 13 and “limited
`
`communication bandwidth computer device” in claim 1) are taught by both Reddy
`
`and Hornbacker. Paper 1 at 23-24, 26-27, 48-49. Bradium does not dispute
`
`Hornbacker teaches these elements, but ignores teachings in Reddy to use the
`
`geographic image viewing functionality described therein in situations such as
`
`military applications or disaster response scenarios. Ex. 1004, ¶ 48. These
`
`teachings would suggest to a POSITA to access the geographic data of Reddy in
`
`bandwidth-limited situations. Ex. 1016, ¶¶ 42-43. Even using Bradium’s
`
`erroneous distinction between limited bandwidth “channels” and “conditions,” a
`
`- 3 -
`
`

`

`POSITA would recognize the scenarios taught in Reddy would frequently
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`encounter limited bandwidth channels under Bradium’s restrictive definition, and
`
`the need for mobility in such scenarios would suggest using portable devices such
`
`as laptop computers, PDAs, etc. Ex. 1016, ¶¶ 46-48.
`
`Bradium’s “limited communication bandwidth computer device” arguments
`
`depend on its flawed construction, and do not respond to the Petition’s showing
`
`that this element is taught according to its plain and ordinary meaning. The Board
`
`already rejected Bradium’s remaining arguments (Paper 9 at 15-19) for the reasons
`
`discussed below.
`
`Bradium does not address the teachings of Reddy as a whole, only separate
`
`teachings of preferred embodiments. Bradium asserts that one embodiment of
`
`Reddy’s teachings, the TerraVision II software, could operate on a graphics
`
`workstation connected to a high-speed ATM network, while a “standard VRML
`
`browser on a laptop machine” is a separate embodiment. Response at 21-26; Ex.
`
`1004, ¶ 48.
`
`The prior art references must be read not just for the features of preferred
`
`embodiments, but for the full scope of what is taught or suggested to a POSITA,
`
`who is a person of ordinary creativity. The question is not whether a particular
`
`software embodiment taught by Reddy (e.g. TerraVision II) would run on a
`
`particular device or under particular conditions, but whether it would be obvious to
`
`- 4 -
`
`

`

`a POSITA in view of Reddy’s teachings to implement the claimed features using a
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`limited bandwidth channel or device. Dr. Michalson explains that a POSITA
`
`would be motivated by Reddy’s teachings of resource-constrained situations to
`
`implement the full range of Reddy’s teachings on mobile devices and with limited
`
`bandwidth. Ex. 1016, ¶¶ 32-40, 42-56.
`
`Bradium argues that the features of TerraVision II would not be viable on a
`
`VRML browser operating on a laptop or other mobile device, yet cites contrary
`
`evidence. Bradium cites pp. 4-5 of Ex. 2066, an April 1999 SRI webpage
`
`describing various ways that the geographic data could be accessed, as “contrasting
`
`TerraVision running on fast graphics workstation with accessing the data only via
`
`a standard browser.” Response at 22. Yet this exhibit states it is “feasible” that
`
`features provided by TerraVision “could be implemented for a standard VRML
`
`browser through the use of various Java scripts embedded in the scene, or running
`
`externally to the browser.” Ex. 2066 at 4. Ex. 2066 further discusses the authors’
`
`desire to extend TerraVision functionality to “commercial, off-the-shelf” software
`
`(id. at 1), enable “open solutions” for a “wide cross-section of users” and integrate
`
`VRML support “directly with Internet browser software” (id. at 2-3). Even if the
`
`Board only considers TerraVision II, Dr. Michalson explains the corresponding
`
`code, cited in Bradium’s Response at p. 5, could operate on a huge range of
`
`computer devices, including mobile devices. Ex. 1016, ¶¶ 50-54. Accordingly,
`
`- 5 -
`
`

`

`Reddy does not teach away from using all features relevant to the claims on a
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`limited communications bandwidth computer device. Id. A POSITA would,
`
`moreover, recognize that operations on a high-bandwidth network could be
`
`constrained to limited bandwidth by concurrent user load. Id., ¶ 43. For claim 3,
`
`Dr. Michalson explains that both laptops (“mobile computing devices”) and tablets
`
`(“PDAs” or “palm-top computers”) existed in 1999 that were capable of hosting
`
`software with the functionality taught by Reddy. Ex. 1020; Ex. 1016, ¶¶ 52-54.
`
`These claim elements are therefore obvious to a POSITA over the teachings
`
`of Reddy and Hornbacker as a whole, both under their plain and ordinary meaning
`
`and Bradium’s incorrect narrow constructions.1
`
`B.
`
`“Said processor operative to select said defined data parcel… and
`render said defined data parcel… for a progressive resolution
`enhancement”
`
`Bradium’s Response (pp. 26-28) incorrectly paraphrases claim 13’s
`
`“processor” elements, which require a processor to select a defined data parcel and
`
`render that data parcel to provide for progressive resolution enhancement. The
`
`claim language does not link selecting a data parcel with progressive resolution
`
`enhancement in the manner argued by Bradium. Reddy, however, meets this claim
`
`element even under Bradium’s interpretation because it teaches that the processor
`
`
`1 The Board did not “misread” Reddy in its Institution Decision. Ex. 1016, ¶ 56.
`
`- 6 -
`
`

`

`selects data parcels (tiles) as it loads the appropriate tiles in a coarse-to-fine
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`manner. Ex. 1004, ¶ 44; Ex. 1016, ¶ 58.
`
`Bradium argues that Reddy teaches requesting tiles based on user
`
`movements using a “fixed” algorithm. Response at 27. Nothing in the claims
`
`excludes the processor selecting data parcels based on user actions; indeed, claim
`
`16, dependent on claim 13, requires this. Ex. 1001 at 14:1-6. Bradium’s “fixed”
`
`algorithm arguments are incorrect because a processor inherently must execute
`
`instructions in an algorithm in order to operate. Ex. 1016, ¶ 60.
`
`Bradium further argues that “TerraVision’s use of available low resolution
`
`data if higher-resolution data has not yet arrived over the network” doesn’t meet
`
`the claim element. Response at 27-28. The claim, however, only requires
`
`selection of one data parcel (“said data parcel”) and rendering that parcel to
`
`provide for progressive resolution enhancement, which Reddy teaches under the
`
`BRI of the claim whenever any given tile is loaded and displayed to replace a
`
`portion of a lower-resolution tile. Bradium further ignores Reddy’s teaching to use
`
`the progressive coarse-to-fine algorithm to “load and display” new data, which a
`
`POSITA would understand means downloading tiles at a series of resolutions to
`
`provide progressive resolution enhancement, even under a restrictive claim
`
`interpretation. Ex. 1004, ¶ 44; Ex. 1016, ¶¶ 57-60.
`
`- 7 -
`
`

`

`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`Prioritization
`
`C.
`As previously explained, Reddy teaches tiles are loaded in a coarse-to-fine
`
`manner, which necessarily means lower-resolution tiles are prioritized over higher-
`
`resolution tiles. Paper 1 at 45-47; Ex. 1004, ¶ 44; Ex. 1005, ¶¶ 190-199. A
`
`POSITA would recognize that loading lower-resolution tiles before higher-
`
`resolution tiles means the lower-resolution tiles are given a higher priority; i.e.,
`
`they are requested first. Ex. 1016, ¶¶ 62-63. The ‘343 Patent’s preferred
`
`embodiment itself utilizes a similar quad-tree data structure to retrieve lower-
`
`resolution tiles followed by higher-resolution “child” tiles. Ex. 1001, 9:1-36.
`
`Bradium’s arguments regarding the “priority” terms of dependent claims 10,
`
`11, and 15 (Response at 28-35) rely on an unexplained narrowing claim
`
`interpretation. Bradium argues that Reddy’s teaching of “pre-fetching” tiles based
`
`on flight path (Response at 30, citing Ex. 1004, ¶ 46) doesn’t disclose
`
`“prioritization,” but fails to consider how a POSITA would interpret such
`
`disclosures in view of teachings of Reddy as a whole and the knowledge of a
`
`POSITA. A POSITA would recognize that such pre-fetching based on an
`
`extrapolated flight path would logically prioritize tiles that are needed to retrieve
`
`earlier views. For example, Fig. 1 of Reddy and the accompanying text (¶¶ 12-17)
`
`teach that resolution is viewpoint-dependent. Therefore, it would be obvious to a
`
`POSITA that when the tile downloading algorithm of Reddy requests tiles based
`
`- 8 -
`
`

`

`on the user’s extrapolated flight path, it would request and download lower
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`resolution (coarser) tiles in the flight path with a higher priority than higher
`
`resolution tiles. For example, as shown below in Fig. 1 (annotated), for a projected
`
`flight path along the path shown by the arrow, the browser would first request the
`
`red lower-resolution tile (high priority), then the green tile, then the blue high-
`
`resolution tile (lowest priority). Ex. 1016, ¶¶ 61-68.
`
`
`
`Regarding claims 10 and 11, Bradium relies on a narrowing construction of
`
`“prioritization value” (“assigning a discrete, quantifiable value to an image parcel
`
`request in the form of a variable in memory space”) which is not the BRI and is
`
`supported only by conclusory expert statements. Response at 34, Ex. 2003, ¶ 107.
`
`Bradium’s claim differentiation argument is incorrect because claims 10/11 and
`
`- 9 -
`
`

`

`15/16 depend from different claims and differ in other respects. Ex. 1016, ¶ 69.
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`Bradium compares Reddy to the ‘343 Patent’s preferred algorithm, even though
`
`“that algorithm is not recited specifically in claim 10.” Response at 35, Paper 9 at
`
`38. Bradium’s attempted distinction between “assigning” and “comparing” values
`
`is unexplained and is insignificant to a POSITA. Ex. 1016, ¶ 62. Accordingly,
`
`Bradium fails to rebut the Petition’s showing or the Board’s previous finding that
`
`this term is taught by Reddy. Paper 1 at 45-48, Paper 9 at 36-39.
`
`D. Data Structure Arguments
`Bradium and its expert ignore the “image pyramid” shown in Fig. 1 of
`
`Reddy and discussed in the Petition (Paper 1 at 31-32, 38-39, 52-54), which shows
`
`an array or “image pyramid” of derivative tiles, each of which is unique to a
`
`particular location and resolution level, just like Fig. 2 of the ‘343 Patent:
`
`
`
`- 10 -
`
`

`

`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`
`
`Dr. Michalson explains that such structures were well-known for 20 years
`
`before the alleged invention date, and that it would be obvious to a POSITA that
`
`two-dimensional coordinates and a level of detail (LOD) are required to identify a
`
`specific tile. Ex. 1016, ¶¶ 70-77.
`
`Bradium admits (Response at 38-39) that Reddy teaches that image textures
`
`(including map images) are specified by URLs containing the image’s coordinates,
`
`which are specific for “a given image, geographical area, and LOD.” (Ex. 1004, ¶¶
`
`18, 22-23.) Bradium also does not address Microsoft’s analysis that in view of
`
`Reddy’s teachings that such tiles are (1) specific to a particular location and LOD
`
`and (2) located by URLs, that it would be obvious to locate tiles using URLs
`
`specifying their location and LOD. Ex. 1005, ¶¶ 138-145, 171-72.
`
`- 11 -
`
`

`

`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`Since it cannot seriously dispute that Reddy teaches a system comprising
`
`tiles stored in arrays based on coordinates and LOD, Bradium instead obfuscates
`
`the issue by pointing to various additional features of Reddy.
`
`Bradium argues that Reddy does not teach the claimed data structure
`
`because in its preferred embodiment, tiles are linked by nodes. Response at 38-39,
`
`41. But the ‘343 Patent’s claims are written in “comprising” format and do not
`
`specify how a data parcel must be located other than including the x, y coordinates
`
`and resolution. Nothing in the claims excludes utilizing quad-tree nodes to locate a
`
`tile at a particular location and resolution, which is how the ‘343 Patent’s preferred
`
`embodiment operates. Ex. 1001, 6:56-59, 7:54-57, 9:1-36; Ex. 1016, ¶¶ 95-97.
`
`Reddy’s Fig. 3 illustrates how arrays of terrain tiles including texture data (e.g. the
`
`map imagery shown in Fig. 1) relate to geotile structures so that the geotiles can be
`
`used to locate specific terrain tiles:
`
`
`
`- 12 -
`
`

`

`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`Nor do the claims of the ’343 Patent exclude the use of two-dimensional
`
`imagery combined with other data, such as a polygon mesh to depict elevation
`
`data. Bradium also argues the system of Reddy might hypothetically incorporate
`
`additional types of data that may occupy more than one X, Y coordinate, such as a
`
`cliff. Response at 39. Although the claims do not even exclude this scenario
`
`because the claim language only requires X, Y coordinates and does not exclude
`
`additional (e.g. Z) values, Reddy teaches displaying textures such as satellite
`
`imagery and aerial photography, which a POSITA would recognize as unique to
`
`particular coordinates. Ex. 1004, ¶¶ 3, 15-17, 22, Fig. 1; Ex. 1016, ¶¶ 40, 100-102.
`
`Therefore, while Reddy teaches some additional features, the simple claimed
`
`image pyramid data structure is taught or suggested to a POSITA by Reddy.
`
`Bradium’s argument that Reddy teaches an embodiment with a geocentric
`
`coordinate system ignores that Reddy teaches incorporating imagery from a wide
`
`variety of sources and coordinate systems, including systems such as Universal
`
`Transverse Mercator (UTM) which include an “X, Y” component, and uses local
`
`coordinate systems usable to locate imagery tiles by position and LOD. Ex. 1004,
`
`¶¶ 27, 29, p. 31 sidebar, p. 35 sidebar; Ex. 1022; Ex. 1016, ¶¶ 84-94.
`
`Bradium argues that Reddy’s coordinate system does not practice the claims
`
`because it is not “conformal” or “cylindrical.” Response at 37. These words
`
`appear nowhere in the ‘343 Patent, but are based on Microsoft products that
`
`- 13 -
`
`

`

`Bradium accused of infringement, violating the principle that claims cannot be
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`construed “in light of the accused device.” SRI Int'l v. Matsushita Elec. Corp. of
`
`Am., 775 F.2d 1107, 1118 (Fed. Cir. 1985). Even if the claims required a
`
`“conformal” or “cylindrical” coordinate system, Reddy teaches coordinate systems
`
`such as UTM meeting this definition. The Board previously rejected similar
`
`arguments and should again. Paper 9 at 31-32; Ex. 1016, ¶¶ 82-93.
`
`Bradium’s assertion that Dr. Michalson interprets the claims to cover “any
`
`file in any configuration” (Response at 40) is false and ignores Dr. Michalson’s
`
`actual statement (Ex. 1005, ¶ 169) that the tiles can meet the claim element if each
`
`tile is stored in a separate file having the required configuration, which Bradium
`
`does not dispute. The later paragraphs (Ex. 1005, ¶¶ 170-77) explain in detail why
`
`the claimed file configuration is obvious over Reddy and further obvious in view
`
`of Hornbacker and the knowledge of a POSITA. Bradium also fails to respond to
`
`Microsoft’s explanation that storing a hierarchy of images within a single file
`
`would have been well-known to a POSITA even under Bradium’s restrictive
`
`construction. Paper 1 at 40; Ex. 1005, ¶¶ 173-77; Ex. 1016, ¶¶ 104-105.
`
`Bradium’s efforts to distinguish Hornbacker’s teachings from this claim also
`
`fail. Ex. 1016, ¶¶ 41, 106-109. Bradium argues that Hornbacker does not create a
`
`“series of derivative images” because an embodiment of Hornbacker creates tiles
`
`“on request” (Response at 41-42). Yet nothing in the claims precludes processing
`
`- 14 -
`
`

`

`tiles on demand, which Claim 2 of the related ‘506 Patent (Ex. 1002) specifically
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`requires. Bradium further attempts to distinguish Hornbacker based on optional
`
`embodiments that “may be encoded” in the view tile name, e.g. rotation angle (Ex.
`
`1003, 9:20-28), but mischaracterizes such embodiments as limitations of
`
`Hornbacker’s teachings. Bradium ignores that Hornbacker also discloses
`
`embodiments in which no rotation angle (etc.) is used (id. at 8:30-9:19). Bradium
`
`also suggests that Hornbacker’s file structure doesn’t satisfy the claim element
`
`because the TILE_NUMBER (which Bradium admits includes X and Y
`
`coordinates) also includes tile width, yet the alleged 3DVU prototype (Ex. 2004, ¶¶
`
`25-30) also required a similar TileSize variable. Ex. 1017, ¶¶ 7-10; Ex. 2020 at 20.
`
`Nothing in the claims or the Board’s constructions requires the KD, X, Y, values to
`
`be “integers,” a word which appears nowhere in the specification and is an effort
`
`by Bradium to require a preferred embodiment rather than the BRI. Ex. 1016, ¶
`
`108. Nor does the BRI of the claims exclude combining the KD, X, and Y into a
`
`single “tile number” as described by Hornbacker, and Bradium itself asserts that
`
`the tile system described in Ex. 2059- which combines X and Y values into a
`
`single “quadkey” value without a separate resolution index- embodies the claims.
`
`Response at 37; Ex. 1016, ¶ 110; Ex. 2059 at 4-6:
`
`- 15 -
`
`

`

`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`
`
`Finally, the named inventors did not even design the ‘343 Patent’s preferred
`
`file structure- their prototype used a pre-existing file format developed by others.
`
`Ex. 2020 at 7; Ex. 1016, ¶ 111; Ex. 1017, ¶¶ 11-12; Ex. 1023. The data structure
`
`claim elements were well known to a POSITA and would have been obvious in
`
`view of Reddy and Hornbacker. Ex. 1016, ¶¶ 70-111.
`
`III. A POSITA WOULD COMBINE REDDY AND HORNBACKER
`A. Reddy and Hornbacker are Analogous art
`The Petition explained at length why Reddy and Hornbacker are analogous
`
`art (Paper 1 at 21-24). Bradium argues a POSITA would not consider the
`
`combination because Hornbacker’s preferred embodiment transmits documents.
`
`Response at 44. But both the GIS and document fields can include transmission of
`
`large-scale raster images (e.g. satellite imagery or pictures of large documents),
`
`and it is the overlap, rather than hypothetical difference, that is relevant to a
`
`- 16 -
`
`

`

`POSITA in order to determine whether to adopt teachings from one reference to
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`another. Bradium and its expert ignore that the TIFF file format (known for
`
`storing documents) was adapted for GIS as the GeoTIFF format. Paper 1 at 7; Ex.
`
`1005, ¶¶ 82-83; Ex. 1005 Apps. GG, HH; Ex. 1016, ¶¶ 120-122.
`
`B. Reddy Does not Teach Away from the Claimed Solutions
`Bradium’s “teaching away” arguments misstate and misapply the law. “A
`
`reference may be said to teach away when a person of ordinary skill, upon reading
`
`the reference, would be discouraged from following the path set out in the
`
`reference, or would be led in a direction divergent from the path that was taken by
`
`the applicant.” DePuy Spine, Inc. v. Medtronic Sofamor Danek, Inc., 567 F.3d
`
`1314, 1327 (Fed. Cir. 2009) (emphasis added) (quotations omitted). Bradium does
`
`not, however, argue that Reddy itself teaches away from progressive resolution
`
`enhancement, only that other, unrelated references suggest possible inefficiencies
`
`associated with “using multiple, smaller images at different resolutions to serve
`
`image and map data” and suggest alternatives. However, the “mere disclosure of
`
`more than one alternative” does not teach away, nor does a known feature such as
`
`using image pyramids “become patentable simply because it has been described as
`
`somewhat inferior to some other product for the same use.” SightSound Techs.,
`
`LLC v. Apple Inc., 809 F.3d 1307, 1320 (Fed. Cir. 2015); In re Mouttet, 686 F.3d
`
`1322, 1334 (Fed. Cir. 2012). Reddy itself specifically teaches a preferred
`
`- 17 -
`
`

`

`embodiment using multiple, smaller images at different resolutions to serve map
`
`PTAB Case No. IPR2016-00448, Patent No. 7,908,343 B2
`Petitioner's Reply to Patent Owner’s Response, Paper 34
`
`
`data, and a POSITA reading Reddy would not ignore that preferred embodiment.
`
`Ex. 1004, ¶¶ 3, 15-17, 20, 42, 44, Figs. 1-3; Ex. 1016, ¶¶ 123-127.
`
`C. Reddy Encourages Use of its Teachings on a Device with Limited
`Bandwidth
`
`Bradium’s argument that Reddy teaches away from a limited
`
`communications bandwidth computer device likewise relies not on Reddy, but on a
`
`different reference, Ex. 2005, which is briefly cited in Reddy. Response at 47-48.
`
`However, Ex. 2005 refers to a different system for retrieving and rendering models
`
`of detailed bathymetric data, not simply retrieving and displaying imagery as
`
`claimed by the ‘343 Patent and taught by Reddy. Ex. 1016, ¶¶ 130-132. A
`
`POSITA would recognize that downloading and rendering imagery for a particula