`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`__________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`__________
`
`
`NETFLIX, INC., ARRIS SOLUTIONS, INC., and
`COMCAST CABLE COMMUNICATIONS, LLC,
`Petitioner,
`
`v.
`
`REALTIME ADAPTIVE STREAMING, LLC,
`Patent Owner.
`
`__________
`
`Case IPR 2018-01169
`Patent 8,934,535 B2
`__________
`
`Record of Oral Hearing
`Held: September 4, 2019
`__________
`
`
`Before KEVIN W. CHERRY, GARTH D. BAER, and
`NABEEL U. KHAN, Administrative Patent Judges.
`
`
`
`
`
`
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`HARPER BATTS, ESQUIRE
`CHRIS PONDER, ESQUIRE
`Sheppard, Mullin, Richter & Hampton LLP
`379 Lytton Avenue
`Palo Alto, California 94301
`
`
`ON BEHALF OF THE PATENT OWNER:
`PHILIP WANG, ESQUIRE
`Russ August & Kabat
`12424 Wilshire Boulevard
`12th Floor
`Los Angeles, California 90025
`
`
`
`The above-entitled matter came on for hearing on Wednesday,
`
`September 4, 2019, commencing at 12:30 p.m. at the U.S. Patent and
`Trademark Office, 600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`
`
`
`
`
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`P R O C E E D I N G S
`- - - - -
`
`
`
`(Proceedings begin at 12:45 p.m.)
`JUDGE BAER: All right. Good afternoon, everybody.
`Again, apologies for the timing delay, but let's go ahead and get
`started.
`We have, this afternoon, our final hearing in IPR2018-01169
`between Petitioner, Netflix, and Patent Owner, Real Time Adaptive
`Streaming.
`Petitioner, Arris, from IPR2019-00674, and Petitioner,
`Comcast, from IPR2019-00684, have also been joined to this
`proceeding.
`The challenged patent number here in this case is US Patent
`No. 8,934,535. I'm Judge Baer. With me is Judge Cherry and
`appearing remotely is Judge Khan.
`Let's go ahead and get the parties' appearances, if we could,
`please. Who do we have for Petitioner, Netflix?
`MR. BATTS: Good afternoon, Your Honor. This is Harper
`Batts. With me is my colleague, Chris Ponder, on behalf of
`Petitioner, Netflix. And also in attendance with the joinder counsel
`is Jim Day from -- for Comcast and Andrew Wilson for Arris.
`JUDGE BAER: Great. Thank you, Mr. Batts.
`And for Patent Owner, Realtime?
`MR. WANG: Good afternoon, Your Honor. Philip Wang
`for Patent Owner, Realtime Adaptive Streaming.
`JUDGE BAER: Thank you, Mr. Wang.
`Welcome, everybody. It's good to have you here. We
`certainly do appreciate you making the effort to be here. We set
`
`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
`27
`
`3
`
`
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`forth the hearing for today's -- we set forth the procedure, rather, for
`today's hearing in our trial order that we issued.
`But just to remind everybody of how everything will go, first of
`all, am I correct that only Petitioner, Netflix, will be presenting on
`behalf of Petitioners today, correct?
`MR. BATTS: Correct, Your Honor. Just myself.
`JUDGE BAER: Great. Thank you.
`All right. The parties will each have -- the presenting parties
`will each have 60 minutes of total time to present. Please keep in
`mind that whatever you project on the screen Judge Khan will not be
`able to see, but he has the materials there in front of him.
`So, if you could just be really clear about what slide number
`you're referencing, what exhibit you're referencing, that will help
`Judge Khan to keep up. And it will also help to keep our record
`clear, if you would, please.
`We remind the parties that under no circumstances are you to
`interrupt the other party while they're making their presentation. If
`there's something that you believe is objectionable, that's fine. You
`can make that objection during your own time, but, please, during
`your own time only.
`I know that, Patent Owner, you had some objections to some of
`Petitioner's demonstratives. That's fine. You can make those
`arguments, but, please, only make those arguments during your time.
`Okay?
`Let's see. Does counsel for Petitioner, Netflix, have any
`questions before we get going?
`MR. BATTS: No, Your Honor. I have copies of the
`demonstratives, if you'd like them.
`
`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
`27
`28
`
`4
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`JUDGE BAER; That would be great. And, Mr. Wang, do
`you have copies as well?
`MR. WANG: Yes, I do, Your Honor.
`JUDGE BAER: Why don't we go ahead and distribute those.
`MR. BATTS: May I approach?
`JUDGE BAER: Please.
`JUDGE BAER: Okay. Petitioner, you're planning on using
`demonstratives as well; is that correct?
`MR. BATTS : Correct, Your Honor.
`JUDGE BAER: Okay. Why don't we take five minutes or
`so, then. Again, sorry for the delay.
`We'll take five minutes, we'll come back and see if we can get
`the Elmo working. Okay? All right. With that, we'll recess for
`five minutes.
`(Whereupon, the above-entitled matter went off the record at
`12:49 p.m. and resumed at 12:53 p.m.)
`JUDGE BAER: Please be seated. Okay. Looks like we
`have everything set up. So, when we left off, I was asking Petitioner
`-- I think I already asked you, but any questions before we get started?
`MR. BATTS: No, Your Honor.
`JUDGE BAER: And for Patent Owner, any questions before
`we get started?
`MR. WANG: No, Your Honor.
`JUDGE BAER: Great. Petitioner, you have the burden, so
`you will go first. Whenever you're ready, you may begin -- actually,
`first, we said 60 minutes of time. Do you want to reserve some time
`for rebuttal?
`MR. BATTS: Yes, Your Honor. I'd like to reserve 20
`minutes.
`
`5
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`JUDGE BAER: 20 minutes. I believe the light will flash
`green at 2 minutes, and then yellow at 1, but sometimes those
`numbers vary a little bit. I'll try and give you a heads up when
`you've got five minutes to go. Okay?
`MR. BATTS: Thank you, Your Honor.
`JUDGE BAER: No hurry, but begin whenever you're ready.
`MR. BATTS: Good afternoon, Your Honors.
`As you noted, this hearing is about the '535 patent. And in
`terms of where we stand for this, I'm putting up Slide No. 2 to give
`you a quick overview of where we are.
`This IPR involves only a single obviousness ground
`challenging all -- well, Claims 1 through 14. And in terms of where
`the arguments lie, and what we'll be discussing today, Patent Owner
`has primarily addressed its arguments related to Independent Claim 1.
`It did have some separate arguments regarding Claims 1 and 4.
`Although, the arguments for Claim 4 were dropped from the surreply,
`so I'm not sure they're still maintaining those arguments.
`And then the last argument is the proper construction of access
`profile. There's only one claim term that appears to be in dispute
`between the parties. And then for context, in a separate IPR, Claims
`15 to 30 have been disclaimed by the Patent Owner.
`And in terms of the obviousness ground that we're dealing with
`here today, we're dealing with a two reference obviousness ground
`relying upon the Imai and Ishii references.
`So, next, I'd like to turn to Slide 3 because I think a lot of our
`focus is going to be on Claim 1 here today and the language of Claim
`1.
`
`And before we get started into the in-depth arguments, I'd just
`like to provide a quick overview of, really, what Claim 1 is talking
`
`6
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`about and requiring, which is, first, determining a parameter attribute
`of some data, Element A; second, the selecting of an access profile --
`and we'll get into what access profile means, but selecting an access
`profile based upon that parameter or attribute; and then compressing
`the data by selecting a compressor based upon the information from
`the access profile. So, I think it's three, kind of, general concepts
`that are required in that claim that we'll get into today.
`And I think, also, it's useful for context to recall that in the file
`history for this patent, the claims that we're challenging here were
`rejected, and Patent Owner overcame that rejection by adding the
`asymmetric aspect to Element 1C.
`And in this IPR, Patent Owner is not contesting that the prior
`art discloses asymmetric compressors, and, instead, really that the
`battle seems to be around access profile and frequency of access
`teachings and a motivation to combine.
`So, I think it's also useful to recall what the patent teaches or
`doesn't teach in terms of the '535 patent that we're challenging here.
`So, on Slide 4, I've given an overview of really how heavily the '535
`patent relies upon the knowledge of a person skilled in the art for
`many of the topics that we're discussing today.
`As an example, the
`specification doesn't even refer to video or audio compressors. It
`only mentions the term video twice. So, there's a lot there that's
`being left to the POSITA and the POSITA's knowledge.
`I don't believe we have any dispute that a POSITA here would
`have video data compression experience as well as a relevant degree
`in the field.
`So, we're looking at POSITAs that would be knowledgeable
`and would have to know things for themselves rather than looking at
`the '535 patent to understand what it's talking about and how things,
`
`7
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`more importantly, are going to be implemented. So, another
`example, on the second bullet, is that the '535 patent doesn't go into
`any detail about how to implement these compression algorithms that
`are now claimed. It just says you use them in the claims rather than
`going into how would you build them, how would you implement
`them, how would you use them.
`The third bullet is going to get more complicated, I'm sure, and
`we'll go into more discussion on access profile, but I do think it's
`useful to look at, from the higher level, that there really isn't a lot of
`teachings in the '535 patent about how to make access profiles.
`Rather, there's some exemplars primarily in Columns 11 and 12 that
`provide you some context of what an access profile is.
`It certainly doesn't get into how would you determine the access
`frequency of a particular data block -- the term that the Patent likes to
`use -- or portions of a data block, it just provides a general, hey, you,
`use an access profile to select which compressor you want to
`compress the data with.
`And, I guess, again, looking at the higher level, if you look at
`the overall '535 patent, much of the patent does not relate to the claims
`or the technology that we're discussing here today.
`There's really a limited -- a number of columns that we're going
`to be discussing where there's anything about access profiles or the
`selection aspect that is relevant for the claim -- the challenged claims
`here.
`
`So, I'd like to go back to the institution decision, also, briefly
`for context. As Your Honors likely recall, the IPR was instituted and
`the Board found that there was sufficient motivation to combine and
`that there are rational explanations for why the Imai and Ishii
`references would be combined.
`
`8
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`And the Board rejected Realtime's argument about access
`frequency and whether the teachings that were provided in the petition
`were sufficient.
`And not only did the Board actually reject the argument in
`terms of what was being presented by Netflix, but the Board went on
`to say, citing the In re Keller decision, that that type of argument was
`an improper bodily incorporation argument.
` And I think what's very notable here is that Petitioner,
`Netflix, re-raised that same bodily incorporation argument in our reply
`brief, we cited the In re Mouttet decision and Keller -- very similar to
`Keller and some of these other decisions that we have about bodily
`incorporation and Realtime -- Patent Owner, Realtime, failed to ever
`address the bodily incorporation argument in its Patent Owner
`response or in its surreply.
`It never addressed Keller. It never addressed In re Mouttet.
`And so, that argument has been waived by Realtime, and I think their
`failure to address that argument concerning their access profile
`argument is fatal to their positions here today.
`So, then I think in terms of -- the first thing I'd like to go over
`with you, if I may, is the construction of access profile and what the
`appropriate construction of access profile should be.
`We'll go to Slide 9. I think Slide 9 provides a good overview
`of the different constructions that are relevant for our discussions here
`today.
`Petitioner proposed a construction, in its petition, that access
`profiles should be information regarding the number or frequency of
`reads or writes.
`And in the response -- in the preliminary response Realtime
`didn't formally propose a construction, but basically advocated a
`
`9
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`construction that it should instead be information that enables the
`controller to select a suitable compression algorithm that provides a
`desired balance between execution speed and efficiency. And at
`institution, the Board sided with Petitioner, Netflix, in adopting the
`Petitioner's construction.
`After institution, Patent Owner formally urged what I would
`call, I guess, a meta construction that it started with in its preliminary
`response and, again, urged the Board here to adopt its construction.
`And I do think it's relevant to note that there are separate IPR
`proceedings between the parties on the '046 patent that we referenced
`in our reply brief, and the Panel in the '046 decision -- '046
`proceeding, upon institution, took a different view as to the term
`access profile and actually adopted a broader construction than
`proposed by Netflix saying, information such as the number of
`frequency of reads or writes.
`JUDGE CHERRY: There's one more construction out there,
`right, in '1342? I think in the '1342 case there's another construction
`floating out there. I don't think we adopted it, but I think that the
`Petitioner in that case has another construction.
`MR. BATTS: I believe that that's the other --
`JUDGE CHERRY: Sling TV.
`MR. BATTS: -- the Sling IPR. I believe that there is still a
`disagreement there about access profile and whether data type can be
`access profile. And I'm going to discuss that as well today --
`JUDGE CHERRY: Okay.
`MR. BATTS: -- but I felt like the one for the '046 was the
`more relevant where a separate panel has made a determination about
`the same claim term and also rejected the same construction that
`Patent Owner is advocating here.
`
`10
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`We're not quite sure whether Patent Owner is still disputing this
`construction. In their reply brief for the '046 that was filed, I think,
`two weeks ago, they said they were no longer contesting the
`construction in that proceeding. And we asked them for a position
`and we're not quite sure, but we'll -- I'm sure we'll hear and I'll address
`it on rebuttal as necessary.
`So, if you go to Slide 12 -- or 11, the Patent, as you are likely
`aware, has a relatively lengthy discussion of access profile across
`Columns 11 and 12 providing some examples of what an access
`profile --
`JUDGE CHERRY: And just to be clear, between the '209
`construction and your construction, do you see any real difference or
`anything material to your case between those two constructions?
`MR. BATTS: I do, Your Honor. I believe that we should
`prevail on our construction alone. But if that broader construction is
`adopted, I think the teachings of Ishii are both teachings about
`frequency of access, as well as using data type for the selection of the
`appropriate compressor.
`So, if we want to go broader and we encompass a data type as
`being within the construction of access profile -- and by data type, I
`mean a file extension, for example, and I'll get into those examples --
`then I think there is multiple reasons why the Petition should be
`rejected under the proposed grounds. But I think either way we
`should --
`JUDGE CHERRY: It should be rejected.
`MR. BATTS: Correct.
`(Laughter.)
`MR. BATTS: As far as my position, but I'm happy to go into
`that in more detail.
`
`11
`
`
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`So, I think -- I was trying to provide some context for the
`access profile discussion here from Columns 11 and 12, and what we
`see is -- these are two tables at the bottom of Columns 11 and 12
`respectively that provided examples of what an access profile would
`be. And, in addition, I think there is some relevant language on 11
`that I'll just read off here that also describes the table.
`So, on Column 11 of the '535 patent starting at lines 35 to 39, it
`says: for instance, the data profiles may comprise a map that
`associates different data types based on e.g., a file extension with a
`preferred one of the compression algorithms. For example, preferred
`access profiles considered by the controller 11 are set forth in the
`following table.
`So, the table -- I don't think there's a dispute that the table is
`showing what access profiles are. And the access profiles in the
`table show -- looking at the relative, basically, slots for rough
`characterizations of what you would expect to be the reads or writes
`of the data that is to be compressed.
`And I think that's certainly relevant as we get into Realtime's
`very implementation-specific arguments later today about knowing
`precise access frequency of a particular data block or portion of a data
`block. If we go to --
`JUDGE KHAN: Counsel, is there any distinction between --
`what's the distinction between your construction and Patent Owner's
`construction here?
`Could it be that your construction is somehow encompassed by
`Patent Owner's construction? Could the frequency of reads and
`writes be included in information that enables the controller to select a
`suitable compression algorithm?
`
`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
`27
`28
`
`12
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`MR. BATTS: I do think it could be encompassed. I think
`that the issue -- and I am going to address the problems also with their
`proposed construction, but I think one of the biggest issues with their
`proposed construction is it really injects a Nautilus-type of uncertainty
`here to the claims where we're left to wonder what is desirable versus
`suitable. There were uncertainty that we believe are unnecessary to
`determine that the claims here -- whether the claims here are invalid.
`So, if we look at Slide 12, Realtime is primarily relying upon
`this one portion of the '535 patent from Column 8 that refers to the
`information enabling the controller to select a suitable compression
`algorithm that provides a desired balance between execution speed
`and efficiency; but the problem with that is that it really does use
`language that would leave uncertain where the balance of the claims
`would be.
`And I think that's demonstrated in Slide 13 where we asked
`Realtime's expert, Dr. Zeger, as to, well, what is the actual meaning?
`What would desired mean and how would we understand the claim
`scope to know what would be the desired balance?
`And as you can see from his answer here, he said, I think the
`understanding of desired would be in the eyes of the practitioner.
`And then he said, a person, for example, might have a particular
`desirable balance.
`So, I think that really highlights that the problem with Patent
`Owner's proposed construction is there's a lot of ambiguity in their
`construction as to what would be appropriate for a particular
`Petitioner versus another practitioner.
`And those types of indefinite and subjective terms, I think,
`really run afoul of Nautilus. And that's why I think their
`construction would be inappropriate.
`
`13
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`And if we go to Slide 13, at a later point in the deposition we
`also asked Dr. Zeger about how does the term suitable relate to
`desirable?
`And as you can see from the question and answer here, he
`really does a circular logic of saying, well, suitable is desirable,
`desirable is suitable.
`So, I think that goes to are we actually -- is Patent Owner's
`construction actually providing any more explanation or certainty or
`understanding as to what is the claim scope or is it doing actually the
`opposite, which is injecting a lot of uncertainty as to what would
`satisfy the claims.
`JUDGE BAER: Mr. Batts, does that matter in an IPR? In
`other words, in a district court claim, construction -- there's a canon of
`claim construction that says, we construe a claim -- one of the canons
`is to construe a claim such that the claim is valid. It's sort of a rather
`weak presumption, but does that matter here?
`I mean, what is your response to the response to your argument
`that, so what? Okay. Maybe our claim is invalid under Nautilus.
`It's not an issue here. We're only addressing prior art and validity
`here, so why does it matter whether their construction injects some
`indefiniteness into the claim?
`MR. BATTS: So, I think a couple of issues, Your Honor.
`One is -- if we look at, I think, Slide 29, one is that the arguments that
`Realtime is relying -- or the evidence that Realtime is arguing on --
`arguing for their construction, they're saying that Petitioner's
`construction is flawed because it's too narrow.
`And you see the language here from their Patent Owner
`response where they say, hey, for example, the specification states
`that an access profile may comprise a data type alone.
`
`14
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`And so, we called them out on that and basically said, in our
`reply brief, well, it can also maybe be data type alone, but then that's
`also satisfied by Ishii.
`So, if you want to broaden it out to include something along the
`lines in the separate IPR proceeding that the other panel has adopted
`and not be limited to information regarding reads or writes, but
`broaden it out somewhere further, that may be -- you know, we don't
`believe it's necessary, but we're not going to necessarily contest that
`because we think our art will satisfy it for even more reasons there.
`But if we go to Slide 3, you'll see that they flip-flopped in their
`surreply about -- the position about whether data type can be an access
`profile. So, they're not really -- the thing that they advocated to say
`that our construction was too narrow, they suddenly turned around
`and said, well, Patent Owner hasn't argued that its construction
`necessarily encompasses data type, we just said it could encompass
`data type.
`Well, under claim construction, it could encompass data type
`and that should be included within their proposed construction. So, I
`think in terms of your specific question about indefiniteness and
`whether that's an issue that the Panel needs to address rather than a
`district court, I do think the answer is still yes because I think it
`doesn't make sense to adopt a construction here that's unnecessarily
`ambiguous and subjective when we have constructions that are
`consistent with the claims in the specification regarding what is access
`profile.
`And that is information that allows -- information we say reads
`or writes or some sort of estimation of what would be the expected
`reads or writes based upon a data type to allow for the selection of
`appropriate compressor.
`
`15
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`JUDGE BAER: Let me rephrase the question, then, a little
`bit. Are you aware of any case law that would suggest a construction
`that is less ambiguous, is preferred over a more -- over a more
`ambiguous construction in an IPR?
`MR. BATTS: I think the clear -- I guess I would say Nautilus.
`I think the clear direction from Nautilus is that you should be looking
`at having a claim scope that provides certainty to the public and to
`Petitioners to know the bounds.
`And what I see typically here is that even if there could be some
`question as to the bounds of the construction, the prior art and the
`teachings here are sufficiently within whatever construction is being
`proposed by Patent Owner, by Petitioner or by the separate panel that
`we would still prevail.
`JUDGE CHERRY: I mean, I guess it's my question.
`So, I mean, I'm looking at your summary -- I'm looking at Slide
`5 and, you know, from the summary of invention of Ishii, and it says,
`you know, selects -- you've highlighted, selects the file compression
`method suitable for the data type and access frequency.
`Why -- I guess I'm just not sure why we're fighting about access
`profile. I mean, is there -- is it that they say that you haven't
`proffered evidence under their construction to meet that or why
`wouldn't --
`MR. BATTS: I think that their -- sorry.
`JUDGE CHERRY: Go ahead. Go ahead.
`MR. BATTS: I think their likely argument, Your Honor, is
`that the petition didn't directly address data type rather than access
`frequency. Even though we did both and we clearly showed that
`Ishii does both, I think -- we'll hear from Realtime's counsel, but that's
`the argument I expect.
`
`16
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`And then I also would note that Ishii itself on Column 7, lines
`17 to 20, actually states that it takes into account the execution speed
`and the compression ratio looking at compression and decompression.
`So, I think that's also relevant teaching to Ishii that -- Ishii is
`teaching to look at to select the suitable compressor based upon data
`type and access frequency.
`But what I do fear, I guess, from Judge Baer's question, would
`be a scenario -- when you say, what's the cite about, I would be a little
`fearful about a final written decision that says, now, we can't
`determine the claim scope and we have to not make a determination
`about whether the prior art invalidates.
`So, I guess that's why I think it's worth giving you a few more
`shots at why I think the construction should not be adopted by Patent
`Owner. And if we go broader than the construction we originally
`proposed, we're okay with that. We just think there's additional
`reasons to invalidate the claims.
`JUDGE CHERRY: Okay.
`MR. BATTS: So, I think the next part that I was planning to
`move on to was the actual proposed grounds and how the proposed
`grounds invalidate the claims regardless, really, of the construction
`adopted.
`If we go to Slide 18, I think it -- a lot of the arguments from
`Realtime stem to be what the Petition actually included in terms of
`arguments or motivations to combine. And I think looking at the
`actual sections of the petition and the company expert declaration, it's
`clear that we included the necessary arguments and rationales to
`prevail here.
`So, what is the Petitioner arguing? Well, the Petitioner is
`arguing that we would use the combined teachings of Imai and Ishii.
`
`17
`
`
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`And if we go to Imai on Slide 20, I don't think there's really much of a
`dispute here that Imai is a system where you're looking at digital
`signals and selecting between multiple compressors to compress data;
`and that Imai teaches various factors to use to determine what the
`appropriate compressor should be, whether it's an asymmetric
`compressor or a symmetric compressor.
`And at the bottom of Slide 20, we've included some samples of
`the different factors that Imai chooses -- Imai includes for the
`selection criteria. And that's the processing capability of the client,
`the transition rate of the network, and the content of the digital
`signals.
`So, I think it's clear that Imai is teaching this selection process
`in selecting between compressor and using various criteria for the
`selection process.
`Then if we turn to Ishii on Slide 23, Ishii is teaching looking at
`the access frequency of data for our files for compression. And it
`teaches both looking at the type of data that you're going to be
`compressing, as well as the frequency for the determination of what
`compressor to apply.
`So, we think that's a very clear teaching, as we set forth in the
`Petition, that would be known to a person of skill in the art. And
`Ishii, in fact, teaches the motivation itself -- beyond the motivations
`that we provided in Dr. Storer's declaration in the Petition, Ishii itself
`teaches the benefits of using access frequency to choose between
`various compressors.
`So, if I go to Slide 7, I think this is a useful kind of side-by-side
`here to see what does the '535 patent say about what an access profile
`does versus what Ishii has to say about an access profile.
`
`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
`27
`28
`
`18
`
`
`
`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
`27
`28
`29
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`And what you see is that the Ishii reference is teaching -- on the
`bottom right there, it's talking about you're using the access frequency
`of a file, and the number of accesses for that file is basically given
`some sort of determination of relative -- or relative determination at
`low, medium or high.
`Similarly, what we see from the access profiles on Columns 11
`and 12 of the '535 patent, its relative amounts of frequencies that are
`being used to apply.
`Sometimes it seems like it could be based upon that type of data
`or it could be based upon the relative frequency for the particular data;
`but, in any event, either way it's showing using access frequency to
`determine the appropriate compressor to apply.
`Now, if we go to Slide 25, I think one of the arguments that you
`likely saw when reading the briefing here was Patent Owner,
`Realtime, was saying that we didn't provide any explanation of how or
`why Imai or Ishii would be combined -- the teachings could be
`combined.
`That was an argument I believe they made both in their
`preliminary response and then again we've seen post institution, but
`we actually provided, I think, across the Petitioner's, at least pages 18
`through 21, provided detailed explanations of how and why the
`references would be combined.
`And it was not simply that they were within the same field of
`art, it wasn't simply that they're both dealing with compression and
`choosing between compressors, but it also explained why the
`combined teachings would provide an improved system where you
`would be using the frequency of access teachings of Ishii to determine
`what would be the appropriate compressor for use within Imai's
`combined Figure 5 and Figure 16 system.
`
`19
`
`
`
`IPR2018-01169
`Patent 8,934,535 B2
`
`
`In Dr. Storer's declaration -- I put an example here, but I think
`Dr. Storer's declaration in paragraphs 120 through 122 provide quite a
`bit of detailed explanation for why a person of skill in the art would
`have wanted to use the access frequency teachings of Ishii within the
`combined teachings of the two systems.
`And I guess I would go back to -- I think it's kind of useful to
`go back to -- I'm a Petitioner's counsel, so I love KSR, right? But if
`we go back to KSR In re Mouttet, I'm just going to read off the quote:
`rather the test for obviousness is what the combined teachings of the
`references would have suggested to those ha