`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`APPLE INC.,
`Petitioner,
`
`v.
`
`QUALCOMM INCORPORATED,
`Patent Owner.
`____________
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`____________
`
`Record of Oral Hearing
`Held: October 10, 2019
`____________
`
`
`
`
`Before TREVOR M. JEFFERSON, AMANDA F. WIEKER, and
`AARON W. MOORE, Administrative Patent Judges.
`
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`GRETCHEN A. DeVRIES, Ph.D., ESQ.
`TIMOTHY W. RIFFE, ESQ.
`W. KARL RENNER, ESQ,
`Fish & Richardson P.C.
`1000 Maine Avenue, S.W.
`Washington, D.C. 20024
`202-626-6447
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`CHAD C. WALTERS, ESQ.
`CHARLES Y. YEH, ESQ.
`Baker Botts, LLP
`2001 Ross Avenue
`Dallas, Texas 75201-2980
`214-953-6511
`
`
`
`
`The above-entitled matter came on for hearing on Thursday, October
`
`10, 2019, commencing at 9:00 a.m., at the U.S. Patent and Trademark
`Office, 600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`
`
`
`
`
`
`
`2
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`
`P R O C E E D I N G S
`- - - - -
`
`
`
`THE USHER: All rise.
`JUDGE WIEKER: Please have a seat. Good morning everyone. We
`are here today to hear argument in Case Number IPR2018-01245, between
`Apple Inc. and Qualcomm Incorporated, concerning U.S. Patent Number
`8,665,239 B2.
`I’m Judge Wieker. And with me is, Judge Jefferson and Judge
`Moore. Counsels for the parties, please introduce yourselves, starting with
`Petitioner.
`MR. RENNER: Good morning, Your Honors. This is Karl Renner,
`from Fish & Richardson. I’m joined by colleagues Tim Riffe and Gretchen
`DeVries.
`JUDGE WIEKER: Thanks for being here. Will you be arguing
`today?
`MR. RENNER: Actually, Gretchen DeVries will be starting us off
`today. And we’ll plan -- just to get the formality out of the way -- to reserve
`25 minutes in redirect.
`JUDGE WIEKER: Thank you.
`MR. RENNER: Thank you.
`JUDGE WIEKER: And for Patent Owner?
`MR. WALTERS: Good morning, Your Honors. This is Chad
`Walters of Baker Botts, for Patent Owner Qualcomm. With me is my
`colleague Charles Yeh. We also have a number of representatives for
`Qualcomm. They have a number of IPRs this week to that they (inaudible).
`Thank you.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`3
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`
`JUDGE WIEKER: Okay. And you’ll be arguing today?
`MR. WALTERS: Yes.
`JUDGE WIEKER: Thanks. As discussed in our September 23rd
`Order, each side will have an hour of total time to argue. Petitioner will go
`first. Patent Owner will go second. Both parties may reserve rebuttal time,
`and I understand that Petitioner would like to reserve 25 minutes. Mr.
`Walters, would you like to reserve rebuttal time?
`MR. WALTERS: I would, Your Honor, 20 minutes, please.
`JUDGE WIEKER: Okay. Thank you. And just a few quick
`reminders. Petitioner bears the burden of proving any proposition of
`unpatentability by a preponderance of the evidence. Also, as a reminder,
`this hearing is open to the public. A full transcript will be part of the record
`when it’s completed. And finally, please remember during your presentation
`to identify each demonstrative slide number as you refer to it, so it is
`reflected in the record.
`And with that, Ms. DeVries, you’re free to begin when you’re ready.
`MS. DeVRIES: Good morning, Your Honors. I’m Gretchen
`DeVries. And I’d like to talk to you about the four claims of the ’239
`Patent, all of which are directed to the broad idea of computer-implemented
`recognizing of gestures that are drawn on a touchscreen interface, and then
`the execution of actions that correspond to those gestures. This gesture
`recognition approach of the ’239 Patent is nothing inventive though, as
`we’ve demonstrated in the two art grounds we advanced in the petition.
`Qualcomm has proposed a set of substitute claims in the motion to
`amend, and we’ve demonstrated also that the substitute claims are
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`4
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`unpatentable, as well as that the motion to amend should be denied on
`procedure.
`Slide 4, please. We advanced two distinct art grounds in the petition,
`each of which addresses all four claims of the ’239 Patent. Today rather than
`walking stepwise through each ground in each claim, we’d like to focus on
`just a few issues that we think would benefit from further discussion.
`Slide 7, please. But before we get into the grounds, I’d like to take
`just a brief look at claim 1 of the ’239 Patent, which recites only generic
`steps related to receiving and responding to touch input. Steps such as
`determining a magnitude of a touch on the touchscreen, that’s in the middle
`of the left-hand column.
`Identifying a gesture based on properties of the touch, which you see
`also in the left-hand column of the claim. And identifying and executing an
`operation that corresponds to that identified gesture, and that’s at the top of
`the right column. These steps are not inventive, gesture recognition was
`well known in the art as of the critical date.
`And even the last two limitations of the claim, which specify that
`magnitude of the touch is involved in either gesture identification or in the
`execution of the operations, are not sufficient to make the claim inventive.
`As our grounds have demonstrated, as of the critical date it was common to
`rely on touch magnitude for gesture recognition and for execution of the
`corresponding actions.
`Slide 9, please. So, let’s turn to the art grounds. In our first ground
`which shows that the claims are obvious over the combination of Hullender
`and Renaud.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`5
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`
`Slide 19, please. Hullender discloses various approaches to computer
`implemented gesture recognition in a touchscreen context. Today I’d like to
`focus on just one aspect of Hullender’s disclosure, which is the role of
`Hullender’s data structure in gesture recognition.
`The claims of the ’239 Patent require that a gesture be identified from
`a collection of user gestures, and as we’ll discuss the evidence has shown
`that it was obvious to identify a gesture from Hullender’s data structure.
`This is the only aspect of Hullender that Qualcomm criticizes in this
`Hullender and Renaud ground.
`Slide 11, please. Hullender discloses two gesture recognition
`embodiments, each of which is a step-by-step process for identifying a
`gesture that’s input on a touchscreen. Looking at the upper left of the slide,
`this is the -- this is a portion of one of Hullender’s step-by-step gesture
`recognition embodiments. And you see that the last step in this process is to
`send a gesture ID to the -- I’m sorry -- system or application.
`Hullender also discloses a data structure that’s shown in the lower left
`of slide 11, and this data structure includes a gesture ID for each gesture
`that’s represented in that data structure.
`Now, Hullender doesn’t explain exactly what is the role of this data
`structure, other than a general indication that the data structure can go along
`with embodiments of the invention, but given, first, that we have
`Hullender’s instructions to send a gesture ID somewhere, as part of a gesture
`recognition process, that’s in the upper left, and that we also a data structure
`that contains gesture IDs, the evidence has shown that it was obvious to
`identify that gesture from the data structure that you see here, in order to
`enable compliance with the last step of that gesture recognition process.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`6
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`
`Slide 14, please. There are just --
`JUDGE MOORE: I’m sorry.
`MS. DeVRIES: Yes.
`JUDGE MOORE: How does Hullender identify a gesture?
`MS. DeVRIES: Hullender identifies a gesture, he describes a number
`of embodiments for gesture identification, one is described with respect to
`Figure 4, and one with respect to Figure 8, and they start with receiving a
`touch input on a screen to go through some processes that describe
`algorithms for gesture identification. And in one of these embodiments, the
`Figure 8 embodiment, it ends with sending the gesture ID through to a
`system or application, that’s what we just discussed.
`JUDGE MOORE: But it doesn’t -- does it describe exactly how the
`gesture is identified?
`MS. DeVRIES: It describes a few ways, let’s see. One of the
`processes of -- process with respect to Figure 4, and that describes how a
`gesture -- strokes that are input on a touchscreen can be identified as a
`gesture. There’s another process in Figure 8 which describes in some more
`detail another embodiment, by which a stroke. So, Hullender isn’t limited to
`just one approach to gesture recognition, it describes a few embodiments
`that can be used to recognize touches on a touchscreen as a gesture.
`JUDGE MOORE: Okay.
`MS. DeVRIES: All right. We are on slide 14. And I’d like to just
`mention two issues that I think are worthwhile to have a brief discussion
`with respect to this Hullender and Renaud ground. Both of these issues
`involve Qualcomm postulating some unsupported interpretations of
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`7
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`Hullender that Qualcomm alleges would preclude the use of Hullender’s
`data structure in gesture recognition.
`First, Qualcomm has argued that Hullender is limited to the use of
`Bayes nets in gesture recognition processes. But as the Board has already
`acknowledged, Bayes net are just an optional aspect of Hullender’s
`disclosure, they’re just an optional portion of one of those gesture
`recognition embodiments that Hullender discloses.
`The second issue relates to Qualcomm’s positing of some other
`unsupported alternatives, they proposed hypothetical explanations for things
`like where that gesture ID might come from, and how the data structure
`might be used. But these arguments failed because we’ve asserted
`obviousness with regard to the role of the data structure in gesture
`identification. And the possibility of some imagined other alternatives
`cannot disprove this obviousness contention.
`JUDGE WIEKER: Can I ask you a question? And I’ll be asking
`Patent Owner’s Counsel the same thing. So, the claim language says,
`“Identifying from the collection of user gestures, at least one user gesture . . .
`.”
`
`MS. DeVRIES: Mm-hmm.
`JUDGE WIEKER: So, the way I understand Patent Owner’s
`argument is that the identifying step must somehow be essentially using the
`data structure to do the identification process, and the problem with
`Hullender, as Qualcomm sees it, is that they use Bayes nets. So, do you
`think that’s a fair reading the claim language? Or do you think as we argued
`in -- or not argued -- as we found in the institution decision that the claim
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`8
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`language simply requires identifying a gesture from among those stored in
`the collection of user gestures?
`MS. DeVRIES: We are looking at the claim language as it reads on
`the page, which it says, identifying a gesture from that collection of user
`gestures.
`JUDGE WIEKER: Right.
`MS. DeVRIES: And we believe that doing so would have been
`obvious given what Hullender tells us. Hullender describes the use of this
`gesture ID, and then only provides one place where that gesture ID can be
`found. So, we believe that POSITA would have found it obvious to identify
`a gesture from that collection of user gestures, in order to -- you know, to
`implement the full process that Hullender has described.
`JUDGE WIEKER: Okay. Thank you.
`MS. DeVRIES: All right. I’m going to jump all the way to slide 36
`now please, and turn to our second ground.
`JUDGE WIEKER: Can we go back actually, and discuss the
`analogous art argument. Qualcomm argues essentially that every computer-
`based technology confronts issues of data storage and data processing, and
`where that happens, and that that alone wouldn’t make Renaud analogous to
`Hullender. Can you address that contention, please?
`MS. DeVRIES: Absolutely. We were planning to leave that to my
`end -- to the end only because my colleague, Tim, was going to do that. But
`we can certainly do it now if you’d like.
`JUDGE WIEKER: That’s okay. I don’t need to disrupt your flow.
`MS. DeVRIES: We just didn’t want to be jumping up and down.
`JUDGE WIEKER: Okay.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`9
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`
`MS. DeVRIES: All right. So well, certainly, we’ll turn to that at the
`end, but in the meantime we’ll turn to our second ground, which is that the
`claims are unpatentable over Kiraly and Agulnick.
`Slide 37, please? Kiraly discloses computer-implemented recognition
`of gestures, and execution of corresponding actions. So, Kiraly’s approach
`to gesture recognition matches well to many of the features of the ’239
`Patent claims. The primary piece that’s missing from Kiraly is that Kiraly is
`not a touch-screen system, Kiraly describes receiving gestures that are
`inputted on some kind of input device that’s distinct from the display, like a
`touchpad input system.
`And this is where Agulnick comes in. Slide 38, please. Agulnick also
`relates to computer-implemented gesture recognition, but this time in the
`context of a notebook computer with a touch-sensitive display surface.
`As you can see in the figure on the right of slide 38, Agulnick’s touch-
`sensitive display is implemented with a position digitizer, that’s pointed
`toward the blue arrow, and that position digitizer is mounted behind an
`LCD, which is shown in purple. This structure of Agulnick’s touch-
`sensitive display screen enables the display to sense the proximity of the
`stylus to the surface and also to sense contact between the stylus and the
`surface.
`And this proximity-sensing technology, both gives richness to the
`gesture available -- I’m sorry -- the gesture vocabulary that’s available in the
`Agulnick system. And also provides good feedback and usability to a user
`of the system.
`Slide 39, please. Now in the context of these two specific references,
`Kiraly and Agulnick, we walked through in the petition how to implement
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`10
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`Kiraly as a touchscreen system, in light of Agulnick’s teachings of
`touchscreen systems. And notably, as you can see in the middle of the left-
`hand column of slide 39, we give a specific example of how the structure of
`Agulnick’s touchscreen display, what we just discussed on the previous
`slide, could be incorporated into Kiraly’s system.
`That is, we discussed incorporating a position digitizer in conjunction
`with an LCD display. Now, naturally when we modify Kiraly to become a
`touchscreen system, the input mechanism of that modified system now
`changes such that gesture input would be provided via that touchscreen.
`And in particular, since we’ve discussed implementing Agulnick’s
`proximity sensing touchscreen in this modified system, the input would
`notbe provided via the proximity detection approach that’s enabled by
`Agulnick’s specific touchscreen.
`Slide 40, please? In the petition we noticed -- we noted both general
`advantages of incorporating a touchscreen into Kiraly’s system, as well as
`specific advantages that arise from integrating Agulnick’s particular
`touchscreen into the system. One advantage of a generic touchscreen that’s
`easy to understand is the usability enhancement that comes about when a
`user in the touchscreen system can draw a gesture directly on the display,
`and directly over the displayed content.
`This direct interaction between the user and the display, and the
`display content, eliminates the disconnect that’s present in Kiraly’s system
`when a user has to use a separate input device like a mouse or a touchpad, to
`control a cursor that’s displayed on a separate display device. And so this
`direct interaction between the user, the display, and the content, results in a
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`11
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`system that provides better feedback and more intuitive operation for the
`user.
`
`And beyond these general advantages that are available by
`incorporating a touchscreen into Kiraly’s system, we also pointed out in the
`Petition how further advantages flow form incorporating Agulnick’s
`particular touchscreen into Kiraly’s system. How we can incorporate the
`particular proximity sensing technology of Agulnick’s touchscreen, and its
`advantages.
`Then as the record developed, when challenged, we expanded on its
`discussion of the advantages that flow from Agulnick’s particular proximity-
`sensing touchscreen. And in particular, Agulnick’s proximity-sensing
`technology enables for seamless sensing of the start and the end of a gesture.
`Kiraly by contrast, before its modification, uses an unwieldy
`triggering event to indicate the start and the end of a gesture. So, when we
`combine Kiraly with Agulnick’s proximity-sensing touchscreen, we obviate
`the need for this complexity of Kiraly’s triggering events, providing a
`simpler and more intuitive user experience.
`JUDGE WIEKER: And you said that the triggering discussion is new
`to the reply and in response to Patent Owner’s argument? Correct?
`MS. DeVRIES: The triggering discussion specifically is in the reply.
`JUDGE WIEKER: Okay.
`MS. DeVRIES: But we expanded on our arguments regarding
`motivation to combine that we presented in the Petition, and that’s shown on
`slide 40, the next step of those arguments.
`JUDGE WIEKER: Thank you.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`12
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`
`MS. DeVRIES: Slide 41, please. Qualcomm has repeatedly argued,
`both for the existing claims, and also later for the substitute claims that
`they’ve proposed in the motion to amend, that Kiraly and Agulnick are not
`combinable because of something called gorilla arm syndrome. This is the
`name for the discomfort and fatigue that users of vertically-oriented
`touchscreens are said by Qualcomm to sometimes experience.
`Vertically-oriented here refers, first to a display that’s more or less
`straight up -- straight up and down, like a standard desktop computer
`display. The record before you, including Qualcomm’s own evidence,
`demonstrates that gorilla arm syndrome, to the extent that it’s relevant at all,
`does not impede or undermine the motivation that drives the combination of
`Kiraly and Agulnick.
`Now, we note that in the Institution Decision you asked for further
`exploration of this issue, so I’d like to talk through this in a little more detail
`just to be sure that you have no lingering concerns here.
`Slide 42, please. Qualcomm’s entire gorilla arm argument is wholly
`reliant on the unsustainable assumption that Kiraly is limited to a vertical
`display orientation. But it’s not. And Qualcomm’s argument collapses
`when Kiraly is read without that artificial display configuration restriction
`that Qualcomm has tried to impose.
`At the very least, Kiraly is silent as to display configuration, and it
`seems that Qualcomm has tried to read this silence as a requirement for the
`display to be vertically oriented, but this makes no sense. Kiraly does refer
`to a computer system that can be a desktop computer, we see that at the
`lower-left of slide 42, but Kiraly makes clear that this is an exemplary
`computer system, and also notes in that very same sentence, that various
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`13
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`types of computer systems can be use. So, certainly the configuration of that
`exemplary computer system, 112, is not a limiting disclosure.
`The relevant question here is how Kiraly would be interpreted by a
`POSITA as of the critical date. And with that in mind, it’s clear that
`Kiraly’s display technology is not limiting to vertically-oriented
`touchscreens. For instance, as you see in the upper-right of slide 42, Kiraly
`lists a number of possible display devices, many of which are compatible
`with a wide-range of display configurations, including non-vertical displays.
`For instance I’ll point out that Kiraly mentions a liquid crystal device,
`and as we just saw, a liquid crystal device is implemented in Agulnick’s
`notebook computer, certainly not a vertical display device. Even
`Qualcomm’s expert, Dr. Wobbrock conceded that Kiraly’s display
`technology can be used in a tablet computer, which is another example of a
`non-vertical display orientation.
`So, why does this matter? It’s because gorilla arm syndrome is
`relevant, at most, to the question of implementing a touchscreen in a
`vertically-oriented display, and that’s not what we have in Kiraly.
`Slide 43, please. For non-vertical displays like portable displays,
`tablets and notebooks, Qualcomm’s own evidence demonstrates that
`touchscreens can perform well, and are not subject to gorilla arm syndrome.
`For instance, we discussed at length in our briefing two of
`Qualcomm’s exhibits, Exhibits 2004 and 2005, the Sears and Ahlstrom
`articles, both of these articles report experimental results that show that
`touchscreens, positioned at relatively low angles, are well liked and are
`comfortable to use.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`14
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`
`So, Qualcomm’s own evidence destroys its argument that
`incorporating a touchscreen into Kiraly would have given rise to gorilla arm
`syndrome. Moreover, Qualcomm cited’ another article, the Wired article,
`which is Exhibit 2003, and the Wired article points out that even vertical
`touchscreens can sometimes be desirable, like in kiosks. So, even if Kiraly
`were to be read as somehow having some kind of vertical display
`requirements, and that would be wrong, there would still be situations in
`which a touchscreen would be compatible with Kiraly’s disclosure.
`Slide 52, please. I’d like now to talk about the feature in the claims
`that recites determining a magnitude of a touch, which is shown in the
`Petition that Kiraly and Agulnick make this feature obvious in a number of
`ways.
`Today I’d like to discuss just one, which is that it was obvious to
`determine stroke size, but I’d also like to point out that that any of the ways
`we advanced in the Petition is sufficient for us to show that Kiraly and
`Agulnick have met this feature of the claim, whereas, Qualcomm has to
`demonstrate flaws with every single one of the positions that we set forth for
`this feature.
`Slide 53, please. To understand why it was obvious to determine
`stroke size, let’s start by looking at Agulnick. Agulnick teaches that the size
`of a gesture can affect how the corresponding action is executed. And this is
`illustrated in the example of Agulnick’s scratch-out gesture. This is shown
`and described on the right-hand side of slide 53. The scratched-out gesture
`has the shape of a vertically-compressed Z, and it invokes a delete action.
`The size of the scratch-out gesture indicates which letters are to be deleted.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`15
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`
`That is that the size of the gesture affects how the delete action is
`executed. As you can imagine having the size of a gesture, affect how the
`delete action is executed. As you can imagine having the size of a gesture
`affect its functionality is valuable in that it gives the user control over what
`the gesture does.
`Without size-dependent execution of this gesture, for example, how
`would a user be able to control the target of that delete action? The evidence
`shows that it was obvious to incorporate the size-dependent capability into
`the Kiraly/Agulnick combination, to retain the ability for the user to control
`their functionality of a gesture.
`Slide 54, please. Now, let’s turn to Kiraly. Kiraly, in the course of
`identifying a gesture, normalizes the gesture to take size out of
`consideration. And Qualcomm has argued that Kiraly, in its disclosure of
`normalization, somehow teaches away from any kind of use of size at all.
`But there’s a fundamental error in this argument, in that it conflates gesture
`identification with the distinct process of executing a corresponding action.
`Normalization for gesture identification does not preclude
`consideration of size for execution of an action. I’d just like to say that
`again. Normalization for gesture identification does not preclude
`consideration of size in the context of executing an action.
`Rather, size-dependent execution can be complementary to and
`subsequent to size independent gesture identification. And to demonstrate
`that this is true, let’s look again, at Agulnick’s scratch out gesture. As we
`said already, the scratch-out gesture invokes a delete action, and execution
`of that delete action depends on the size of the gesture. But identification of
`the gesture is not dependent on its size, two differently sized gestures, both
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`16
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`shaped like vertically compressed Zs, will both be identified as scratch-out
`gestures, as Qualcomm’s expert has confirmed.
`So this indicates that Agulnick has contemplated some sort of size
`independent gesture identification process that occurs before the delete
`action is executed. So, the example of Agulnick’s scratch-out gesture,
`demonstrates the feasibility of first taking size out of consideration for
`identifying gestures. This is what Kiraly does with its normalization,
`While, second, still taking size into account for the subsequent
`execution of actions. That is, the evidence demonstrates the compatibility of
`Kiraly’s normalization process, with a complementary process of having
`size affect the execution of an action. This is a far cry from the teaching
`away that Qualcomm has alleged.
`JUDGE MOORE: Is there agreement that the magnitude in the claim
`can refer to the magnitude of anything?
`MS. DeVRIES: The magnitude in the claim refers to the magnitude
`of the touch of the (crosstalk) --
`JUDGE MOORE: But any aspect of the touch?
`MS. DeVRIES: It appears that magnitude is broad enough in the
`claim to cover nearly any type of magnitude. In claim 2, which is the
`dependent claim, a few examples of magnitudes are specified, and in the
`context of claim 2, in the specification we don’t believe that there’s a
`specific type of magnitude that’s required in claim 1.
`JUDGE MOORE: But does the specification shed any light on what
`magnitude means in the claim?
`MS. DeVRIES: The specification gives a list of some examples of
`magnitudes, it discusses stroke length, stroke area, stroke speed, the force of
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`17
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`a touch on a surface, but there’s no definition of magnitude in the
`specification.
`JUDGE MOORE: Is there agreement between the parties of
`magnitude can be a magnitude of anything -- anything related to the stroke?
`MS. DeVRIES: I don’t think there’s been an explicit discussion on
`that point, but my understanding is Qualcomm, in regards to stroke size, is
`not contesting stroke size being a magnitude, they’re contesting the
`combinability of stroke size with Kiraly’s normalization process.
`JUDGE MOORE: Thank you.
`MS. DeVRIES: All right. Let’s turn to slide 63, please. Now, I’d
`like to turn to the motion to amend and discuss the set of substitute claims
`that Qualcomm has presented.
`Slide 64, please. The substitute claims introduced the idea that the
`identification of a gesture is based on a comparison with the collection of
`user gestures. Our briefing has addressed a number of issues with the
`substitute claims, but I’d like to focus here on just a few, in particular the
`obviousness of the claims over the art, and the non-responsiveness of claims
`6 and 8.
`We’ve introduced two new references here with regard to the
`substitute claims, Rojewski and Cortopassi, and these references apply, not
`only to features of the substitute claims, but they also apply well to many
`features of the existing claims, to the extent that we may have considered
`advancing grounds based on these references in the Petition had we had
`them available.
`Slide 69, please. We’ve shown first that all four substitute claims are
`unpatentable over Hullender and Rojewski. Interestingly, Qualcomm has
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`18
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`not contested the substance of Rojewski’s applicability to the features of the
`substitute claims, or the articulated motivation for the combination.
`Slide 70 please. Qualcomm’s one and only argument against the
`application of Hullender and Rojewski to the substitute claims is this:
`Qualcomm frames Hullender’s principle of operation as relating to accurate
`gesture recognition, and claims the incorporation of Rojewski would destroy
`that principle of operation. This argument attempts to read into Hullender a
`supposed principle of operations that is far too narrow, and unsupported by
`Hullender’s disclosure.
`A principle of operation is something significant. The Federal Circuit
`has said in In re Mouttet that it can be a high-level ability of a system. And
`there’s no indication that accuracy rises to this level of significance in
`Hullender.
`Hullender addresses accuracy only once, in paragraph 62, that’s at the
`lower right of slide 70, and there it is aspirationally, accuracy is presented as
`a goal at best, there’s nothing more in Hullender to support Qualcomm’s
`attempt to elevate this goal into the significance of a principle of operation.
`So, Qualcomm has failed to even identify a reasonable principle of
`operation, and that’s the basis on which its single argument rests and is
`fundamentally flawed.
`JUDGE WIEKER: What would your position be regarding
`Hullender’s principle of operation?
`MS. DeVRIES: We believe that Hullender’s principle of operation
`relates to the computer-implemented recognition of gestures, and execution
`of actions. And we think this reflects a high-level ability of the system and
`it’s supported by the way that Hullender articulates its disclosure.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`19
`
`
`
`Case IPR2018-01245
`Patent 8,665,239 B2
`
`
`Now, even if we accept accuracy, as Hullender’s principle of
`operation Qualcomm’s argument still fails because it has not demonstrated
`that Rojewski is incapable of accurate gesture of recognition.
`Slide 71, please. In fact, it’s worth noting that Rojewski is capable of
`accurate gesture recognition. Rojewski describes the use of its gestures in a
`digital security context, like for digital signatures. Now, you can’t rely on
`gestures to be used as digital signatures if you lack the ability to accurately
`identify those gestures, or at least you must be aiming for accuracy. And
`this is consistent with the aspirational role of accuracy, in Hullender’s
`system. So, even if Hullender’s principle of operation does relate to
`accuracy, there is still no indication that incorporating Rojewski would
`diminish the accuracy of the combination.
`Slide 72, please. Moreover, Rojewski brings yet other benefits to this
`combination, in terms of improved usability. For instance, Rojewski enables
`the pe
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
Still Working On It
This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.
Give it another minute or two to complete, and then try the refresh button.
A few More Minutes ... Still Working
It can take up to 5 minutes for us to download a document if the court servers are running slowly.
Thank you for your continued patience.
This document could not be displayed.
We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.
Your account does not support viewing this document.
You need a Paid Account to view this document. Click here to change your account type.
Your account does not support viewing this document.
Set your membership
status to view this document.
With a Docket Alarm membership, you'll
get a whole lot more, including:
- Up-to-date information for this case.
- Email alerts whenever there is an update.
- Full text search for other cases.
- Get email alerts whenever a new case matches your search.
One Moment Please
The filing “” is large (MB) and is being downloaded.
Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.
Your document is on its way!
If you do not receive the document in five minutes, contact support at support@docketalarm.com.
Sealed Document
We are unable to display this document, it may be under a court ordered seal.
If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.
Access Government Site
