`571-272-7822
`
`
`
`RECORD OF ORAL RECORD
`
`IPR2016-00159, Paper No. 49
`March 9, 2017
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`------
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`------
`
`PALO ALTO NETWORKS, INC. And
`
`BLUE COAT SYSTEMS, LLC
`
`Petitioner,
`
`V.
`
`FINJAN, INC.,
`
`Patent Owner.
`
`------
`
`Case IPR2016-00159
`
`Patent 8,677,494 B2
`
`------
`
`
`
`Before: CHARLES J. BOUDREAU, ZHENYU YANG, and
`SHEILA F. McSHANE (by Videoconference), Administrative Patent
`Judges.
`
`
`The above-entitled matter was heard on Thursday,
`February 16, 2017 at the U.S. Patent and Trademark Office, 26 South Fourth
`Street, Courtroom 322, San Jose, California at 10:01 a.m.
`
`
`
`APPEARANCES:
`
`FOR THE PETITIONER:
`
` Cooley, LLP
` By: ORION ARMON
` oarmon@cooley.com
` 380 Interlocken Crescent
` Suite 900
` Broomfield, Colorado 80021-8023
` (720) 566-4119
`FOR THE PATENT-OWNER:
`
` Kramer Levin
` By: JAMES HANNAH
` jhannah@kramerlevin.com
` 990 Marsh Road
` Menlo Park, California 94025-1949
` (650) 752-1712
`
`
`
`2
`
`
`
`P R O C E E D I N G S
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`(10:01 a.m.)
`JUDGE BOUDREAU: Good morning. This is the
`oral hearing in case IPR2016- 00159, captioned Palo Alto
`Networks, Incorporated versus Finjan, Incorporated, to which
`IPR2016-01174, captioned Blue Coat Systems, Incorporated
`versus Finjan, Incorporated has been joined.
`I'm Judge Boudreau. We're joined on the video
`screen from Virginia by Judges Yang and McShane. We seem
`to have some additional interference here from Virginia.
`JUDGE YANG: No, Judge Boudreau. It's not from
`us. We're very quiet here.
`JUDGE BOUDREAU: Okay. It seems to have
`
`stopped.
`
`All right. Anyway, because of the limitations on
`our audio technology Judges Yang and McShane will only be
`able to hear you if you're standing at the podium speaking into
`the microphone. So we ask that anytime you're speaking you
`remain at the podium.
`Let's have counsel come up and enter an
`appearance and state your names for the record. We'll start
`with counsel for the Petitioner.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`3
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`MR. ARMON: Good morning. Orion Armon, lead
`counsel for Petitioner Palo Alto Networks. And I have my
`colleague Mikaela Stone with me, and Michael Skrzypek here
`as our trial tech sitting next to me at counsel table.
`JUDGE BOUDREAU: And for Patent Owner?
`MR. HANNAH: Good morning, Your Honor. Nice
`to see you in person instead of over a video screen.
`Good morning, Your Honors. James Hannah on
`behalf of Finjan and with me is Hien Lien.
`JUDGE BOUDREAU: Thank you.
`MR. HANNAH: Thanks.
`JUDGE BOUDREAU: W e set out the procedure for
`today's hearing in the Hearing Order, which was Paper 46, but
`I'll just take a few minutes to remind everyone. Petitioner and
`Patent Owner will each have 30 minutes to present arguments.
`Petitioner will go first and present its case regarding the
`challenged claims and may reserve time for rebuttal. Patent
`Owner will then respond to Petitioner's presentation.
`Petitioner will then have the opportunity to use any reserved
`rebuttal time to respond to Patent Owner's arguments.
`The parties may also discuss their respective
`motions to exclude evidence, and in Patent Owner's case, its
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`4
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`Motion for Observations on Cross Examination, during their
`initially allotted time.
`And if you do bring up any of those motions during
`your initial presentation, then the other party will have the
`opportunity to respond during its response or rebuttal time.
`Just a few other points. Keep in mind that Judges
`Yang and McShane won't be able to see the screen over there.
`So anytime that you're referring to any demonstratives or
`papers in the case, please make sure that you identify them by
`the exhibit number or the page number of the demonstratives
`just so that the record is clear for the transcript as well as for
`Judges Yang and McShane's benefit.
`Lastly, just as a common courtesy, if you believe
`that the other party is making a new argument that wasn't
`presented in the briefs or if you have any our objections, we
`just ask that you reserve them until the end or until your
`presentation, rather than interrupting.
`With that, unless there are any questions from any
`of the parties, we may proceed.
`And Mr. Armon, just before you begin, would you
`like to reserve any of your time for rebutt al?
`MR. ARMON: I'd like to reserve ten minutes for
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`5
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`rebuttal, please.
`JUDGE BOUDREAU: Okay.
`MR. ARMON: Good morning. And may it please
`
`the Board.
`
`Turning to slide 2 of our demonstratives, I intend
`to briefly speak to the priority date of the '494 patent, then
`turn briefly to the issue of claim construction and specifically
`the plain and ordinary meaning of the "suspicious computer
`operation" clause in the petition claims before reaching the
`issues of obviousness including specifically questions
`concerning the key claim construction issue in dispute, as well
`as hopefully reaching the issue of database.
`Turning first to the priority date of the petition
`claims, Palo Alto Networks' position is that the proper priority
`date for the petition claims is March 30th, 2000.
`In the institution decision the Board preliminarily
`found that the petition claims are entitled to an earlier priority
`date, and it's our position that the Board should reach this
`decision and find that the March 30th, 2000 date is the correct
`date.
`
`The background here, of course, is that Patent
`Owner filed a certificate of correction in the '822 priority
`
`
`
`6
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`patent and the question is the effect of that certificate of
`correction in this case.
`JUDGE BOUDREAU: But that petition was
`granted on July 24th -- 25th, 2014; is that right?
`MR. ARMON: Yes. Turning to slide 6 of our
`demonstratives, there's a timeline that shows the history. A
`petition to correct the priority claim was originally filed in
`2014. It's the first bullet. That's Exhibit 1017 in our
`materials. The petition was preliminarily granted on July 25th,
`2014, and on September 8th, 2014, the examiner concluded in a
`final Office action that the claims were entitled to a May 2000
`date. And we know subsequently the Board reversed the
`determination of the examiner, but the last three bullets on
`slide 6 are the material bullets for this issue.
`Palo Alto Networks was served with a complaint in
`District Court in November of 2014 and the Petition in this
`case, in this proceeding, was filed on November 6th, 2015.
`Those are key dates because they are both before the
`reexamination certificate's issuance on February 16th, 2016.
`Turning very briefly to the decision to institute, in
`the decision the Board preliminarily found two things:
`Preliminarily found that the date that the examiner entered the
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`7
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`certificate of correction was an effective date for correcting
`the priority chain, or that it was effective retroactively once
`the reexamination certificate issued.
`Our position and, again as stated in the Petition , is
`that the statute governing certificates of correction as well as
`the regulations on this issue do not allow this type of
`correction to have retroactive effect, and because the IPR in
`this case was filed prior to the issuance of the reexamination
`certificate, that certificate has no effect here.
`Now, the operative statutory language is in section
`
`255.
`
`JUDGE BOUDREAU: Why wasn't the petition
`effective as of the date it was granted?
`MR. ARMON: That relates to the language of the
`statute as it's been interpreted by the Federal Circuit as well as
`the CFR. On the face of section 255 it states that a certificate
`of correction is only effective for causes thereafter arising, and
`that language has been interpreted by the Federal Circuit in the
`H-W Tech. case. For the panel I'm on slide 8.
`And the key portion of the Federal Circuit's
`holding in that case was that certificate of correction is only
`effective for causes of action arising after a certificate was
`
`
`
`8
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`issued.
`
`JUDGE BOUDREAU: Did that case involve a
`certificate of correction with respect to a priority date or was
`that the correction of a claim?
`MR. ARMON: It did, Your Honor. Sorry. It did,
`but it was related to a District Court proceeding, not an IPR
`proceeding. But there was a priority date of correction there.
`JUDGE BOUDREAU: In any event, when wouldn't
`Swimmer and Martin both be prior art regardless of which
`priority date we ultimately determine is correct?
`MR. ARMON: Finjan takes the position -- and
`that's one of the reasons I think this is an important issue for
`the Board to reach in this case. Finjan's position with respect
`to Martin is that the stamp from a library that received the
`Martin publication in 1998 is the actual publication date and
`Finjan takes the position that the preliminary 1997 priority
`date that the Board afforded the '494 patent claims subject to
`this proceeding is therefore a date earlier than the Martin
`publication date.
`So we have an actual dispute on our hands not only
`concerning whether Martin qualifies as prior art for the '494
`patent. We also have the uncertainty regarding issue
`
`
`
`9
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`preclusion. Courts have not yet decided the effect, if any, of
`issues like this when they are adjudicated in one forum and
`there is underlying litigation going in District Court.
`So it's important to Petitioner to have this issue
`decided properly for that reason.
`The third issue is that the sole reason why the
`Board decided not to institute our lead grounds, the Touboul
`grounds, was because of this priority date issue. So having a
`clear record on that is very important to us.
`I'll just add on slide 9 that the CFR also aligns
`with both the Federal Circuit's interpretation of the statute and
`the statutory language itself and it states that the proposed
`amendments will not be effective until the reexamination
`certificate is issued and published.
`So very briefly jumping back, what section 255 and
`the CFR as well as the Federal Circuit's interpretations hold is
`because the reexamination certificate issued after this Petition
`was filed it should have no retroactive effect in this case. And
`although perhaps for future cases Finjan may be entitled to that
`priority date, our position is that in this particular case it
`cannot be retroactive and therefore the proper priority date
`should be the 2000 date that's argued in the Petition .
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`10
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`JUDGE YANG: Counsel, I have a question.
`Earlier you mentioned the H-W Technology case from the
`Federal Circuit. I missed that. Did you say it relates to a
`priority issue or it doesn't?
`MR. ARMON: Your Honor, it does, but it was a
`District Court-related issue. To my knowledge, there's been no
`case deciding this issue specifically with respect to the fact
`scenario we have here where the prior proceeding is an IPR
`proceeding.
`JUDGE YANG: That is fine. Do you by any
`chance recall what H -W -- what the case is about? I have to
`admit I'm not very familiar with the case.
`MR. ARMON: Your Honor, it's been some time
`since I read the full case.
`JUDGE YANG: That is fine.
`MR. ARMON: But as I recall, the facts are
`analogous to the situation we have here in the sense that you
`had overlap between a certificate of correction that had been
`filed earlier, in other words, a petition for a certificate of
`correction had been filed earlier, I believe, but then a District
`Court proceeding began and the certificate of correction wasn't
`actually issued until after the proceeding had commenced.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`11
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`And so you had the same type of overlap that you
`have in this case. And the question became whether or not it
`was the filing date for the certificate of correction or its actual
`final issuance that mattered. And the key holding that I think
`just reiterates the plain text of the statutory language is that
`these certificates of correction never have retroactive effect.
`And so what we have here is a circumstance where the -- the
`issuance of the reexam certificate coming in 2016 -- let me hop
`back to the timeline -- is after both the initiation of the District
`Court litigation against Palo Alto Networks as well as after the
`filing of the Petition.
`And there is an important policy consideration
`here. Petitioner filed its Petition right at the statutory
`one-year deadline. We couldn't have waited any longer for that
`proceeding to conclude or we would have passed our bar date.
`So it's a matter of policy as well in addition to the statutory
`language. I think as a matter of fairness we had to petition on
`the patent as it then stood uncorrected.
`So we submit that this is a material issue both with
`respect to the priority date of Martin, also because of the issue
`preclusion effect, and finally because the Board's
`non-institution of the Touboul grounds was solely because of
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`12
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`this issue.
`
`I'd like to now turn briefly to claim construction, if
`I may. There's one issue in this case that remains disputed.
`The patent has expired. The parties agree on that and so the
`claim construction standard in this case has converted to the
`Phillips standard.
`JUDGE BOUDREAU: Will that affect any of the
`constructions instanced -- any of the constructions that we
`determined in the decision to institute?
`MR. ARMON: Your Honor, there's two
`constructions that are agreed. I believe that the parties still
`have agreement on those. And so I don't think that the change
`would affect that. There's no briefing on that issue and we're
`certainly not taking the position that either of the agreed
`constructions would change. With respect to -- and that
`specifically relates to "downloadable" and "database." Those
`constructions are not disputed.
`There is a dispute about the plain and ordinary
`meaning of -- again slide 13 summarizes the -- the positions of
`the parties on the two agreed terms.
`There is a dispute between the parties on the
`limitation "deriving security profile data for the
`
`
`
`13
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`Downloadable" -- and this is the key portion -- "including a list
`of suspicious computer operations that may be attempted by the
`Downloadable."
`The Petitioner proposed a construction for this
`term. The Board at institution went with plain and ordinary
`meaning. We now have a dispute about what the plain and
`ordinary meaning of what this term is. And to summarize
`briefly, Patent Owner focuses on the notion that operations
`need to be deemed suspicious.
`Our position is that deeming is an inappropriate
`characterization of this claim limitation because it suggests
`affirmative action would be required. And if you look at the
`'194 patent that contains the key language that provides the
`support for this limitation, you see that it's actually very
`broadly disclosed.
`So I'll turn to slide -- first of all very briefly to
`slide 16. It's worth reminding the panel that we have a claim
`differentiation issue here. Claims 7 and 8 indicate that the
`Downloadable Security Profile data can include a URL or a
`digital certificate. I think it goes without saying that those
`things wouldn't be deemed suspicious. So DSP data certainly
`can include other information.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`14
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`But turning to the specification of the priority
`patent for this disclosure, which is the '194 patent on slide 17.
`There's broad disclosure in multiple places in the '194 patent
`and I'm speaking to the '194 patent because the '494 patent
`does not directly address Downloadable Security Profile data.
`That disclosure is more about mobile protection code.
`But here in the '194 specification there's several
`disclosures that describe DSP data. On slide 17 we have one
`and it says "DSP data 310 includes a list of all." I think even
`under a plain and ordinary meaning construction as in Phillips
`is important. All potentially hostile or suspicious computer
`operations. So the word "all" and "potentially" are important
`to us here. That may be attempted by a specific Downloadable.
`JUDGE BOUDREAU: Now, the claims of the '494
`patent use the term "suspicious ," whereas the '194 patent uses
`"potentially hostile." Is there any difference between
`"suspicious" and "potentially hostile"?
`MR. ARMON: Your Honor, we think that the
`description of "suspicious computer operation" as "potentially
`hostile" informs the meaning of the claim terms in the '494. So
`we have the language here. We also have other language in
`Exhibit 1013 which is the '194 patent.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`15
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`At column 4, for example, lines 32 to 35, it says:
`"A list of all potentially hostile or suspicious computer
`operations." At column 5, lines 50 to 54 there's a reference to
`"a list of all operations which could ever be deemed potentially
`hostile." The key point here from Petitioner's perspective is
`that the plain and ordinary meaning construction even under
`Phillips should not read in a limitation from the spec that
`requires some affirmative determination.
`The broad disclosure here focuses on "potentially
`hostile or suspicious operations" including the word "all."
`It means that you could certainly have a wide range
`of information or activities captured in the DSP data without
`an actual determination of suspiciousness.
`JUDGE BOUDREAU: Is it your position that all
`MS- DOS functions are suspicious operations?
`MR. ARMON: No, Your Honor, it's not. I would
`agree that if a disclosure focused on capturing literally
`everything, that wouldn't satisfy this limitation. But it's
`important. I'd like to turn to the disclosure in Swimmer and
`show you why Swimmer doesn't disclose capturing everything.
`Now Swimmer actually teaches away from
`capturing "all," and Swimmer teaches -- and this is Exhibit
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`16
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`1006. In Exhibit 1006 you look at page 7. There's actually a
`description in section 4.1 that describes a very first
`implementation of Swimmer. And what it says about the very
`first implementation was that it was an auditing system that he
`used a filter to catch everything. It says specifically the very
`first implementation of an auditing system was a filter which
`was placed before DOS services and registered all calls to DOS
`functions. Okay?
`Swimmer explains that that system was scrapped.
`In fact, the very next line on the bottom of page 7
`says, "This limitation was soon scrapped." So a filter catching
`everything was the proof of concept that they did not use and
`that's not the focus of Swimmer.
`Let me hop to slide 22 here. What you see instead
`with Swimmer is a design that is intended to capture
`potentially suspicious activity by using hooks into the op
`codes. And the effect of those hooks in the op codes is to
`catch potentially suspicious operations. Now, we've explained
`that a person of ordinary skill in the art would have understood
`the examples of the function calls that are being hooked
`including those that were known suspicious.
`It's also important to note, as the Board noted both
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`17
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`in its institution decision at 21, and as you'll see at Exhibit
`1006 at page 13, is that Swimmer, in our view, teaches not
`only designing the system using these hooks into the op codes
`to capture suspicious activity, but Finjan also -- I'm sorry, but
`Swimmer also discloses on page 13 that -- and I'll jump to that
`language right now. Over to slide 25.
`On slide 25 Swimmer also teaches tuning the audit
`system to derive only the data necessary for detecting viruses.
`You'll see that's the language here highlighted in yellow that
`the audit system can be tuned to provide only the necessary
`data in order to eliminate overhead.
`So we have a disclosure in Swimmer that first
`notes on page 7 that the original system did catch everything.
`That was scrapped. We then have the disclosure describing
`that -- on page 9 of Exhibit 1006 that Swimmer hooks into the
`op codes to catch the potentially suspicious functions; and then
`on page 13 of Exhibit 1006 an explanation, an express teaching
`of tuning the system to provide only the necessary data, which
`again would be the suspicious computer information.
`JUDGE BOUDREAU: How does that tuning relate
`to Figure 3 of Swimmer?
`MR. ARMON: So, Your Honor, in F igure 3 -- I'll
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`18
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`jump back here. Figure 3 is on slide 22, at least an excerpt of
`it. Figure 3 is a depiction in plain English text of the readout
`of the audit trail that's generated by the emulator in Swimmer.
`JUDGE BOUDREAU: That actually raises a
`question. Does F igure 3 depict a raw audit trail or does it
`depict an NADF file?
`MR. ARMON: Your Honor, our understanding is
`that it depicts an audit trail. Figure 3 on page 9 of Exhibit
`1006, we don't have the whole thing here, but it's expressly
`captioned an excerpt from an audit trial. You'll see in the text
`on page 9 that it explains that the audit record attributes of
`record as collected by the PC emulator have the following
`structure. And that's the structure that's shown in Figure 3.
`Now, Swimmer explains. So, in other words, this
`structure that you see is read out by the emulator as the audit
`trail. What Swimmer explains a few pages later on page 12 of
`Exhibit 1006, section 5.1, regarding universality, is that the
`audit trail values are included in what's called the normalized
`audit data format. So the normalized audit data format is
`actually created by ASAX as an additional publication step to
`assist with the ASAX analysis of the data.
`The actual audit trail values as shown on section
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`19
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`5.1 are included in the NADF file, but what's read out by the
`emulator is the structure shown in F igure 3. I should note, as
`Swimmer explains on page 10, I believe, that that is a
`somewhat simplified plain English version. There's some
`additional information that he said he left out for
`simplification purposes.
`JUDGE BOUDREAU: Well, on page 10 it also
`says an example of an audit trail as depicted in F igure 3. This
`is a human- readable representation of the binary NADF file.
`So that's why I think there's a little bit of confusion as to
`whether Figure 3 is the audit trail or if it's the NADF file.
`MR. ARMON: Your Honor, our understanding is
`that the -- because as explained in section 5.1, the actual audit
`data value is included in the NADF file. The entirety of the
`audit data is included in the NADF file, but you have the
`additional identifier lengths and for the integer information
`that's bulleted out in section 5.1 that's expressly characterized
`as being used to assist ASAX with its processing of the data.
`So we're getting into the issue of database that I'd
`like to speak to very quickly before I reserve the remainder of
`our time. Our claim construction for database --
`JUDGE BOUDREAU: Can a log file be interpreted
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`20
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`as a database?
`MR. ARMON: Yes, Your Honor. We think it can
`be, if it has the right characteristics. During cross- examination
`of Dr. Goodrich, one of Patent Owner's experts, he agreed that
`labels are less important than structure, and the focus should
`be on the structure that's disclosed and not on whether it's
`labeled as a log file, labeled as a flat file , or something else.
`Our position is that the description in Swimmer is
`of a flat-file database. And let me jump briefly to -- let's see
`if I perhaps went too far here. I'd like to briefly touch on the
`construction that explained why Swimmer's disclosure satisfies
`the construction.
`The construction that's agreed by the parties is a
`collection of interrelated data and -- that's one, interrelated
`data. Organized according to a database schema. That's part 2.
`To serve one or more applications is part 3.
`I don't think there's any dispute between the parties
`regarding "interrelated data."
`JUDGE BOUDREAU: What is the interrelationship
`between the data that's listed in Figure 3 here of Swimmer?
`MR. ARMON: Figure 3 I think is our best example
`of the data that's captured. It's related in the sense that these
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`21
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`are all function calls relating to the same code segments. And
`so you'll notice, Your Honor, that there is an interrelationship
`because the CS in the first field or column in Figure 3 has a
`shared number because you're going to be capturing
`information about the same Downloadable. You might be
`capturing multiple bits of information about activity that that
`Downloadable will generate. So there is an interrelation as
`shown by the fact that you have the same code segment
`number.
`
`The second requirement is a database schema. The
`schema that is applied to the audit trail information is
`described on page 9 of Exhibit 1006, and our view is that the
`Denning model that's characterized there unquestionably is a
`schema. And we pointed to three bits of information to support
`that interpretation. One is Dr. Rubin, the Petitioner's expert,
`who supports that view.
`We also point to Exhibit 1086, which is a Chang
`reference. Chang is a prior art patent that characterizes at
`column 3, lines 15 to 20, that a flat-file database is made up of
`records composed of fields. And I think that's what we have
`here. We have records or rows shown in F igure 3 and fields or
`columns shown as well.
`
`
`
`22
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`Finally, again, as I mentioned earlier,
`Dr. Goodrich in his deposition transcript, which
`you find in Exhibit 1098, page 63, lines 14 to 22, agrees that
`the focus should be on the structure, not on the name.
`The last piece that we need to show to satisfy the
`construction is serving one or more applications. And our
`slide 32 here speaks to that issue. Swimmer discloses again in
`1006 at 2 and several other places that the ASAX system uses
`the activity data, in other words, the audit system data that's
`collected by the PC emulator.
`So we have ASAX as the one or more applications
`using the data, and I think that the disclosure concerning the
`NADF or normalized audit data format on page 12 supports
`that view. The entire discussion in section 5.1 is about the fact
`that the data needs to be normalized so that ASAX can use it.
`Unless the panel has any questions, I'd like to
`reserve my remaining five minutes for rebuttal.
`JUDGE BOUDREAU: Yes, you can.
`MR. ARMON: Thank you.
`MR. HANNAH: Just give me a second to set up
`here. Thanks.
`First of all, I just wanted to say thank you very
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`23
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`much to the Board. Thank you for your time. Thank you for
`your consideration. I know all of you have a lot of cases and a
`lot of things to do, but this case is very important to Finjan as
`is all the other cases that we're involved with.
`So I do want to thank you for your time and I
`appreciate your consideration in viewing the record anew,
`looking at all the evidence and considering the references as a
`whole to make the determination in this case.
`Now, may it please the Court here, I've heard from
`some people that proving a patent is obvious is easier than
`proving a patent is anticipated. For anticipation you have to
`match each element to the claim to the prior art and for
`obviousness you have to consider the patent and the prior art as
`a whole.
`
`I think in this case proving obviousness is much
`more difficult because of the Swimmer reference and because
`you must consider Swimmer as a whole and what it teaches.
`All you have to do is look to the abstract of
`Swimmer to see exactly what it is. It's Exhibit 1006.
`Swimmer has two stages. The first stage is
`emulating a file, which results in this activity data. The
`second stage is applying rules to that activity data to determine
`
`
`
`24
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`if there's a virus present. Petitioner's entire case is based on
`stage 1, the emulator generating the activity data. But when
`you consider Swimmer as a whole Petitioner's argument
`completely falls apart because in Swimmer -- and it teaches in
`Swimmer you don't know if you have a virus. You don't know
`if anything is suspicious until you apply those rules to the
`activity data.
`And for that reason alone you don't have a list of
`suspicious operations. If I asked the court -- if the court
`reporter here asked me can you list all of the females in this
`room and I gave him a list of everyone in this room, the court
`reporter would tell me, "You gave me the wrong list.
`You did not give me a list of all the females."
`And that's exactly what Swimmer does and how
`Petitioner uses Swimmer. If you look at that figure, F igure 3
`-- which is everything. It's the list of suspicious operations.
`It's the database. It's the storage in the database. Everything
`is in Figure 3. And you look to the description of it right
`below -- I'm looking at page 9.
`It says, "The audit system was integrated into an
`existing PC emulation by placing hooks into the module for
`processing all op codes."
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`25
`
`
`
`Case IPR2016-00159
`Patent 8,677,494 B2
`
`
`
`I heard during the Petitioner's presentation that
`that does not meet the claim element. He said that that process
`was scrapped. It absolutely was not scrapped.
`Swimmer specifically says that you process all op
`codes and by processing all op codes you have a -- you have an
`audit trail, a log file of all the system activity. There's no
`determination of a list of suspicious operations. It simply does
`not happen until later in the process.
`The other reason that obviousness is much tougher
`to prove in this case in considering Swimmer as a whole is the
`database. When Swimmer talks about a database Swimmer
`says that it's inefficient. It doesn't work. It only works -- it
`won't work for polymorph