`
`Case 3:17-cv-05659-WHA Document 412-18 Filed 03/29/19 Page 1 of 4
`
`EXHIBIT 16
`
`
`
`Case 3:17-cv-05659-WHA Document 412-18 Filed 03/29/19 Page 2 of 4
`
` Volume 4
` Pages 615 - 831
`UNITED STATES DISTRICT COURT
`NORTHERN DISTRICT OF CALIFORNIA
`BEFORE THE HONORABLE WILLIAM H. ALSUP, JUDGE
`
`)
`FINJAN, INC.,
` )
` Plaintiff,
`)
` )
` VS. ) No. C 17-5659 WHA
` )
`JUNIPER NETWORKS, INC.,
`)
` )
` Defendant.
`)
` ) San Francisco, California
` Thursday, December 13, 2018
`
`
`TRANSCRIPT OF PROCEEDINGS
`
`
`APPEARANCES:
`
`For Plaintiff: KRAMER, LEVIN, NAFTALIS & FRANKEL LLP
` 990 Marsh Road
` Menlo Park, California 94025
` BY: PAUL J. ANDRE, ESQ.
` LISA KOBIALKA, ESQ.
` JAMES HANNAH, ESQ.
` KRISTOPHER B. KASTENS, ESQ.
`
` KRAMER LEVIN NAFTALIS AND FRANKEL LLP
` 1177 Avenue of the Americas
` New York, New York 10036
` BY: CRISTINA LYNN MARTINEZ, ESQ.
`
`
`(Appearances continued on next page)
`
`
`
`
`Reported By: Katherine Powell Sullivan, CSR No. 5812, RMR, CRR
` Jo Ann Bryce, CSR No. 3321, RMR, CRR
` Official Reporters
`
`I N D E X
`
`
`Thursday, December 13, 2018 - Volume 4
`
` PAGE VOL.
`Plaintiff Rests
`643
`4
`
`PLAINTIFF'S WITNESSES PAGE VOL.
`
`ICASIANO, ALEX
`By Videotaped Deposition
`
`GUPTA, SHELLY
`By Videotaped Deposition
`
`DEFENDANT'S WITNESSES PAGE VOL.
`
`BUSHONG, MICHAEL
`(SWORN)
`Direct Examination by Ms. Carson
`Cross-Examination by Mr. Hannah
`Redirect Examination by Ms. Carson
`
`RUBIN, AVIEL
`(SWORN)
`Direct Examination by Mr. Heinrich
`Cross-Examination by Mr. Andre
`Redirect Examination by Mr. Heinrich
`
`UGONE, KEITH RAYMOND
`(SWORN)
`Direct Examination by Ms. Curran
`Cross-Examination by Ms. Kobialka
`
`642
`
`643
`
`644
`645
`698
`709
`
`710
`711
`771
`796
`
`797
`798
`820
`
`4
`
`4
`
`4
`4
`4
`4
`
`4
`4
`4
`4
`
`4
`4
`4
`
`APPEARANCES (CONTINUED):
`
`For Defendant: IRELL & MANELLA LLP
` 1800 Avenue of the Stars, Suite 900
` Los Angeles, California 90067-4276
` BY: JONATHAN S. KAGAN, ESQ.
` ALAN J. HEINRICH, ESQ.
` JOSHUA GLUCOFT, ESQ.
` CASEY CURRAN, ESQ.
`
` IRELL & MANELLA LLP
` 840 Newport Center Drive, Suite 400
` Newport Beach, California 92660
` BY: REBECCA CARSON, ESQ.
` KEVIN X. WANG, ESQ.
`
`I N D E X
`
` E X H I B I T S
`
`
`TRIAL EXHIBITS IDEN EVID VOL.
`
`
`58
`
`685
`
`4
`
`182
`
`1059
`
`1070
`
`1170
`
`1241, Title Page & Table of Contents
`
`1241, Page 39
`
`1248
`
`1264
`
`1347
`
`1552
`
`2197
`
`686
`
`727
`
`720
`
`690
`
`730
`
`763
`
`760
`
`751
`
`682
`
`726
`
`724
`
`4
`
`4
`
`4
`
`4
`
`4
`
`4
`
`4
`
`4
`
`4
`
`4
`
`4
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Case 3:17-cv-05659-WHA Document 412-18 Filed 03/29/19 Page 3 of 4
`
` 740
`
`RUBIN - DIRECT / HEINRICH
`by sending one number, and so the table that contains the
`verdicts is consulted. There's a lookup in that table. It's a
`very fast process. That is return -- it returns an integer and
`that is sent to the SRX.
`THE COURT: Wait. Wait. But the Claim 10 doesn't
`require -- it doesn't -- let me put it differently -- it
`doesn't address what gets sent back to SRX. Claim 10 ends with
`the security profile data being put into a database. So what
`difference does it make what gets sent to the SRX? Claim 10
`doesn't address that.
`THE WITNESS: So, Your Honor, let me explain why I'm
`making this point.
`Dr. Cole claimed that there was this huge efficiency gain
`that Juniper was getting by being able to query a database very
`quickly, and that would allow the list of suspicious operations
`to be retrieved very quickly, and he used those words.
`And I am countering that by saying that that is not a
`requirement of the system. It doesn't need to retrieve the
`list of suspicious operations very quickly like he said it does
`because you only need to get a verdict.
`THE COURT: All right. Well, that may -- you may
`differ with him on that point, but I have a different question,
`which was: Does Claim 10 in your view even address what gets
`sent down to the SRX?
`THE WITNESS: It does not.
`
`RUBIN - DIRECT / HEINRICH
` 742
`Q. Now, what does it mean to be organized according to a
`database schema?
`A. So a database schema is a very specific thing. If you
`have a database -- I want to give an example of a -- say you
`have a school and the school wants to have a database of the
`students that are in the school. And what do they want to keep
`track of? The student's name, gender, age, GPA. If you're
`going to have a schema in a database, you have to figure out
`what all those fields are in advance.
`So let's say there's someone who's the database
`administrator at the school and they define those fields that I
`just said and they create that database and they start entering
`all the students' names and GPAs and gender and all that stuff
`and it all goes into the database. That's a very strict
`structure for that database.
`If you then wanted to add the country of birth for that
`student, you couldn't add it to that database. Why? Because
`there's no field for that. We've defined the fields. That is
`our schema.
`And so why would we restrict ourselves in this way? Why
`build a database that's so inflexible that we can't easily add
`fields to it without making a whole new database? The reason
`is that those databases have efficiencies for querying. Okay?
`You want to ask the question "Show me all the students who are
`18 years old and who are GPA of 3.5 or better." In a database
`
` 1
` 2
` 3
` 4
` 5
` 6
` 7
` 8
`
`
`91
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` 1
` 2
` 3
` 4
` 5
` 6
` 7
` 8
`
`
`91
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`RUBIN - DIRECT / HEINRICH
` 739
`these tables are generated -- there's a separate table that
`just contains some bookkeeping information and a verdict and no
`suspicious operations. They have separate names and they're in
`separate tables in DynamoDB.
`THE COURT: Wait a minute. Don't say anything for a
`
`second.
`
`(Pause in proceedings.)
`THE COURT: All right. For now I'll hold my further
`questions. Thank you.
`Go ahead.
`BY MR. HEINRICH:
`Q. And just a few clarifications here. So does the verdict
`contain or include a list of suspicious computer operations?
`A.
`It's a simple number. It doesn't contain anything except
`that number.
`Q. Now, you mentioned there are these two tables. When a
`request goes up to Sky ATP and the file is already seen, which
`one of those two tables that you mentioned is consulted?
`A.
`It only looks at the table that contains the verdicts, and
`there's a really good reason for that. The Sky -- the SRX
`device doesn't need to know what to do with suspicious
`operations. The list of suspicious operations is huge. It
`would create a tremendous amount of overhead to try to send
`that down to the SRX.
`And everything that needs to be done can be accomplished
`
` 741
`
`RUBIN - DIRECT / HEINRICH
`THE COURT: It does not. Okay. Thank you.
`Go ahead.
`BY MR. HEINRICH:
`Q. So let's kind of tie this together now by getting right
`into the specific reason for your noninfringement opinion, and
`what is that?
`A. So my reason is that Sky ATP does not have a database as
`the term "database" has been agreed to by the parties.
`Q. And according to Claim 10, what has to be stored in the
`database?
`A. So the database has to contain a list of suspicious
`operations. If there's no list of suspicious operations -- I
`can't stress this enough -- if there's no list of suspicious
`operations, then it's not considered a database of Claim 10
`because Claim 10 has been agreed upon to be a database that
`contains a list of suspicious operations.
`Q. Okay. And what's the parties' agreed definition for
`"database"?
`A. The agreed construction is a collection of interrelated
`data organized according to a database schema to serve one or
`more applications.
`Q. Now, in Sky ATP is a security profile that includes a list
`of suspicious computer operations stored in a database that
`meets this definition?
`A. No.
`
` 1
` 2
` 3
` 4
` 5
` 6
` 7
` 8
`
`
`91
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` 1
` 2
` 3
` 4
` 5
` 6
` 7
` 8
`
`
`91
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`Case 3:17-cv-05659-WHA Document 412-18 Filed 03/29/19 Page 4 of 4
`
`RUBIN - DIRECT / HEINRICH
` 744
`THE WITNESS: Let me try this without the microphone,
`and if anybody has any trouble hearing me, I'll pick it up.
`Okay? The court reporter too. You'll hear me.
`So first I'd like to describe a regular schema database
`like the one that I just talked about, and it's going to have a
`very specific structure. Let's say we have a student ID, age,
`gender, and GPA. It doesn't really matter what these fields
`are. I'm just giving an example.
`So we have these fields over here. Oh, I forgot the name
`but it doesn't matter. We have all of the fields that you
`would need for this example.
`And let's say that the university or high school, whatever
`the school is, starts entering the students into it. So the ID
`might be 12, the age might be 18, gender male, and GPA 3.8,
`good student. Okay?
`And we can imagine that this database is very, very big,
`but it can't have any information other than these things. If
`it did, then the schema would break. So if you tell me, "Okay.
`I want you to enter for student number 12 their country of
`origin?", I say, "Well, this database doesn't support that
`feature." Okay?
`This is a type of database called a schema database. This
`is the only database that is in question in Claim 10 because
`Claim 10 is defined, agreed to by the parties, as being a
`database that has a database schema.
`
`RUBIN - DIRECT / HEINRICH
` 746
`that I had previously put in my schema database into a
`schema-less database. I can do that but now you ask me, "Okay.
`Student 12 is from Germany. Can you put that in the database?"
`Sure. I'll just add some more information to my database.
`There's no schema. It's schema-less. So I can put whatever I
`want whenever I want into it.
`Now, it's not without its drawbacks because now if I say,
`"Well, show me all the students who are 18 years old who have a
`3.5 GPA," it's a lot of work. Right? I've got to go through
`the entire database. It's not structured with the schema. And
`that was the advantages of schema.
`And so a schema-less database is this kind of more
`free-form thing and it has a key.
`THE COURT: Can you elaborate on the key part again?
`Take the example where there's the one with all the circles.
`THE WITNESS: Sure. I'm going to look at this row
`right here (indicating) in the schema-less database.
`THE COURT: Yes. Is that entered sequentially into
`the file? Does normally go into -- in other words, when it's
`stored, does it all go in sequentially?
`THE WITNESS: So the way that it works is, in the code
`that manages this database -- and I am calling this a database
`somewhat carelessly because in the case a database has to have
`a schema; but I fall into the trap, and so would anybody
`working in this field, of using the term "database" more
`
` 1
` 2
` 3
` 4
` 5
` 6
` 7
` 8
`
`
`91
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` 1
` 2
` 3
` 4
` 5
` 6
` 7
` 8
`
`
`91
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`RUBIN - DIRECT / HEINRICH
` 745
`Now, there's another class of databases called schema-less
`databases. One of my primary disagreements with Dr. Cole was
`his definition of a schema-less database, which I believe is
`incorrect. He said you have an application where you don't
`enter some field, then you have a schema-less database, but
`that's not true.
`A schema-less database is a form of database that's a
`little bit newer, and a lot of companies like Amazon are
`putting a lot of resources into building these type of database
`because they match today's applications of big data, and big
`data is a very different concept from having a school -- a
`school student database.
`So let me show you what I consider to be and what I
`believe is standard in the industry as a schema-less database.
`In a schema-less database, you have keys. Okay? And the
`database is said to be keyed off of the key, and you can take
`whatever information you want. I can take a packet of
`information in any format, and I can stick that in the database
`and have a key here (indicating).
`Okay? We've heard the term "hash." That's relevant here.
`You can put the hash of whatever you're putting here
`(indicating) over here (indicating) to help you find it later.
`You could even put several things of different types in a
`schema-less database.
`So now let's say that I want to put the same information
`
` 1
` 2
` 3
` 4
` 5
` 6
` 7
` 8
`
`
`91
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`RUBIN - DIRECT / HEINRICH
` 743
`with a schema you can create a query exactly like the one I
`just described and immediately it will spit out all the
`students that meet that. That's very powerful. Okay? And
`that is why you have a schema in a database.
`Q. Now, are there also schema-less databases?
`A. Yes. So there's another type of database which is called
`schema-less. It doesn't have the properties that I just showed
`you.
`Q. What's the difference between a database with a schema and
`a schema-less database?
`A. Your Honor, I would like to be able to show them that
`with --
`
`THE COURT: Please.
`(Pause in proceedings.)
`THE COURT: Why don't you put it closer to the jury
`
`MR. KAGAN: Over there, Your Honor?
`THE COURT: No. I'm going to come around where my law
`clerk is and look so just you need to face the jury.
`All right. So where is our witness? Oh, there he is.
`All right. So go ahead and keep your voice up.
`MR. ANDRE: Your Honor, may I stand where I can see it
`
`THE COURT: Of course.
`MR. ANDRE: Actually, if it's there, I'm fine here.
`
` 1
` 2
` 3
` 4
` 5
` 6
` 7
` 8
`
`
`91
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`box.
`
`also?
`
`