`Case 1:16-cv-00455-RGA Document 394-1 Filed 02/07/18 Page 1 of 21 PagelD #: 29730
`
`
`
`
`
`
`
`
`
`EXHIBIT A
`EXHIBIT A
`
`
`
`Case 1:16-cv-00455-RGA Document 394-1 Filed 02/07/18 Page 2 of 21 PageID #: 29731
`1
`
`(The proceedings occurred at 10:05 o'clock a.m. as
`
`follows:)
`
`THE COURT: All right. Please be seated.
`
`This is time the Court has set for a hearing with
`
`witnesses inAcceleration Bay v. Activision Blizzard, Civil
`
`Action No. 16-453, and two related cases.
`
`Can we have counsel's appearances, Mr. Rovner?
`
`MR. ROVNER: Good morning, your Honor.
`
`Phil Rovner from Potter Anderson for Plaintiff
`
`Acceleration Bay.
`
`With me from Kramer Levin is Jim Hannah.
`
`MR. HANNAH: Good morning, your Honor.
`
`MR. ROVNER: Paul Andre and Aaron Frankel.
`
`And from my office Alan Silverstein.
`
`Dr. Mitzenmacher, who you'll be hearing from, is in the
`
`back.
`
`THE COURT: All right. Thank you.
`
`Mr. Kraftschik?
`
`MR. KRATSCHIK: Good morning, your Honor.
`
` UNITED STATES DISTRICT COURT
` FOR THE DISTRICT OF DELAWARE
`
` CA NO. 16-453-RGA,
` 16-454-RGA,
` 16-455-RGA
`
` January 29, 2018
`
`
`
` 10:05 o'clock a.m.
`
`
`: : : : : : : :
`
`ACCELERATION BAY LLC,
`
` Plaintiff,
`
` v.
`
`ACTIVISION BLIZZARD INC.,
`
`: :
`
` Defendant,
`.............................
`
`TRANSCRIPT OF MOTION FOR CLARIFICATION
`BEFORE THE HONORABLE RICHARD G. ANDREWS
`UNITED STATES DISTRICT JUDGE
`
`APPEARANCES:
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`3
`
`P R O C E E D I N G S
`
`12
`
`3
`4
`10:05:15 5
`6
`7
`8
`9
`10:05:33 10
`11
`12
`13
`14
`10:05:41 15
`16
`17
`18
`19
`10:05:53 20
`21
`22
`23
`24
`10:06:07 25
`
`2
`
`1
`2
`3
`4
`10:06:17 5
`6
`7
`8
`9
`10:06:27 10
`11
`12
`13
`14
`10:06:39 15
`16
`17
`18
`19
`10:06:45 20
`21
`22
`23
`24
`10:06:56 25
`Page 1 to 4 of 79
`
`Stephen Kraftschik from Morris Nichols for defendants.
`
` With me today from Winston & Strawn are Michael Murray
`
`and David Enzminger.
`
`MR. MURRAY: Good morning, your Honor.
`
`4
`
`MR. ENZMINGER: Good morning, your Honor.
`
`THE COURT: Good morning to you all.
`
`Plaintiff, I presume you're going first?
`
`MR. MURRAY: Excuse me, your Honor.
`
`May I say one thing for the record before we get
`
`started?
`
`THE COURT: Yes.
`
`MR. MURRAY: We were just told right before the hearing
`
`started that the witness who put the Declaration in addressing
`
`this term to the Court is actually not the witness that they're
`
`calling today.
`
`THE COURT: Well, they said Dr. Mitzenmacher. I
`
`thought it was probably going to be Dr. Medvidovic.
`
`MR. MURRAY: So we were under the assumption that they
`
`were bringing the witness in for the Declaration on its term,
`
`and they did not.
`
`I just want that to be on the record.
`
`THE COURT: Okay.
`
`MR. HANNAH: Thank you, your Honor.
`
`THE COURT: Go ahead, Mr. Hannah.
`
`MR. HANNAH: So, your Honor, we were just calling to
`
`call the witness and have him testify on direct examination, and
`
`let them do cross-examination and then redirect.
`
`THE COURT: That's what I had in mind.
`
`MR. HANNAH: We would like to call to the stand Dr.
`
`02/01/2018 12:05:57 PM
`
`For Plaintiff: POTTER, ANDERSON & CORROON
` BY: PHILIP A. ROVNER, ESQ
`
`25
`
` BY: ALAN R. SILVERSTEIN, ESQ
`
` -and-
`
` KRAMER LEVIN NAFTALIS & FRANKEL
`
` BY: JAMES R. HANNAH, ESQ
`
` BY: PAUL J. ANDRE, ESQ
`
` BY: AARON M. FRANKEL, ESQ
`
`1
`2
`3
`4
`5
`
`678
`
`For Defendant: MORRIS NICHOLS ARSHT & TUNNELL
`
` BY: STEPHEN J. KRAFTSCHIK, ESQ
`
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`1 of 31 sheets
`
` -and-
`
` WINSTON & STRAWN LLP
`
` BY: MICHAEL M. MURRAY, ESQ
`
` BY: DAVID P. ENZMINGER, ESQ
`
`Court Reporter: LEONARD A. DIBBS
`
` Official Court Reporter
`
`
`
`Case 1:16-cv-00455-RGA Document 394-1 Filed 02/07/18 Page 3 of 21 PageID #: 29732
`5
`
`7
`
`A.
`
`Sure.
`
`So one time, a couple of years ago, I had my last
`
`graduate student that graduated. We were working on
`
`verification systems for cloud computing.
`
`And the idea there is that these days, whether you're a
`
`scientist or a business, you have may have some big problem that
`
`you're going to ask a lot of cloud system providers to compute
`
`the answer for you. Someone like Amazon, or Google, and so on.
`
`And, so, you send your process over, and it does this
`
`large computation for you, and it will return some sort of
`
`Mitzenmacher, please.
`
`DR. MICHAEL MITZENMACHER, after having been first duly
`
`sworn as a witness for the plaintiff, was examined and testified
`
`as follows:
`
`MR. HANNAH: Your Honor, would I like to pass up some
`
`slides.
`
`THE COURT: Sure.
`
`MR. HANNAH: I have a couple of clean copies of the
`
`'344 patent, if you don't have them readily accessible.
`
`THE COURT: Okay.
`
`MR. HANNAH: May it please the Court?
`
`THE COURT: Yes.
`
` DIRECT EXAMINATION
`
`Good morning, Dr. Mitzenmacher.
`
`Good morning.
`
`Can you please start out by describing your educational
`
`Sure.
`
`I was an Undergraduate at Harvard University. I
`
`received an Undergraduate Degree in Mathematics and Computer
`
`Science in 1991.
`
`I then went to the other Cambridge, the University of
`
`Cambridge for a year in England where I did a mathematics
`
`program called the Part 3 Triopods, and received a certificate
`
`1
`2
`3
`4
`10:10:20 5
`6
`7
`8
`9
`10:10:40 10
`11
`12
`13
`14
`10:10:51 15
`16
`17
`18
`19
`10:11:09 20
`21
`22
`23
`24
`answer.
`10:11:21 25 Q.
`
`answer.
`
`So it might say the answer is 37.
`
`I assume the question is on your side, why should we
`
`trust that answer?
`
`So part of it may be a security issue. You may have
`
`concerns that some malicious entity is taking over part of the
`
`cloud, and we try to give you a background answer. Or it may
`
`just be a concern that there's some sort of flaw in how they ran
`
`the software. We have the software and give you back the
`
`answer.
`
`So what we devise is some sort of an algorithm and
`
`protocol that you communicate back and forth, so that the cloud
`
`system can convince you that it's actually presented, the right
`
`What type of organizations are you involved with?
`
`1
`2
`3
`4
`10:07:47 5
`6
`7
`8
`9
`10:07:59 10
`11
`12
`13
`14
`BY MR. HANNAH:
`10:08:22 15 Q.
`16
`A.
`17 Q.
`18
`background?
`19
`A.
`10:08:30 20
`21
`22
`23
`24
`10:08:46 25
`
`6
`
`8
`
`of Math Studies, sort of the equivalent of a Master's.
`
`I then went to the University of California at Berkeley
`
`Can you please tell us about your educational and your
`
`Sure.
`
`After I finished my Ph.D., I went to work as a research
`
`scientist at Digital Systems Research Center. That was part of
`
`the Digital Equipment Corporation in Palo Alto.
`
`And I switched over to being a professor. I joined
`
`Harvard in January of 1999. I was promoted to an Associate
`
`Professor in 2002. And then a full Professor in 2005.
`
`I have been -- I've stayed at Harvard since. I have
`
`1
`2
`3
`4
`10:11:39 5
`6
`7
`8
`9
`10:12:01 10
`11
`12
`13
`14
`10:12:13 15
`regularly on program committees for conferences and so on.
`16 Q.
`17
`your work.
`18
`19
`10:12:26 20
`21
`22
`23
`24
`10:12:53 25
`Page 5 to 8 of 79
`
`A.
`
`I'm a fellow for the Association of Computer Machinery.
`
`That's the ACM. They're a a flagship organization for Computing
`
`Professionals.
`
`It's a fellow and honor that they give their best 30 or
`
`40 people a year. It's like the top one percent.
`
`I'm a member of the Editorial Board for a class of
`
`publications, The Communications of the ACM.
`
`I'm actually the Chair of SIGACT. That's the ACM
`
`special interest group on algorithms and computation theories.
`
`There are some various groups within the ACM. This is the main
`
`one for algorithms.
`
`And, of course, I also do other sorts of organizational
`
`work.
`
`I'm a member of various editorial boards. I serve
`
`Let's talk about some of the awards that you received for
`
`Can you explain what is being shown here on the slide?
`
`Right.
`
`So, again, I'm an ACM Fellow. And, again, that's given
`
`to like 30 or 40 people a year and given to something like the
`
`top one percent.
`
`My citations -- the reason I've been given this award
`
`or honor, was for contributions to coding theories, hashing out
`
`algorithms and data structures, and networking algorithms.
`
`2 of 31 sheets
`
`1
`2
`3
`where I got a Ph.D. in Computer Science in 1996.
`4 Q.
`10:09:03 5
`appointment background?
`6
`A.
`7
`8
`9
`10:09:16 10
`11
`12
`13
`14
`10:09:38 15
`16
`17
`18
`19
`10:09:55 20
`Fountain, eHarmony, Akamai, and so on.
`21 Q.
`22
`A.
`23
`networks and communication.
`24 Q.
`10:10:09 25
`project you were working on?
`02/01/2018 12:05:57 PM
`
`served as Area Dean For Computer Science. That's the equivalent
`
`of a Department Chair, and it's called area Dean there, from
`
`2010 to 2013.
`
`I've also done, while I've been employed at Harvard, a
`
`variety of consulting. Some of it legal expert work, some
`
`technology work for companies such as Microsoft, Digital
`
`And at Harvard, what does your research focus on?
`
`My research focus is on algorithms, particularly for
`
`Can you describe some of the projects or at least one of the
`
`
`
`Case 1:16-cv-00455-RGA Document 394-1 Filed 02/07/18 Page 4 of 21 PageID #: 29733
`9
`
`11
`
`1
`2
`3
`4
`10:13:10 5
`6
`7
`8
`9
`10:13:26 10
`11
`12
`well.
`13 Q.
`14
`you teach at Harvard?
`10:13:38 15
`A.
`16
`17
`18
`19
`10:13:51 20
`21
`22
`23
`24
`10:14:13 25
`
`Again, those are my main areas of research.
`
`I've run several best paper awards, including the best
`
`paper award. And, in general, for the Institute of Electrical
`
`and Electronics Engineers Information Theory Society for my work
`
`on codings.
`
`I've won the test of time award. That's an award
`
`looking back say like ten years for conventional papers, so my
`
`limitations of coding algorithms within network systems.
`
`I received the National Science Foundation Career
`
`Award. That was for, you know, young faculty members starting
`
`out and doing research. I have several other NSF grants as
`
`And can you please tell the Court what kind of courses do
`
`Sure.
`
`So I teach the undergraduate algorithms course. I
`
`pretty much taught that every year at Harvard since 1999, except
`
`for two years where I taught sort of a special honors versions
`
`of that course.
`
`I also teach graduate courses. I tend to revolve two
`
`graduate courses, one is on Randomized Algorithms and
`
`Probabilistic Analysis. What the advanced algorithms responses
`
`were. I've actually written a textbook -- one of the main
`
`textbooks on the subject.
`
`My other graduate level class is specifically on
`
`And in forming that analysis, what level of the skill did
`
`Right.
`
`So to one of ordinary skill in the art, I used the
`
`following definitions.
`
`I assumed it was someone who had a Bachelor's Degree in
`
`Computer Science or a related filed. Plus, some additional work
`
`or experience beyond just the Undergraduate Degree, which would
`
`be either a two or more years of industry experience or an
`
`advanced degree in computer science or related field, or
`
`Master's Degree.
`
`Although, you know, my opinions wouldn't change given
`
`1
`the function that was recited in the Court's claim construction.
`2 Q.
`3
`you apply?
`4
`A.
`10:15:47 5
`6
`7
`8
`9
`10:16:06 10
`11
`12
`13
`14
`the small variations in this definition.
`10:16:21 15 Q.
`16
`someone take to get a Bachelor's Degree in Computer Science?
`17
`A.
`18
`19
`10:16:39 20
`21
`22
`23
`graphics or other areas, depending on their interests.
`24 Q.
`10:16:55 25
`understanding of computer science?
`
`And based on this level of skill, what type of courses does
`
`Like I said, you know, for instance at Harvard, you would be
`
`required to take some of the programming courses, a number of
`
`courses related to algorithms and data structures.
`
`You would also be expected to take courses in some
`
`number of advanced subjects, such as networking, for example.
`
`Artificial intelligence would be another category. Perhaps
`
`So is it fair to say that they would have a basic
`
`1
`2
`3
`4
`10:14:33 5
`6
`7
`8
`9
`10:14:45 10
`11
`BY MR. HANNAH:
`12 Q.
`13
`for today?
`14
`A.
`10:15:00 15
`16
`17
`18
`19
`10:15:15 20
`21
`22
`23
`24
`10:15:32 25
`3 of 31 sheets
`
`So, Doctor, can you briefly tell us what was your assignment
`
`So my assignment for today was -- I understand there's an
`
`issue in particular about how algorithms are described and help
`
`me understand the descriptions of algorithms to the extent that
`
`I need to present background or answer questions related to --
`
`to help you understand algorithms. I appear to do that.
`
` Specifically, I understand that there was one of
`
`the claim terms, Term 4.
`
`And, so, I examined Claim 4 and looked at the
`
`corresponding function and structure on, and why the Court's
`
`claim construction, in this particular -- I think one of the
`
`main issues is to determine whether the algorithm description at
`
`Column 5 of the '344 patent contains sufficient structure for
`
`1
`2
`3
`4
`10:17:08 5
`used.
`6 Q.
`7
`A.
`8
`9
`10:17:26 10
`11
`12
`13
`14
`10:17:40 15
`16
`17
`18
`corresponding output data.
`19 Q.
`10:17:54 20
`used?
`21
`A.
`22
`23
`24
`10:18:07 25
`Page 9 to 12 of 79
`
`10
`
`12
`
`A.
`
`Yes. Someone at this level would certainly have a good
`
`understanding of computer science.
`
`And, in particular, would have an understanding or
`
`experience with algorithms and how algorithms are described and
`
`algorithms for networks and communications.
`
`This is a class in 2002, current research and current
`
`research papers, a lot of discussion with what the recent work
`
`is on. And the goal there is for the graduate students to enter
`
`the class by doing projects that can lead to research
`
`publications, and do research in general.
`
`MR. HANNAH: Your Honor, at this time, we would like to
`
`tender Dr. Mitzenmacher as an expert in networking and
`
`algorithms?
`
`THE COURT: Yes. Okay.
`
`And, so, what is an algorithm?
`
`An algorithm is really a process to accomplish some task,
`
`you know, a sequence of steps. It's often like if you were to
`
`use metaphor analogy we use for interesting students -- we say
`
`it's just like a recipe.
`
`So just as you would have -- in a recipe, you have
`
`ingredients, such as like flour, and sugar, and chocolate. And
`
`those are your inputs and your outputs that wind up being in the
`
`cake.
`
`Except in computer science, it's like a recipe, but
`
`your inputs are the data, right, so you have some types of data
`
`elements that are your inputs, and you're producing some sort of
`
`What's an example of an algorithm that would be commonly
`
`Yes. So what I teach in my undergraduate class, and some
`
`variations of it are used on a regular basis, it would be like a
`
`shortest path algorithm.
`
`A shortest path algorithm would be like if you go to
`
`Google Map and you say, I would like directions from here to
`
`02/01/2018 12:05:57 PM
`
`
`
`Case 1:16-cv-00455-RGA Document 394-1 Filed 02/07/18 Page 5 of 21 PageID #: 29734
`13
`
`15
`
`there.
`
`And what Google will return to you is -- they will give
`
`the shortest path in terms of finding the quickest route for you
`
`to get there. In fact, it will find you sort a mix of several
`
`quickest routes.
`
`And as you can imagine, they have a lot of data as to
`
`maybe the individual distances and so on. That's the different
`
`points in the network. And then they have to use that basic
`
`information to compute this, you know, larger scale information
`
`What's an example of an algorithm taught in interim classes
`
`1
`2
`3
`4
`10:18:23 5
`6
`7
`8
`9
`10:18:38 10
`of what is the shortest path between these two points.
`11 Q.
`12
`for algorithm?
`13
`A.
`14
`10:18:53 15
`16
`17
`18
`19
`10:19:12 20
`21
`22
`23
`24
`10:19:31 25
`
`All right.
`
`So when we teach -- in other classes as well, one of
`
`the ways likening this to this recipe description that I gave
`
`before is, when we sort of teach algorithms to newcomers, one
`
`example we sometimes use is, well, how do you make a peanut
`
`butter and jelly sandwich?
`
`And the idea there is that you can give a very natural
`
`high-level description, right, you start with two pieces of
`
`bread, your peanut and jelly, and then you take the peanut
`
`butter, spread it on one side of one of the pieces of bread.
`
`Take the jelly, spread it on one side of the other piece of
`
`bread. You put the peanut butter side and the jelly side
`
`together and then you have your sandwich.
`
`Yes, that's my understanding.
`
`And you understand there's a corresponding limitation in the
`
`Yes.
`
`Can you understand that in order to satisfy a
`
`1
`for connecting to the identified broadcast channel?
`2
`A.
`3 Q.
`4
`'966 patent?
`10:20:58 5
`A.
`6 Q.
`7
`means-plus-function, you need to identify the function and the
`8
`structure, is that right?
`9
`A.
`10:21:08 10 Q.
`11
`12
`the identified broadcast channel, do you see that?
`13
`A.
`14 Q.
`10:21:22 15
`A.
`16
`17
`18
`19
`10:21:37 20
`21
`22
`23
`24
`10:21:53 25
`
`Yes, that's my understanding.
`
`So let's take a look at the function.
`
`So the function has been construed as, connecting to
`
`Yes.
`
`What's your understanding of what type of function that is?
`
`That's a very high level of function. We're talking about
`
`making a connection within a network, a particular type of
`
`connection.
`
`They identify a broadcast channel. But, again, that's
`
`a very high level of function, one that people understand
`
`readily or skill in the art, because they have to deal with
`
`managing connections to the networks. They would have seen that
`
`and understood that.
`
`And, so, one would expect that, you know, the
`
`corresponding description of an algorithm to do that would be a
`
`correspondingly high level.
`
`And that is a, you know, a natural way to describe an
`
`algorithm for making a peanut butter and jelly sandwich.
`
`14
`
`16
`
`And, so, you keep mentioning high-level function.
`
`Can you explain what a high-level function is versus a
`
`low-level function?
`A.
`
`Yes.
`
`So, if we're talking about a high-level function, by
`
`connecting to the identified broadcast channel, you would expect
`
`to see stuff like, well, you know, who do I connect to first?
`
`And then what sort of individual connections do I make among the
`
`discipline.
`
`If you are talking about a things at a lower level,
`
`okay, where we're dealing with something that needed to be
`
`implemented on a chip, right, then you might need to get into
`
`issues of like, well, what gets put where, and what does -- this
`
`whole population needs to occur.
`
`But here we're talking about a high level networking
`
`function where there is already in place various protocols such
`
`as TCP/IP and so on for individual connections -- in their
`
`sockets, TCP/IP and so on.
`
`And, so, people understand the methodologies for doing
`
`connections. You would expect a description to take into
`
`account that people have this knowledge.
`
`THE COURT: Mr. Hannah, can you hold on just a second?
`
`MR. HANNAH: Sure.
`
`(Pause).
`
`MR. HANNAH: Is this better? Is it okay?
`
`4 of 31 sheets
`
`1
`2
`3
`4
`10:19:47 5
`6
`7
`8
`9
`10:20:03 10
`11
`12
`13
`detail in the algorithm.
`14 Q.
`10:20:18 15
`required for an algorithm?
`16
`A.
`17
`18
`19
`10:20:29 20
`21
`22
`high-level description.
`23 Q.
`24
`10:20:47 25
`02/01/2018 12:05:57 PM
`
`And then when we go into that we say, well, okay, when
`
`you're implementing that, there may be additional details that
`
`you need to provide the computer, because the computer is
`
`somewhat literal. And you don't really even need them to
`
`describe the algorithm, but there are things to think about.
`
`For instance, when you're talking about peanut butter,
`
`you would think about looking in the jar first. And certainly
`
`one of skill in the art in making sandwiches would understand
`
`that you need to open the jar to make the peanut butter and
`
`jelly sandwich, without having to actually put that level of
`
`So how does the function determine the level of detail
`
`Right.
`
`So when we're describing algorithms, we tend to focus
`
`on the high level, like what's the main idea, what's the -- the
`
`important steps that you need to know.
`
`And, in particular, when we're talking about a
`
`high-level functionality, you expect to see a corresponding
`
`So let's turn to Claim 13.
`
`And you understand that Claim 13 is one of the claims
`
`that we're here to talk about today, and specifically the means
`
`1 Q.
`2
`3
`4
`10:22:01 5
`6
`7
`8
`9
`10:22:17 10
`11
`12
`13
`14
`10:22:33 15
`16
`17
`18
`19
`10:22:51 20
`21
`22
`23
`24
`10:22:58 25
`Page 13 to 16 of 79
`
`
`
`Case 1:16-cv-00455-RGA Document 394-1 Filed 02/07/18 Page 6 of 21 PageID #: 29735
`17
`
`19
`
`1
`THE COURT: It sounds better here.
`2
`Go ahead, Mr. Hannah.
`3
`MR. HANNAH: Sure.
`4
`BY MR. HANNAH:
`10:23:48 5 Q.
`Sir, I believe you were just describing the high-level
`6
`function versus the low-level function.
`7
`Let's turn to the Court's claim construction here.
`8
`Can you explain what is being shown here?
`9
`And we can focus on the '344 patent just for brevity
`10:24:01 10
`11
`12
`13
`14
`10:24:20 15
`16
`know, what sort of steps are necessary.
`17 Q.
`So, if we look at -- let's focus -- well, for the record, if
`18
`I talk about the Column 5 algorithm, you understand that I'm
`19
`talking about the algorithm at Column 5, Lines 33 through 55?
`10:24:41 20
`A.
`21 Q.
`And if I talk about the Column 18 algorithm, that I'm
`22
`talking about the first algorithm that is found in the Court's
`23
`claim construction here?
`24
`A.
`10:24:50 25 Q.
`
`Yes.
`
`Yes, I would understand that.
`
`So let's talk about the Column 5 algorithm.
`
`here.
`A.
`
`So the Court in the claim construction has picked out two
`
`descriptions of the algorithm.
`
`One given at around Column 5. And one starting around
`
`the end of Column 17 or beginning Column 18.
`
`And further in the description it has described, you
`
`Sure. Again, starting I guess around Line 33 or so, the
`
`language from the specification is the following.
`
`"Thus, the process of connecting to the broadcast
`
`channel includes locating the broadcast channel, identifying the
`
`neighbors of the connecting computer, and then connecting to
`
`each identified neighbor."
`
`So it's broken down that that is connecting to the
`
`identified broadcast channel into three corresponding set of
`
`steps.
`
`And then it further describes or explains those
`
`substeps in a manner so that one of skill in the art can
`
`understand what was needed or what was required.
`
`In particular, you'll see what follows next is, "Each
`
`computer is aware of one or more "portal computers" through
`
`which that computer may locate the broadcast channel. A seeking
`
`computer locates the broadcast channel by contacting the portal
`
`computers until it finds one that is currently fully connected
`
`to the broadcast channel."
`
`So here we have this first step right at this location
`
`step. And the specification, besides just stating that there's
`
`1
`2
`3
`4
`10:28:03 5
`6
`7
`8
`9
`10:28:22 10
`11
`12
`13
`14
`10:28:38 15
`16
`further needs to be made in order to have the connection.
`17 Q.
`Can you explain how these steps of the algorithm correspond
`18
`to the identified function that is being shown on this Slide 11?
`19
`A.
`10:28:53 20
`21
`22
`23
`24
`10:29:11 25
`Page 17 to 20 of 79
`
`a location step, gives the necessary sufficient position to give
`
`the understanding of what one needs to do that location step by
`
`use of the portal computer.
`
`It then says, "The found portal computer then directs
`
`the identifying of four computers, (i.e., to be the seeking
`
`computer's neighbors), to which the seeking computer is to
`
`connect."
`
`Then, again, that's some sort of description of the
`
`second step of identifying the neighbor for the connecting
`
`computer. And, again, this is by way of the portal computer,
`
`and the portal computer does the direction it's identifying.
`
`Then "Each of these four computers then cooperates with
`
`the seeking computer to effect the connecting of the seeking
`
`computer to the broadcast channel," right.
`
`Again, that's an elaboration of the description of the
`
`last step is that the connecting computer then has to connect to
`
`the identified neighbors.
`
`So you can see that, you know, what the steps are here
`
`is, again, a description of the algorithm first by breaking it
`
`down into three substeps. And then, of course, we get an
`
`elaboration of these three substeps in a way that, again, one of
`
`skill in the art would understand what needed to be done or what
`
`needed to be accomplished, as given specifically in the
`
`1
`2
`3
`4
`10:26:38 5
`6
`7
`8
`9
`10:26:54 10
`11
`12
`13
`14
`10:27:10 15
`16
`17
`18
`19
`10:27:32 20
`21
`22
`23
`24
`specification.
`10:27:48 25 Q.
`
`Can you explain specifically why this disclosed algorithm is
`
`20
`
`sufficient to define the structure and make the bounds of the
`claims understandable?
`A.
`
`All right.
`
`So, again, I think this is understandable. This is a
`
`description. I can imagine giving it to a student, say a
`
`high-level undergraduate who is taking a networking and
`
`algorithms course and who has some experience to say that is in
`
`the framework, this is the description.
`
`And they could build it from there, because it
`
`describes, in sufficient detail, what is needed and what is
`
`required.
`
`In particular, the relationship with the portal
`
`computers. What steps the portal computers partake in, in order
`
`to generate a connection to the broadcast channel.
`
`And, in particular, the individual connections that
`
`Right.
`
`I mean, clearly they correspond. They explain how to
`
`accomplish the corresponding function. And, again, this is not
`
`surprising that the connection function is a high-level
`
`functionality, and as one would expect, with computers
`
`generally, but also in particular given the high-level of
`
`functionality, it does describe it in high-level terms of making
`
`02/01/2018 12:05:57 PM
`
`18
`And do you have a copy of the '344 patent with you?
`
`Yes, I do.
`
`1
`2
`A.
`3 Q.
`So looking at the Column 5 algorithm, can you please explain
`4
`how one of skill in the art would understand the algorithm that
`10:25:07 5
`is described there?
`6
`A.
`7
`8
`9
`10:25:27 10
`11
`12
`13
`14
`10:25:42 15
`16
`17
`18
`19
`10:26:01 20
`21
`22
`23
`24
`10:26:18 25
`5 of 31 sheets
`
`
`
`Case 1:16-cv-00455-RGA Document 394-1 Filed 02/07/18 Page 7 of 21 PageID #: 29736
`21
`
`23
`
`various connections and working through the portal computer.
`
` But these are high-level descriptions that, again, one
`
`Well, again, I think this lays out in clear sequence the
`
`steps that needs to be taken, and the manner in which they need
`
`to be taken by way of the portal computer.
`
`Now, when giving it an implementation, one would expect
`
`that there would be further work in the implementation level to
`
`make these connections.
`
`But, again, the patent describes, for instance, how
`
`those things are done. How to make an individual paired
`
`connection on a TCP/IP connection or other network protocols.
`
`So, there's other information in the patent that when
`
`you think about, okay, well, if I'm going to get to a more
`
`detailed level, what I need to do.
`
`There's information there within the patent and one of
`
`skill in the art would already know how to do these things based
`
`1
`2
`3
`of skill in the art would readily understand.
`4 Q.
`Why don't you need any additional details other than those
`10:29:28 5
`as described in the Column 5 algorithm to -- to make it
`6
`understandable?
`7
`A.
`8
`9
`10:29:46 10
`11
`12
`13
`14
`10:30:02 15
`16
`17
`18
`19
`10:30:17 20
`21
`on this description.
`22 Q.
`So is it fair to say that your students would understand the
`23
`bounds of this algorithm and the bounds of the claims in light
`24
`of this algorithm on Column 5?
`10:30:33 25
`A.
`
`Yes, they would both understand how to do it and they
`
`And because it's an entire embodiment, it's not just
`
`details on the connecting systems, but details on certain
`
`interactions with the other algorithm steps required other
`
`claims or other parts of this patent and the other related
`
`patents, right?
`
`So this is, again, going into more detail, but it's
`
`more detail than is needed for than just the connecting step and
`
`1
`2
`3
`4
`10:32:28 5
`6
`7
`8
`the embodiment.
`9 Q.
`Can you please explain -- well, you said this actually talks
`10:32:44 10
`about a series of algorithms -- can you explain what you mean in
`11
`relation to Figure 8?
`12
`A.
`13
`14
`10:32:57 15
`16
`17
`18
`19
`10:33:19 20
`21
`22
`23
`24
`10:33:42 25
`
`Okay.
`
`So there are various aspects of that.
`
`So one of the steps -- the algorithm where it talked
`
`about before was focused on issues related to the large regime.
`
`So here this algorithm, because it's s specific embodiment going
`
`into more detail, discusses issues related to both what has been
`
`called a small regime and a large regime.
`
`It goes into some detail about if you're in the small
`
`regime, which corresponds, for instance, to the contacts of zero
`
`or 805. It's the 805 block in this chart.
`
`That what it will do when -- say you're the first
`
`person to come, so you are the broadcast channel. After there
`
`are various sorts of specific error corrections or dealing with
`
`faults or errors that are set up in this diagram, and in
`
`22
`
`24
`
`understand what the corresponding bounds or limitations are by
`
`way of the portal computer. The limitations are by how the
`
`seeking computer interacts with the portal computer in order to
`
`identify the neighbors, and the bounds about what sort of
`
`connections are subsequently made to connect to the broadcast
`
`to it.
`
`1
`2
`3
`4
`10:30:53 5
`6
`channel.
`7 Q.
`Let's turn to the next algorithm that we've identified as
`8
`the Column 18 algorithm.
`9
`And you've got the '344 patent up there. You can refer
`10:31:08 10
`11
`12
`13
`14
`10:31:34 15
`16
`17
`18
`19
`10:31:49 20
`21
`22
`23
`24
`10:32:11 25
`02/01/2018 12:05:57 PM
`
`subsequent diagrams, one of the things here is the setting of
`
`the connect time. That's 802.
`
`So the use of the setting of the connect time that is
`
`described in the patent, it deals with issues relating to timing
`
`of messages, which is in Column 18, around Line 29.
`
`It says, in Block 802, the regime sets the connect time
`
`to the current time. The connect time is used to identify the
`
`instance of the process that is connected through this external
`
`port. One process may connect to a broadcast channel of a
`
`certain channel type and channel instance using one call-in port
`
`and then disconnects, and another process may then connect to
`
`that same broadcast channel using the same call-in port. Before
`
`the other process becomes fully connected, another process may
`
`try to communicate with it thinking it is the fully connected
`
`old process.
`
`So using the connect time as a method of an error
`
`correction, or a way of handling possible fault issues.
`
`But there are other ways that that can be handled. And
`
`there are other sort of fault situations that, again, are not
`
`needed, and that one of skill in the art would understand that
`
`Yes, it's describing the connecting function, but it's also
`
`6 of 31 sheets
`
`1
`2
`3
`4
`10:34:03 5
`6
`7
`8
`9
`10:34:21 10
`11
`12
`13
`14
`10:34:36 15
`16
`17
`18
`19
`10:34:56 20
`21
`there are various ways to accomplish those goals.
`22 Q.
`So in looking at the Column 18 algorithms, is it fair to say
`23
`that it describes the connecting to the identified broadcast
`24
`function in addition to other functions?
`10:35:10 25
`A.
`Page 21 to 24 of 79
`
`Can you please explain what is being described in the
`Column 18 algorithm?
`A.
`
`Sure.
`
`At the bottom of Column 17, and the beginning of 18, it
`
`sort of lays out what Figures 8 through 345 are. And, in fact,
`
`the sequences they originally talked about.
`
`It says, Figures 8 through 34 are flow diagrams
`
`illustrating the processing of the broadcast component in one
`
`embodiment.
`
`Figure 8 is the flow diagram illustrating the
`
`processing of the connect regime in one embodiment.
`
`What we see here is a description of Figure 8, and
`
`subsequent figures, are going to describe a specific embodiment.
`
`And one would expect that the specific embodiment would provide
`
`additional details.
`
`
`
`Case 1:16-cv-00455-RGA Document 394-1 Filed 02/07/18 Page 8 of 21 PageID #: 29737
`25
`
`27
`
`describing an embodiment. It's describing it also in the
`
`context of several other functions that need to take place for
`
`So, in summary, Doctor, can you please tell us, what is your
`
`it now stands?
`A.
`
`I think the Court's claim construction is fine. I think it
`
`perfectly answers the question and lays out very clearly what
`
`the means-plus-function are and what's required of it.
`
`I think that these choices that were used for the
`
`Column 5 algorithm and the Column 18 algorithm. They both
`
`represent interesting methods or algorithm descriptions of how
`
`to achieve the functionality as required in the claim
`
`And is it your opinion that the Column 5 algorithm on its
`
`1
`2
`3
`additional claims within this patent and the other patents.
`4 Q.
`10:35:29 5
`opinion with regard to the Court's claim construction order as
`6
`7
`8
`9
`10:35:49 10
`11
`12
`13
`14
`construction.
`10:36:07 15 Q.
`16
`own is sufficient to define the structure and make the claim
`17
`18
`19
`10:36:25 20
`21
`22
`23
`24
`10:36:40 25
`
`understandable for both the '344 patent and the '966 patent?
`A.
`
`Yes, I believe that that description that I've gone through
`
`-- and, again, the subst