`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`DELL INC.; EMC CORPORATION; HEWLETT-PACKARD
`ENTERPRISE CO.; HP ENTERPRISE SERVICES, LLC; TERADATA
`OPERATIONS, INC.; and VERITAS TECHNOLOGIES, LLC,
`Petitioner,
`
`v.
`
`REALTIME DATA LLC,
`Patent Owner.
`____________
`
`Case IPR2017-00176 (Patent 7,161,506 C2)
`Case IPR2017-00179 (Patent 9,054,728 B2)
`Case IPR2017-00806 (Patent 7,161,506 C2)
`Case IPR2017-00808 (Patent 9,054,728 B2)
`____________
`
`Record of Oral Hearing
`Held: February 20, 2018
`____________
`
`
`
`
`
`Before JASON J. CHUNG, SCOTT C. MOORE, SHEILA F.
`McSHANE, and KAMRAN JIVANI, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`MICHAEL WOODS, ESQUIRE
`ANDREW R. SOMMER, ESQUIRE
`Winston & Strawn, LLP
`1700 K Street, N.W.
`Washington, D.C. 20006
`
`ON BEHALF OF PATENT OWNER:
`
`
`KAYVAN B. NOROOZI, ESQUIRE
`Noroozi, P.C.
`1299 Ocean Avenue
`Suite 450
`Santa Monica, California 90401
`
`
`
`The above-entitled matter came on for hearing on Tuesday,
`February 20, 2018, commencing at 2:38 p.m., at the U.S. Patent and
`Trademark Office, 600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
` 2
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`
`
`P R O C E E D I N G S
`- - - - -
`JUDGE CHUNG: We are convened for the consolidated
`hearing of IPR2017-00176, which has been consolidated with
`IPR2017-00806, and joined with IPR2017-01688. The judges for that
`panel consist of Judges Scott Moore and Sheila McShane and I. The
`other three cases in this consolidated hearing are IPR2017-00179, which
`has been consolidated with IPR2017-00808 and has been joined with
`IPR2017-01690. The panel for those three cases consists of Judges Scott
`Moore and Kamran Jivani and I. The presence of four judges here is not
`an indication of an expanded panel. The judges will be presiding on
`cases only as noted.
`Because three of my colleagues are joining remotely, we
`respectfully request counsel for each side to speak clearly into the
`microphone and identify the slide number that they are referring to. To
`the extent that petitioner wants to use the ELMO projection device,
`please keep in mind that there's a chance that the remote judges cannot
`see what you are referring to on the ELMO device and to please speak
`clearly and describe clearly what you are referencing in the ELMO
`device.
`
`As for objections, we want the free flow of conversation of each
`party, so to the extent that each party has an objection, please preserve
`them until it is their time to speak.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`
`
` 3
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`
`At this time, will petitioner please identify themselves?
`MR. SOMMER: Thank you, Your Honor. Andrew Sommer on
`behalf of petitioners.
`JUDGE CHUNG: Patent owner?
`MR. NOROOZI: Kayvan Noroozi on behalf of Realtime Data.
`JUDGE CHUNG: Thank you. Each party will have
`45 minutes to present their arguments. And because petitioner has the
`burden, petitioner will go first. Petitioner has the option of reserving
`some rebuttal time which will be taken away from -- which will be part
`of their 45 minutes total. At this time, does petitioner know how much
`rebuttal time they would like to reserve?
`MR. SOMMER: Yes, Your Honor, we would like to reserve
`18 minutes for rebuttal.
`JUDGE CHUNG: Thank you. Because there is no timer on the
`wall, I will let petitioner know when they have a few minutes left and I
`will also do the same for patent owner. Because petitioner reserved
`18 minutes of rebuttal time, I set the timer for 27 minutes. At this time,
`petitioner, you may begin.
`MR. SOMMER: Thank you, Your Honor. Can all the remote
`judges hear me?
`JUDGE JIVANI: Yes, please proceed.
`JUDGE McSHANE: Yes, we can.
`MR. SOMMER: May it please the Board, Andrew Sommer on
`behalf of petitioner. Today we are going to be discussing two patents,
`
`
`
`
` 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
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`the '506 and the '728 patent. And both share a very common
`specification. I think they are identical but for some of the formalities of
`cross-referencing the related applications. And there are some slight
`differences in the claims, some of which matter in these proceedings.
`And we'll talk about those in a bit.
`So here are the topics that I would like to cover today here in
`our argument. This is slide 2. And the first thing I would like to do is
`give a brief overview of the '506 and '728 patents. No doubt Your
`Honors are familiar with the disclosure and the claims of this particular
`patent. So I will be expeditious.
`Then, I will give an overview of Franaszek, Hsu and Sebastian
`and the relevant teachings of these references with respect to the
`invalidity arguments that are presented in the petition and explained
`further in the reply.
`Finally, I'll conclude about giving our explanation about why
`the evidence in this proceeding establishes that the challenged claims are
`obvious over the prior art.
`So turning to slide 3, we have here Figures 13A and 13B of the
`challenged patents, and it's identical between the two. What happens
`here is a data stream is received by this compression device, and after
`some buffering and some counting of the blocks, there's a decision made.
`It looks at the content of this data and it determines whether it recognizes
`it or not. And if the system recognizes the content of the data, it sends it
`into the branch called content-dependent encoders. And the idea in the
`
`
`
`
` 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
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`patent is that if you know the type of data, then you can better tailor the
`compression techniques to the data type, and so you have these
`content-independent encoders. And if you don't recognize the type of the
`data in the data stream, you feed it into what they call
`content-independent encoders. The claims variously refer to these as a
`default or single encoder in certain circumstances. We don't see the
`phrase "content-independent encoding" in the claims at issue in this
`proceeding.
`So the prior art, turning to slide 4, the primary prior art
`reference that we are relying on is a reference to Franaszek. And
`Franaszek teaches a system in which you look at the data stream, you
`determine if you can understand the type of the data in the stream and
`then you compress it using either a default compression technique or a
`compression technique that is tailored to the data type if it's available.
`Skipping to slide 6, we see this process flow. And this process
`flow it appears in a few places in the record in addition to Franaszek
`itself. This is Figure 4A. And here is an exemplary process flow of what
`is described in Franaszek. It looks to see if the data type is available.
`And in the context of the Franaszek reference, this is a data type that is
`written into a field in the block itself. It's labeled 205 in Figure 2 of the
`patent. And we can see that on slide 5. And if the type is not available
`for that particular block, what Franaszek will do is it sends it to a default
`set of compressors. It does not sample to determine which of the default
`is the best and then will compress that using a single or a default
`
`
`
`
` 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
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`compressor for that particular data block. If Franaszek sees that the data
`type for that block is available, then it will select a list of compression
`techniques that is tailored to that data block just like the patents-in-suit.
`And it will then compress that data block according to these tailored
`compression techniques.
`So the claims at issue here require looking to something other
`than a descriptor. And I don't think that the parties dispute that Franaszek
`is looking at a descriptor. In fact, that was a point that was made when
`these claims, at least, in the '506 patent were allowed in the first place.
`Franaszek is different because Franaszek is only looking at a descriptor
`to determine data type.
`So the question that we submit the Board needs to resolve in
`this proceeding is would it have been obvious to look at something other
`than solely a descriptor in the context of these claims such that the claims
`would have been obvious? And we submit that the evidence shows that
`it absolutely would have been. And the primary teaching that we rely on
`in this regard is Hsu. Hsu teaches a system that looks at something other
`than a descriptor to compress what they call heterogeneous files. And
`Hsu identifies this problem in the prior art. He says, well, sometimes
`files have diverse data types within them. And we could have audio and
`we could have images and text, and those are all going to be compressed
`using different compression algorithms that are optimized for those types.
`And so what we take is the string of data, the data stream, and we
`determine a compression plan for that stream by looking at each block
`
`
`
`
` 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
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`and saying which way or which way should I compress this particular
`data block? What compression technique should I use? And it
`determines metrics and types.
`Now, the metrics deal with the statistical properties of the data.
`The data type is based on an analysis of discrete portions of each block
`and it tries to categorize it into these predetermined sequences. So Hsu
`will then look at the contents of that data block to determine the data type
`and these metrics. And he does that because he recognizes that each type
`is slightly different. Each type is expressed in a different way. So I'm
`going to look at the contents of the data. And in case the contents of the
`data changes within a particular block, he uses a majority vote to try to
`determine what is the optimal compression for this particular block. And
`the results that he reports are an improvement over using a generic
`compressor for the entire file.
`JUDGE CHUNG: Counselor, can you point to us the testimony
`from Dr. Creusere that says that you would only use Hsu's teaching with
`the decision to -- when you recognize that type of block in your
`Franaszek reference and to not use it when you don't recognize it? So
`stated differently, could you point to us the best testimony from
`Dr. Creusere that says that you would only use Hsu for the if yes branch
`and not for the if no branch?
`MR. SOMMER: So we submit that that is the legal conclusion
`of obviousness in this case. And I'll answer your question directly. Dr.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`
`
` 8
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`Creusere does not say in any of his testimony, no, you shouldn't use Hsu
`when you go down the default branch.
`JUDGE CHUNG: So he makes no mention of that in the
`
`record?
`
`MR. SOMMER: Not that I'm aware of. Certainly not in his
`direct testimony. He does not say do not use Hsu to go down the default
`branch in Franaszek. Now, it's our position that that doesn't matter. And
`the reason that it doesn't matter is because there are numerous facts in the
`record that support that conclusion, the legal conclusion of obviousness
`as to whether the claimed subject matter would have been obvious based
`on the testimony of record.
`JUDGE CHUNG: Are any of those facts in the record
`supported by Dr. Creusere's testimony?
`MR. SOMMER: I'm sorry?
`JUDGE CHUNG: You are saying that there's numerous facts
`in the record that support your position that you would only use it for the
`if yes branch but you would omit the teachings of the Hsu reference for
`the if no branch. Are any of those facts in the record, are they supported
`by Dr. Creusere?
`MR. SOMMER: Well, the conclusion is not supported by
`Dr. Creusere's testimony directly, if that's your question, Your Honor.
`Does he come out and say you would not use Hsu going down the default
`branch --
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`
`
` 9
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`
`JUDGE MOORE: Counselor, one moment, please. It sounds
`like your microphone is turned off. I'm unable to hear you.
`JUDGE CHUNG: Off the record.
`(Discussion off the record.)
`JUDGE CHUNG: Let me repeat the question for the benefit of
`my fellow panel members. My question was, is there any testimony on
`the record that says that you would only use the teachings of Hsu for the
`if yes branch of Franaszek but you would omit the teachings of Hsu for
`the if no branch of Franaszek? Is there any testimony from Dr. Creusere?
`MR. SOMMER: On that particular issue he doesn't say that
`expressly. What we do submit is that there are facts of record --
`JUDGE CHUNG: You just said a few seconds ago that he did
`not say that, that he doesn't say it at all. I'm not sure if my fellow panel
`members heard that, but right now you are saying that he doesn't say it
`expressly?
`MR. SOMMER: He does not say in his direct testimony that it
`would have been obvious not to use Hsu going down the Franaszek
`default branch of Figure 4A. If we go back to slide 6 of our presentation,
`patent owner is saying that if you use Hsu in the context of Franaszek,
`you would insert it at block 401 and you would always flow to the right
`of this diagram.
`And we submit that the facts of record as opposed to the legal
`conclusion of what would have been obvious based on those facts shows
`that you absolutely would have very good reason not to use Hsu going
`
`
`
`
` 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
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`down that default branch. And the evidence includes that the number of
`types of data were dramatically increasing in the time period in question.
`There's evidence from Dr. Creusere in paragraphs 131 and 133 of his
`declaration discussing the profusion of new file formats and iterating
`through Hsu's collection of these known data types, these 10, could slow
`the program down. Dr. Creusere explains how pattern matching in the
`context of Unix and the magic file command slowed programs down.
`This is found in 131 of his declaration. It's found in Exhibit 1028,
`page 185, and it's found in Exhibit 1029, page 5 talking about exact
`pattern matching used to identify block types.
`Now, so why wouldn't you use this given these facts of record
`going down the default branch? Well, if you take patent owner at what
`they say, and we don't materially disagree with this, Hsu is entirely
`content-dependent. We don't have to disagree with that point. If, as
`Dr. Zeger says, one of ordinary skill in the art reading Hsu would always
`identify one of those 10 data types regardless of the content of the block,
`then what would happen is you would get the wrong type. You would
`inject inefficiencies and you would basically be tossing aside Franaszek's
`teaching that the default branch should be used when you don't know the
`data type.
`And so if you look at Dr. Zeger's testimony on this particular
`issue -- and I think in the '506 proceeding it's Exhibit 1035. In the '728
`patent IPRs, it's Exhibit 1031. At pages 54 and 55 he talks about how the
`Franaszek system is likely operating based on a file extension that that is
`
`
`
`
` 11
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`his reasonable conclusion based on what he understands Franaszek to be
`teaching. He says in this case, you definitely go down the left branch if
`you are looking at Franaszek. Franaszek doesn't know the type, so you
`use the default encoder.
`Now, Dr. Creusere says well, of course, if you did know the
`type, then you could use Hsu as a shortcut. You could short-circuit this
`and get a better identification of the data type when you do know the type
`of the overall file. And Dr. Creusere says that could provide a shortcut.
`That's in paragraph 132 of his declaration. And this could be the most
`efficient and least time-intensive process in the context of using
`Franaszek and Hsu.
`So what does the record evidence show? It shows that
`Franaszek discloses content-dependent and content-independent. That's
`definitely undisputed. It also says Hsu discloses a way of using a
`content-dependent encoder. And what patent owner is saying is that if
`you were to put Hsu into Franaszek, something that we don't think is
`actually a legally appropriate analysis, then when you were to do that,
`you would always guess the data type and you would always proceed
`down the right-hand side of the flowchart here in Franaszek. And we
`submit that that's legally erroneous. That's bodily incorporation, what
`happens when you put one reference into another. And that's what
`Dr. Creusere was testifying about in the context of the deposition
`testimony that they have cited extensively at page 131 of his deposition.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`
`
` 12
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`
`JUDGE MOORE: Counselor, I would like to drill down a little
`bit on this subject. So I would like to take you through your obviousness
`theory a little bit and just make sure I'm understanding it correctly.
`So you concede that Franaszek does not calculate parameters or
`attributes of the type required by claims 1 and 24, correct?
`MR. SOMMER: I don't know that that is something we
`concede because Franaszek does determine a data type of a data block
`when it's available.
`JUDGE MOORE: Right. But that's not a parameter attribute of
`the type required by claims 1 and 24. Isn't that just -- those claims both
`exclude data, analyzing data based solely on a descriptor that's indicative
`of one or more parameters or attributes, and that's the whole basis of you
`looking to Hsu.
`MR. SOMMER: Absolutely. When you get to that point of the
`claim that it's got to exclude solely a descriptor, yes, you are correct,
`Franaszek does not disclose that.
`JUDGE MOORE: And Hsu discloses calculating four different
`parameters or attributes for a given data block; is that correct?
`MR. SOMMER: Yes.
`JUDGE MOORE: Data type and three redundancy matrices,
`
`correct?
`
`MR. SOMMER: That is correct.
`JUDGE MOORE: So your expert testifies that it would have
`been useful to calculate these four different parameters or attributes
`
`
`
`
` 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
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`regardless of whether or not Franaszek's data type information was
`present.
`MR. SOMMER: I don't believe that that is something that our
`expert testified to.
`JUDGE MOORE: Well, perhaps I misstated. He said even if --
`what your expert testified to, I believe, my apologies if I misstated it, is
`that it would be useful to calculate these four parameters or attributes
`even if Franaszek's data type is present.
`MR. SOMMER: That's a consistent or accurate way to
`characterize his direct testimony, yes.
`JUDGE MOORE: And then the argument you are trying to
`make now, which your expert's testimony does not directly support, is
`that a person of ordinary skill in the art would have refrained from
`calculating Hsu's parameters or attributes in circumstances when
`Franaszek's data type is not present, correct?
`MR. SOMMER: That is our position, yes. That's the legal
`conclusion that we are asking the Board to reach.
`JUDGE MOORE: And the evidence and argument you have
`presented, as I understand it, is that it would have been complicated to
`calculate Franaszek's type, data type information because of the profusion
`of file types, correct?
`MR. SOMMER: That's not quite the argument that we are
`presenting. The argument is that if it's not on Hsu's list, even if it were to
`implement the process of Hsu, it would not get the correct data type and
`
`
`
`
` 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
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`therefore, these concepts of tailoring your compression type to the data
`type wouldn't be achieved. You are just guessing at that point.
`JUDGE MOORE: You are estimating, right? And an estimate
`is not correct a hundred percent of the time, so I'm not sure it's correct to
`say that you would not get the correct data type. Let's say for a moment
`that I accept your argument as credible that a person of ordinary skill in
`the art would have realized that calculating Hsu's data type is
`complicated and time consuming when you don't have a shortcut in the
`form of Franaszek's data type. So assume that I accept that. What about
`the three redundancy matrices? Why wouldn't you calculate those?
`MR. SOMMER: Well, those don't tell you anything apart from
`data type in the context of Hsu. And you can see that here on our slide 8.
`JUDGE MOORE: Why would you calculate them at all if they
`didn't tell you any additional information that's useful for compression?
`MR. SOMMER: I'm sorry, I guess I don't understand the
`question. I don't think you would calculate them at all if you are not able
`to determine a data type.
`JUDGE MOORE: What is your support for that?
`MR. SOMMER: Well, the way that Hsu operates is it uses this
`table that we have shown here on slide 8. It's table 1 in the Hsu
`reference, Exhibit 1005. In this case, if you don't have the data type, you
`can't index into the table and figure out which compression type to use in
`the first place. I'm sorry. If I misunderstood Your Honor's question, I do
`apologize.
`
`
`
`
` 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
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`
`JUDGE MOORE: No, I think you are getting to the point.
`Please continue. So your theory is that because of the algorithm that Hsu
`uses, that you would not be able to use the redundancy matrices in the
`absence of a calculated data type?
`MR. SOMMER: Right. If you don't have a data type, you can't
`index into this table and select any of the encoding types without that. So
`what Hsu is doing is it calculates these redundancy metrics and selects
`the highest one, I think it says, assuming it's over 2.5. And if it's over
`2.5, then you know you are going to get some degree of valuable data
`compression and we're going to use that. So pick the highest one of these
`redundancy metrics, but in order to select the encoding type, you have to
`know the data type within Hsu. Otherwise, you can't index into the table
`and access the data properly to identify that compression technique.
`JUDGE MOORE: Thank you. I think I understand that better.
`So then I think the whole issue in this case, at least the cases involving
`the '728 patent, the focus is whether or not you have offered adequate
`support for your position that a person of ordinary skill in the art would
`have refrained from calculating Hsu's four parameters or attributes in the
`situation where Franaszek's type information isn't present. I would like
`to take you through the evidence here a little bit and give you a chance to
`express it, because my first reaction is it seems rather odd that you would
`want to calculate type information for a data block in circumstances
`where type information was already present and that you would not want
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`
`
` 16
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`to calculate it in circumstances where it wasn't present. That seems
`backwards.
`JUDGE CHUNG: And to piggy-back on what Judge Moore is
`saying, my concern is, and I'm looking at the 176 IPR, for instance,
`paragraph 65, it seems that Dr. Creusere comes very close to saying why
`you would use it for the if yes branch but not the if no branch, but then
`Dr. Creusere states, Nevertheless, as I discuss below, I do not believe
`such information would be material to my opinions regarding the
`obviousness of the subject matter of the claims I have analyzed.
`So it doesn't seem -- my concern is it doesn't seem like
`Dr. Creusere has really drilled down on this point why you would use it
`for the if yes branch and then all of a sudden you would omit Hsu's
`teaching from the if no branch.
`MR. SOMMER: So if I understood your question, you want to
`understand kind of the lay of the land on the evidence in the record and --
`JUDGE CHUNG: Yes. I was just talking out loud.
`MR. SOMMER: I just want to make sure that I have got the
`questions on the table here to be able to respond fully. So Dr. Creusere
`says in paragraph 65 of the declaration in the '506 proceeding, look, it
`doesn't really matter. There was a fight in a related IPR over the
`teachings of Hsu. And the question there was does this Unix file
`command return data for unknown data types or doesn't it? Does it
`always try to cabin it into one of the 10 things on our list here in slide 8
`or does it return an unknown? And Dr. Creusere said, look, Unix did
`
`
`
`
` 17
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`return unknown data types when it was unable to properly classify things,
`but it doesn't matter to my analysis, as Your Honor pointed out. And the
`reason that he does say that in the '506 proceeding is it simply doesn't
`matter to the claim. The claim is met by performing either one of the
`steps required by the "if" statement or if/if not, right, required by that
`particular claim. And that's under the Board's precedent in Schulhauser.
`JUDGE CHUNG: So let's touch on that for a moment. I will
`note that you made that argument in the 176 case and the consolidated
`and joined case for that. But you did not make it in the 179 case, which
`-- and my guess for that omission is because those claims are directed to
`a system claim. Whereas, in the 176 and all the joined and consolidated
`proceedings with the 176, those recite a method claim.
`Now, pertaining to that Ex Parte Schulhauser issue without
`going too much on a tangent here, was Ex Parte Schulhauser cited in the
`petition?
`MR. SOMMER: I think the record will establish it was not
`cited in the petition.
`JUDGE CHUNG: Okay. And what is the reason for bringing it
`up -- well, why bring it up at the reply if it, I guess -- is it petitioner's
`position that patent owner opened up the door for petitioner to bring up
`that argument at that juncture?
`MR. SOMMER: Absolutely. They said clearly in their patent
`owner response, look, we lose because if you incorporate Hsu into
`Franaszek, which we think is not the right legal analysis, but setting that
`
`
`
`
` 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
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`aside, if you incorporate it in, then you are always going to go down
`step 1C of the claim and you are never going to hit, I think in this
`particular -- or it's 104C as opposed to 104F, I think. I'm sorry. You'll
`have to excuse me.
`JUDGE CHUNG: Why shouldn't that be brought up at the
`petition? It seems kind of tenuous to me that patent owner's argument
`would open the door to your argument of citing Ex Parte Schulhauser in
`addition to citing Schulhauser or Cybersettle, Incorporated versus
`National Arbitration Forum, Incorporated, and that's a Federal Circuit
`case that you cited on page 22 of your reply in the 176 IPR. Was that
`Cybersettle versus National Arbitration Forum also cited in the petition
`or is this the first time it's being cited?
`MR. SOMMER: It was cited in response to the patent owner's
`response, Your Honor. They said we lose because if you do insert Hsu
`into Franaszek, you always proceed down the one fork of the decision of
`Franaszek. And as we explained in our reply, look, that's not even the
`right legal analysis. That's not what the obviousness inquiry is designed
`to do. It's not what you are to consider. It's what do the teachings of
`these references as a whole suggest to the person of ordinary skill in the
`art and what ultimately as a legal question would have been obvious to
`that person based on the facts of the record. And if that answers Your
`Honor's question, I would like to get back to Judge Moore's question.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`
`
` 19
`
`
`
`Case IPR2017-00176 (Patent 7,161,506 C2)1
`Case IPR2017-00179 (Patent 9,054,728 B2)2
`Case IPR2017-00806 (Patent 7,161,506 C2)1
`Case IPR2017-00808 (Patent 9,054,728 B2)2
`
`
`JUDGE CHUNG: We could touch on that more during your
`rebuttal time. You have 2 minutes and 14 seconds left. You can use that
`remaining time to address Judge Moore's question.
`MR. SOMMER: Thank you very much, Your Honor. So
`Judge Moore, in terms of the evidence of record, this is laid out in our
`reply and in the petition at pages 40 to 41, in the reply, we explain how
`the complete record, developed record shows and establishes the
`inefficiencies of using Hsu in the context of using Hsu to avoid default
`data compression. Franaszek has a system whereby if you know the data
`type and you have these encoders tailored to that data type, you compress
`using those tailored encoders. And if you don't, then you use this default
`set. And the evidence of record including --
`JUDGE MOORE: You test. You run tests and then you try
`and choose a default. You run tests on multiple ones an
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
