`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`CELLCO PARTNERSHIP d/b/a VERIZON WIRELESS,
`Petitioner,
`
`v.
`
`BRIDGE AND POST, INC.,
`Patent Owner.
`____________
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`___________
`
`Record of Oral Hearing
`Held: January 17, 2019
`____________
`
`
`
`
`Before JONI Y. CHANG, BARBARA A. PARVIS, and KEVIN C.
`TROCK, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`
`
`
`APPEARANCES:
`ON BEHALF OF THE PETITIONER:
`JAY I. ALEXANDER, ESQUIRE
`PETER P. CHEN, ESQUIRE
`Covington & Burling, LLP
`One City Center
`850 Tenth Street, N.W.
`Washington, D.C. 20001-4956
`
`
`ON BEHALF OF PATENT OWNER:
`LAUREN N. ROBINSON, ESQUIRE
`DENISE M. De MORY, ESQUIRE
`Bunsow, De Mory, Smith & Allison, LLP
`701 El Camino Real
`Redwood City, California 94063
`
`
`The above-entitled matter came on for hearing on Thursday,
`
`
`January 17, 2019, commencing at 1:05 p.m., at the U.S. Patent and
`Trademark Office, 600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
` 2
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`
`
`
`
`P R O C E E D I N G S
`- - - - -
`JUDGE CHANG: Good afternoon. I'm administrative patent
`judge Joni Chang. Here with me is Judge Barbara Parvis. Judge Kevin
`Trock is joining us remotely from San Jose, California. And please
`introduce yourself at this time, starting with the petitioner.
`MR. ALEXANDER: Good afternoon, Your Honor. Jay
`Alexander, counsel for the petitioner. With me today is my co-counsel,
`Mr. Peter Chen.
`MS. ROBINSON: Lauren Robinson, counsel for patent owner,
`Bridge and Post. With me today is my partner, Denise De Mory. And also
`with us is Nitin Shah, the named inventor of the '747 patent and founder of
`Feeva, who invented the technology.
`JUDGE CHANG: Thank you so much and welcome. This is a
`consolidated oral hearing for IPR2018-00054 and 55 involving patent
`8,862,747. This hearing is open to the public. The transcript will be entered
`in both cases and is usable across both cases. Please note that the
`demonstratives, exhibits, are neither evidence nor substantive briefs. Rather,
`they are merely visual aids. We did enter into our files, and there was no
`objections from either party. May I ask, did you provide a hard copy to the
`court reporter?
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`
`
`
`
`
` 3
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`MR. ALEXANDER: We did, Your Honor. Would either you or
`Judge Parvis like an extra hard copy today?
`JUDGE CHANG: I would like to, thank you.
`JUDGE PARVIS: I would like one as well.
`JUDGE CHANG: Because Judge Trock is participating remotely,
`I just want to remind counsel that please speak only at the podium. And also
`for clarity, when you present, please speak clearly and also refer to the slide
`number.
`Also, consistent with our prior order, each party has a total of
`60 minutes for both cases to present their argument. And petitioner will
`proceed first to present its case as to the challenged claims in both cases, and
`thereafter, patent owner will respond to the petitioner's case. Petitioner may
`reserve a small portion of time for rebuttal.
`Is there any questions at this time? No, okay. Counsel for
`petitioner, you may start whenever you are ready.
`MR. ALEXANDER: Thank you, Your Honors. Although the
`patent at issue and the claims at issue have many limitations, the parties have
`narrowed the issues to a relatively few number, and we have --
`JUDGE CHANG: Sorry to interrupt. Would you like to reserve a
`rebuttal time?
`MR. ALEXANDER: I'm sorry. Yes, we would like to reserve
`15 minutes, if we could.
`
`
`
`
`
`
`
`
` 4
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`JUDGE CHANG: Okay. Let me start the timer for you. Hold on.
`Okay, you may begin.
`MR. ALEXANDER: Thank you. We have listed the issues in
`dispute in slide 3. I'm Jay Alexander. I will be speaking toward the issues
`specific to the combination of Harada, Roker and Brijesh in ground 1. My
`co-counsel, Mr. Chen, will speak to the issues regarding the priority date of
`the '747 patent and the ground 2 based on Candelore.
`JUDGE CHANG: Before you begin, so you asserted that one of
`the prior art is prior art because the patent is not entitled to the priority date.
`But also in the petition, you asserted the Brijesh reference as prior art under
`102(a) or (e). Now, are you abandoning that argument?
`MR. ALEXANDER: Not at all, Your Honor.
`JUDGE CHANG: Because I don't see that in here.
`MR. ALEXANDER: Yeah, we did not list it in the slides.
`Mr. Chen is going to speak to that in more detail. But we do contend Brijesh
`is prior art under 102(a), (b) and (e). We think because the patent should not
`get priority date, it's 102(b). But even if the Board were to decide otherwise,
`it's still 102(a) prior art and because the Brijesh and the '747 patent are
`different inventive entities and therefore, the invention was by another
`before the provisional date.
`JUDGE CHANG: I just don't see it in your slide or the
`presentation. So I just wondered.
`
`
`
`
`
`
`
`
` 5
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`MR. ALEXANDER: We do intend to preserve that. And
`Mr. Chen will speak to that.
`In terms of the disputes regarding ground 1, the first dispute is over
`what the parties have called the device identifier element which is --
`essentially the argument is that a POSA, a person of skill in the art, would
`have been led away from using the MAC address as an identifier because of
`Harada's so-called very heavy reliance on personal information to identify
`the user. We disagree with that because Harada's reliance on personal
`information is not so pronounced. In fact, we think the critical passage --
`and I'm going to turn to slide 8 now. The critical passage from Harada says
`that the user profile may be selected from the database based on identifying
`information associated with a particular computer or user of that computer.
`So Harada is really contemplating using either type of information. If the
`information is relating to the particular computer, it would, by definition, be
`nonpersonal information as opposed to information related to the user.
`Now, Harada goes on to give an example where actually both types
`of information are being used, both the user name and a TCP/IP network
`address. However, the fact that the network address is being used indicates
`that Harada is not so reliant on use of personal information as the patent
`owner has argued. And in fact, the expert for patent owner, Mr. Smoot,
`admitted that the network connection information used by Harada is, in fact,
`not personal information.
`
`
`
`
`
`
`
`
` 6
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`But more importantly, this argument is an attack on Harada
`individually. Whereas, the petition is relying on a combination of the
`teachings of both Harada as well as Roker and Brijesh. So it's not
`appropriate to sort of pick out the individual deficiencies of one reference.
`The Board really should consider the combination as a whole.
`And in particular, we rely on two major teachings from Roker.
`Number one is that the identifier really can be arbitrary as long as it's static
`and unique. And the passage we rely most heavily on is reproduced at slide
`11. It talks about a static IP address. It talks about some billing or
`accounting code and basically states that it's really arbitrary as long as it's as
`static and unique.
`The second teaching that we rely on from Roker is the fact that the
`service provider is a trusted keeper of the user's personal information, and
`therefore, any information about the user that leaves the provider's network
`should be deidentified or anonymized before leaving the network. The
`passage we rely most heavily on is reproduced at slide 12 from Roker.
`And then the teaching from Brijesh, which is also a very similar
`system -- I should mention that Harada, Roker and Brijesh are all extremely
`similar in that they are addressing the problem of being able to target content
`back to a user by sending some form of identification of who the user is or
`that user's computer out to the destination server so that the destination
`server can choose content to deliver back.
`
`
`
`
`
`
`
`
` 7
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`All of them recognize the shortcomings of cookies that had been
`used previously for that purpose, and so all of these references are tagging
`the outbound message with some indication of identity. Brijesh specifically
`talks about making use of the device's MAC address as that identifier, and
`we've reproduced two of the key passages on slide 13 from Brijesh.
`So really the overall combination or the notion that a POSA would
`have been motivated to use a MAC address to identify the sender of the
`outgoing message, the outgoing request for some resource from the server is,
`you know, make use of the MAC address because it fills Roker's teaching of
`having the identifier be anonymous, static and unique, and therefore,
`basically checks all of the boxes that Roker tells you that you would want.
`And our expert, Mr. Gray, went on at length about the reasons why
`the person of ordinary skill in the art would combine those teachings
`together to arrive at the claimed invention, including the device identifier
`limitations. And we've taken an excerpt from that testimony on slide 14
`from paragraph 85 of Mr. Gray's declaration. He actually addresses it in
`additional text besides this, but this is sort of the key piece of the testimony
`that is on slide 14.
`By the way, I should mention that Mr. Gray's testimony is
`unchallenged. Patent owner did not elect to take his deposition. So there's
`no contrary testimony from Mr. Gray. All we have is patent owner's expert's
`rebuttal.
`
`
`
`
`
`
`
`
` 8
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`There is a second argument that patent owner makes for why
`Harada relies heavily on personal information, and that's Harada's use of this
`IDENT protocol. However, as we pointed out, the IDENT disclosure in
`Harada is simply an alternative embodiment that we don't rely upon in the
`petition. Even patent owner's expert admitted that. We've got that excerpt
`of his testimony on slide 16.
`But importantly, what we rely on is Harada's teaching that you can
`identify the sender of this message as either the particular computer or the
`user. Roker teaches you want it to be some arbitrary identifier. And Brijesh
`has the MAC addresses. So that's the combination that the petition proposes.
`Patent owner makes another argument which is essentially another
`attack on Harada individually. They say that Harada's proxy server would
`not have had access to the MAC address of the sender, and therefore, the
`person of skill in the art would be led away from modifying Harada to use
`the MAC address.
`This also ignores the fact that the petition relies on a combination
`of teachings in the prior art. And in particular, as we pointed out in slide 17,
`the petition basically says that a person of skill in the art would have
`architected this combined system in one of two ways: The POSA could have
`just used Roker's architecture itself or it could have architected the system to
`have a proxy server similar to Harada which performs the tagging function
`located at the edge of the service provider's network facing the client. And
`therefore, in either instance, the tagging device, whether it's Roker's network
`
`
`
`
`
`
`
`
` 9
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`device 70 or a proxy server such as Harada within the edge of the network
`provider system, in either case they would have had access to the MAC
`address. In fact, Mr. Smoot, the patent owner's expert, admitted that in
`Roker's architecture, the MAC address would have been available to the
`network device 70 in Roker which does the tagging. And that admission is
`reproduced on slide 18.
`In addition, we've got a couple slides talking about what Roker
`teaches about his architecture. In fact, Roker notes that really the tagging
`device and the router could be integral. They could be one and the same
`hardware device. And that passage is reproduced on slide 19. And the fact
`that the network device and the router is located on the edge of the service
`provider's network is disclosed in the passage that's reproduced at slide 20.
`So what that leads to, the point that leads to is the fact that that
`architecture in Roker is virtually indistinguishable from the architecture
`that's laid out in the '747 patent. If you look at Figure 2 of the '747 patent,
`which we've reproduced at slide 21, the '747 also teaches an architecture
`where the tagging device and the router can be co-located on the same side
`of the network as the client and therefore, have access to the MAC address
`and be able to use the MAC address. Roker is the same, and we've
`illustrated that graphically on slide 22.
`Now, in addition, the second way that this combination could have
`been architected by the POSA would be to include something like a proxy
`server co-located with the router in a system such as Roker's. And the fact
`
`
`
`
`
`
`
`
` 10
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`that proxy servers could be co-located with routers is actually shown in the
`very exhibit that patent owner's expert cites in his reply. And that's shown
`on slide 23. So a POSA would have a lot of flexibility in architecting the
`system once you accept the fact that the POSA is motivated to use the MAC
`address as the device identifier in order to tag the message.
`So if there are no questions on the device identifier limitation, I'll
`move on to the second area of dispute, and that is whether or not the request
`identifier needs to be in a single string in a single field. Patent owner has
`made the argument that the claim requires that the request identifier be
`composed of a single alphanumeric string embedded in a single extensible
`field of the outgoing HTTP message. That's really a claim construction they
`are making, but they don't identify it as such in their papers. That's what
`they are saying. They are saying that Harada, which uses two fields, which
`is shown, for example, on slide 27, part of the information that's outgoing is
`in an unencrypted field. The other part is in an encrypted field, 304 and 305.
`They are saying that cannot meet the claim limitation because it's more than
`one field.
`We disagree with that as a matter of claim construction because it's
`black-letter law. And we've reproduced the 01 Communique case on slide
`28 that when a claim uses the word "a" or "an", it means one or more. And
`in this case, we'll go back to slide 26, the alphanumeric string is identified in
`the claim as an alphanumeric string, as is the extensible field. So this claim
`would permit that the request identifier being embedded in one or more
`
`
`
`
`
`
`
`
` 11
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`alphanumeric strings in one or more extensible fields. The only exception to
`that would be if there was some clear indication in the claim language or in
`the intrinsic evidence that would lead to a different construction.
`And I should mention that we are still under the BRI standard in
`this case, and so the question is whether our construction is a reasonable one.
`And we think it is because patent owner has identified no -- nothing in the
`claim language or anything in the intrinsic evidence that would absolutely
`limit the alphanumeric string or the extensible field to a single one.
`JUDGE CHANG: Counsel, so does it matter if we apply Phillips
`or BRI?
`MR. ALEXANDER: We would contend it does not.
`JUDGE CHANG: Because the citations to the case law that you
`have, are those BRI cases or Phillips cases?
`MR. ALEXANDER: Those are Phillips cases. 01 Communique
`was a case arising out of a District Court case. But yeah, we don't think it
`makes a difference. If anything, it's a little bit stronger for us under BRI, but
`we should prevail either way.
`There's nothing in the claim language that patent owner has
`identified that requires a single string. I mean, certainly patent owner could
`have used those words. They could have said encrypt and embed the request
`identifier in a single alphanumeric string and a single extensible field, but
`the claim doesn't say that.
`
`
`
`
`
`
`
`
` 12
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`And the intrinsic evidence indicates that there's nothing really
`critical about there being a single string or a single field. There's absolutely
`no indication that that is a critical feature of this alleged invention. In fact,
`to the contrary, the specification describes Figure 5 which shows the
`make-up of the header as an example, exemplary HTTP header. And we've
`highlighted that passage at slide 29. And it also indicates that there's some
`flexibility in both the length and the position of this request ID tag within the
`field. So there's at least far from being -- having it be critical that it's a
`single field in a set position, there seems to be an indication of quite some
`flexibility.
`JUDGE CHANG: How about the prosecution history? Is there
`anything in the prosecution history?
`MR. ALEXANDER: Neither side cited anything in the
`prosecution history. We didn't see anything that would indicate that the
`Board should depart from the Federal Circuit's rule. And patent owner has
`not cited any.
`And furthermore, we think the Board should rule for us on the
`claim construction issue, but even if not, we think that the petition makes
`enough of a case of obviousness anyway. Roker teaches that all of the
`information should be encrypted to protect the privacy of the user. And
`we've reproduced that teaching at slide 31. And our expert, Mr. Gray, again,
`unchallenged, not deposed, has said that if one were to encrypt all of the
`information, the most logical and natural thing to do would be to put the
`
`
`
`
`
`
`
`
` 13
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`result, the resulted encrypted information in a single field, because that
`would be basically the easiest thing for a POSA to do. And we've produced
`the key testimony at slide 32, which is paragraph 102 of his opening
`declaration, and slide 33, paragraph 9 of the reply declaration.
`JUDGE CHANG: Why did Harada put it in two different headers?
`MR. ALEXANDER: I'm sorry?
`JUDGE CHANG: Why did the primary reference, Harada, put it
`in two different fields?
`MR. ALEXANDER: Harada doesn't say why. One could surmise
`that because some of the information was unencrypted and some was
`encrypted, that led to using two different fields. But if one were to accept
`the teaching of Roker that you want to encrypt all of this information, then it
`would be logical just to use a single field for all of the encrypted
`information.
`And also, I should add, this is not in our slide, but Mr. Smoot, the
`patent owner's expert, conceded that combining all of the request identifier
`information into a single field would be well within the level of skill in the
`art. And that testimony is at Exhibit 1036, page 41:20 to page 43:5. And it's
`argued in our reply at page 8.
`JUDGE CHANG: Did you argue that in your petition?
`MR. ALEXANDER: Well, no, we didn't address -- well, no. I
`should say we did. If you go back to paragraph 102 of Exhibit 1008 on slide
`32, we talked about -- Mr. Gray talks about having motivation to encrypt the
`
`
`
`
`
`
`
`
` 14
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`information in field 304 and talks about why. He gives three reasons. There
`was not a specific piece of testimony about putting it in a single field in the
`petition because under our view and we thought the natural reading of the
`claim, and we were actually quite surprised it was disputed, is that if the
`information that was in more and more fields, that would beat the claim
`limitations. So there wasn't anything specific about putting it into a single
`field in the petition. But once patent owner raised this argument that the
`claim could be construed to require a single field, we showed in reply that
`that would be obvious.
`If there are no further questions on that issue, I'll move to the third
`issue, which is the more general teaching away argument that patent owner
`makes. Essentially what they have said is that there are different functions
`and different levels of complexity if you consider both Harada's proxy server
`and then Roker's system such that the person of skill in the art would be
`taught away from the combination. We mentioned significant complexity,
`added cost, for example, in Harada's system. And they also mention the fact
`that Roker performs some additional optimizing of the message, and they
`say that that's something that would somehow lead away from the
`combination.
`It's black-letter law that to make a teaching away argument
`successfully, the patent owner has to show some criticism, discrediting or
`discouragement of the combination in the prior art. And we've cited the
`Meiresonne case on slide 37 as an example of that law. In addition, even if
`
`
`
`
`
`
`
`
` 15
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`the alleged combination is somehow inferior, that still wouldn't negate
`obviousness. And we cite the In re Mouttet case on slide 38 for that
`proposition.
`The record here is completely devoid of any criticism,
`discouragement or discrediting of putting the teachings together in the way
`that the petitioner proposes. In fact, as I mentioned before and as we have
`summarized on slide 39, both Harada and Roker are very similar systems.
`They essentially do the same thing in terms of tagging the outbound
`message. They both recognize the shortcomings of cookies. They tag the
`outbound request to identify the requester. They teach encryption of certain
`of the information. And they teach the tagging device located between the
`client and the server to intercept the message, place the tag on it and then
`send it out to the server.
`JUDGE CHANG: On slide 39, you say both teach encrypting
`certain personal information. So here the claim requires nonpersonal
`information. So if somehow you modify Harada to only encrypting
`nonpersonal information, would it make the invention inoperable?
`MR. ALEXANDER: No. The fact that they both teach encrypting
`personal information really goes to the fact that the person of skill in the art
`would understand that it's important to protect personal information. So
`therefore, really the preferred course would be to use a nonpersonal
`identifier. In fact, Roker teaches that directly. Roker teaches encrypting the
`static IP address or teaches using the static IP address or some other arbitrary
`
`
`
`
`
`
`
`
` 16
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`code to identify the user. So the fact that they are both teaching encryption
`really goes to the notion that the POSA understands that it's important to
`anonymize or hide in whatever ways are available any personal information
`about the user. So that supports our combination, in our view.
`JUDGE PARVIS: With respect to the performance improvement,
`is there testimony by patent owner's expert that there would be -- that Roker,
`the device, the graphing device handles higher traffic and so there would be
`issues with combining Roker and Harada?
`MR. ALEXANDER: Yes. That goes to -- Roker has some
`different functions. For example, it does this enhancement. They call it
`preemption and module events and other things that go to optimizing the
`traffic in some other way. But that doesn't detract from Roker's teaching of
`tagging the outgoing message with information that's anonymous about who
`the sender is.
`JUDGE PARVIS: Mr. Gray testifies that it's a performance
`improvement as one of the reasons to combine the two; is that correct?
`MR. ALEXANDER: Yes, he does. But that doesn't go to this
`issue, though. These optional functions that Roker can do, it doesn't have to
`do. It doesn't have anything to do with the performance improvement. The
`performance improvement goes to co-locating whatever device is doing the
`tagging with the router so that they are in close proximity. That reduces
`latency which is essentially reducing signal delay between the two devices,
`so that would lead to some increased performance. That really goes to the
`
`
`
`
`
`
`
`
` 17
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`architecture point that I made earlier. And it's also included in the general
`teaching away argument where they again raise the architecture issue. Did
`that answer your question?
`JUDGE PARVIS: The only -- with respect to teaching away, it's
`not just teaching away. I just want to make sure that the reasoning for why
`one of ordinary skill in the art, the question pertains to that. Does the
`testimony of Mr. Smoot undermine in any way Mr. Gray's testimony, in
`petitioner's view, with respect to the performance? Even if it's not teaching
`away, is it undermining that testimony?
`MR. ALEXANDER: No, we don't think so because the
`performance improvement is gained by co-locating the router and the
`tagging device. What Mr. Smoot is talking about is, well, in Roker you
`could have a lot of additional functions in addition to this tagging function
`being performed by this device, this same device that does the tagging. In
`our view, those are just optional alternative functions. They don't have to be
`there, and we certainly don't rely on them being there. But there's nothing in
`having those options available that in any way detracts from Mr. Gray's
`point, which is that you can co-locate the devices to reduce latency.
`JUDGE PARVIS: Thank you.
`MR. ALEXANDER: If the Board has no other questions, I'll turn
`over the remainder of the time to Mr. Chen.
`MR. CHEN: Thank you. And Your Honors, I'll be speaking to
`two points. One is the status of the Brijesh reference as prior art, and two,
`
`
`
`
`
`
`
`
` 18
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`the Candelore reference which is part of our second ground in combination
`with Harada and Roker.
`So let me advance the slides. So first on the topic of Brijesh, I
`appreciate Your Honor's observation at the outset, and certainly, as my
`colleague, Mr. Alexander, noted, petitioner absolutely continues to maintain
`that Brijesh qualifies as a prior art reference under multiple subsections of
`Section 102. As the trial was instituted and as evidence and argument were
`compiled, a lot of the focus happened to be on the Section 102(b) issue.
`JUDGE CHANG: I'm sorry to interrupt, but we would like to hear
`why is it prior art under 102(a).
`MR. CHEN: Our position under Section 102(a) is that there's been
`no evidence or argument from the patent owner regarding the date of
`invention of the '747. And so even if the earliest possible date of invention,
`which we believe is the date of application that would be a March 2007 date.
`That's the earliest possible date that would exist for the '747. That's the date
`that the provisional was filed. However, the Brijesh reference was published
`in November of 2006, and there's been no effort to antedate Brijesh with
`evidence of prior invention, prior reduction to practice.
`JUDGE CHANG: Is there any argument or evidence to show that
`the subject matter that's disclosed in Brijesh is not by another?
`MR. CHEN: I don't believe that the parties focused on that, but
`certainly our position is that there are multiple inventors here on the
`provisional and on the '747, and that we believe that there is not perfect
`
`
`
`
`
`
`
`
` 19
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`alignment between the '747 provisional and Brijesh. There hasn't been any
`evidence produced by the patent owner to go to that issue.
`JUDGE CHANG: So the inventive entity of the provisional
`application and the inventive entity of Brijesh, are they the same?
`MR. CHEN: No, they differ in -- they have commonality as to
`three individuals, but they differ as to the fourth.
`JUDGE CHANG: Okay.
`MR. CHEN: So that is why petitioner continues to maintain its
`position as to Brijesh on a 102(a) basis.
`JUDGE CHANG: What is your position regarding the patent
`owner's evidence that both the provisional and Brijesh was commonly
`known at the time of invention?
`MR. CHEN: Yes. So certainly under pre-AIA Section 103(c), the
`evidence that the patent owner has offered would possibly be relevant to
`whether Brijesh has status as a 102(e) reference. So we, the petitioner,
`evaluated that evidence as it came in on the patent owner response, and
`while we are not -- we chose not to pursue that in our reply, it's not depicted
`here in the slides, we are not conceding that position and rather would ask
`that the Board evaluate that evidence to determine whether indeed the patent
`owner has established common ownership.
`JUDGE CHANG: So why should the Board decide on that issue if
`Brijesh is prior art under 102(a)?
`
`
`
`
`
`
`
`
` 20
`
`
`
`Case IPR2018-00054 (Patent 8,862,747 B21)
`Case IPR2018-00055 (Patent 8,862,747 B21)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`MR. CHEN: Absolutely, Your Honor, no question that we
`proffered Brijesh as a reference under any one of the three subsections.
`Because the institution decision from the panel focused heavily on 102(b)
`over the course of the trial, the evidence that was developed naturally was
`focused on 102(b) with some additional evidence regarding Brijesh as a
`102(e) reference. But it's not necessary to focus on the common ownership
`issue and whether that has been a satisfactory showing. We believe that
`Brijesh qualifies either under 102(a) or 102(b).
`Quite a number of the slides that I intend to present focus on the
`102(b) issue, which was the subject of the Board's analysis at institution and
`thereby led to a fair amount of evidentiary back-and-forth as the