`
`
`1
`
`
`
`UNITED STATES DISTRICT COURT
`FOR THE DISTRICT OF DELAWARE
`
`CA NO. 16-453-RGA,
`16-454-RGA,
`16-455-RGA
`
`January 29, 2018
`
`: : : : : : : :
`
`
`
`ACCELERATION BAY LLC,
`
`Plaintiff,
`
` v.
`
`ACTIVISION BLIZZARD INC.,
`
`10:05 o'clock a.m.
`
`
`
`: :
`
`Defendant,
`.............................
`
`TRANSCRIPT OF MOTION FOR CLARIFICATION
`BEFORE THE HONORABLE RICHARD G. ANDREWS
`UNITED STATES DISTRICT JUDGE
`
`APPEARANCES:
`
`For Plaintiff:
`
`POTTER, ANDERSON & CORROON
`BY: PHILIP A. ROVNER, ESQ
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 2 of 79 PageID #: 24596
`
`
`2
`
`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
`
`MORRIS NICHOLS ARSHT & TUNNELL
`BY: STEPHEN J. KRAFTSCHIK, ESQ
`-and-
`WINSTON & STRAWN LLP
`BY: MICHAEL M. MURRAY, ESQ
`BY: DAVID P. ENZMINGER, ESQ
`
`For Defendant:
`
`Court Reporter:
`
`LEONARD A. DIBBS
`Official Court Reporter
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 3 of 79 PageID #: 24597
`
`
`3
`
` P R O C E E D I N G S
`
`(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 in Acceleration 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.
`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.
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:05:15
`
`10:05:33
`
`10:05:41
`
`10:05:53
`
`10:06:07
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 4 of 79 PageID #: 24598
`
`
`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.
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:06:17
`
`10:06:27
`
`10:06:39
`
`10:06:45
`
`10:06:56
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 5 of 79 PageID #: 24599
`
`
`5
`
`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
`
`BY MR. HANNAH:
`Q.
`Good morning, Dr. Mitzenmacher.
`A.
`Good morning.
`Q.
`Can you please start out by describing your educational
`background?
`A.
`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 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:07:47
`
`10:07:59
`
`10:08:22
`
`10:08:30
`
`10:08:46
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 6 of 79 PageID #: 24600
`
`
`6
`
`of Math Studies, sort of the equivalent of a Master's.
`I then went to the University of California at Berkeley
`where I got a Ph.D. in Computer Science in 1996.
`Q.
`Can you please tell us about your educational and your
`appointment background?
`A.
`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
`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
`Fountain, eHarmony, Akamai, and so on.
`Q.
`And at Harvard, what does your research focus on?
`A.
`My research focus is on algorithms, particularly for
`networks and communication.
`Q.
`Can you describe some of the projects or at least one of the
`project you were working on?
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:09:03
`
`10:09:16
`
`10:09:38
`
`10:09:55
`
`10:10:09
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 7 of 79 PageID #: 24601
`
`
`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
`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
`answer.
`Q.
`What type of organizations are you involved with?
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:10:20
`
`10:10:40
`
`10:10:51
`
`10:11:09
`
`10:11:21
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 8 of 79 PageID #: 24602
`
`
`8
`
`I'm a fellow for the Association of Computer Machinery.
`A.
`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
`regularly on program committees for conferences and so on.
`Q.
`Let's talk about some of the awards that you received for
`your work.
`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.
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:11:39
`
`10:12:01
`
`10:12:13
`
`10:12:26
`
`10:12:53
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 9 of 79 PageID #: 24603
`
`
`9
`
`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
`well.
`Q.
`And can you please tell the Court what kind of courses do
`you teach at Harvard?
`A.
`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
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:13:10
`
`10:13:26
`
`10:13:38
`
`10:13:51
`
`10:14:13
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 10 of 79 PageID #: 24604
`
`
`10
`
`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.
`BY MR. HANNAH:
`Q.
`So, Doctor, can you briefly tell us what was your assignment
`for today?
`A.
`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 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:14:33
`
`10:14:45
`
`10:15:00
`
`10:15:15
`
`10:15:32
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 11 of 79 PageID #: 24605
`
`
`11
`
`the function that was recited in the Court's claim construction.
`Q.
`And in forming that analysis, what level of the skill did
`you apply?
`A.
`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
`the small variations in this definition.
`Q.
`And based on this level of skill, what type of courses does
`someone take to get a Bachelor's Degree in Computer Science?
`A.
`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
`graphics or other areas, depending on their interests.
`Q.
`So is it fair to say that they would have a basic
`understanding of computer science?
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:15:47
`
`10:16:06
`
`10:16:21
`
`10:16:39
`
`10:16:55
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 12 of 79 PageID #: 24606
`
`
`12
`
`Yes. Someone at this level would certainly have a good
`A.
`understanding of computer science.
`And, in particular, would have an understanding or
`experience with algorithms and how algorithms are described and
`used.
`Q.
`And, so, what is an algorithm?
`A.
`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
`corresponding output data.
`Q.
`What's an example of an algorithm that would be commonly
`used?
`A.
`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
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:17:08
`
`10:17:26
`
`10:17:40
`
`10:17:54
`
`10:18:07
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 13 of 79 PageID #: 24607
`
`
`13
`
`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
`of what is the shortest path between these two points.
`Q.
`What's an example of an algorithm taught in interim classes
`for algorithm?
`A.
`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.
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:18:23
`
`10:18:38
`
`10:18:53
`
`10:19:12
`
`10:19:31
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 14 of 79 PageID #: 24608
`
`
`14
`
`And that is a, you know, a natural way to describe an
`algorithm for making a peanut butter and jelly sandwich.
`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
`detail in the algorithm.
`Q.
`So how does the function determine the level of detail
`required for an algorithm?
`A.
`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
`high-level description.
`Q.
`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 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:19:47
`
`10:20:03
`
`10:20:18
`
`10:20:29
`
`10:20:47
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 15 of 79 PageID #: 24609
`
`
`15
`
`for connecting to the identified broadcast channel?
`A.
`Yes, that's my understanding.
`Q.
`And you understand there's a corresponding limitation in the
`'966 patent?
`A.
`Yes.
`Q.
`Can you understand that in order to satisfy a
`means-plus-function, you need to identify the function and the
`structure, is that right?
`A.
`Yes, that's my understanding.
`Q.
`So let's take a look at the function.
`So the function has been construed as, connecting to
`the identified broadcast channel, do you see that?
`A.
`Yes.
`Q.
`What's your understanding of what type of function that is?
`A.
`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.
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:20:58
`
`10:21:08
`
`10:21:22
`
`10:21:37
`
`10:21:53
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 16 of 79 PageID #: 24610
`
`
`16
`
`Q.
`
`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?
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:22:01
`
`10:22:17
`
`10:22:33
`
`10:22:51
`
`10:22:58
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 17 of 79 PageID #: 24611
`
`
`17
`
`THE COURT: It sounds better here.
`Go ahead, Mr. Hannah.
`MR. HANNAH: Sure.
`BY MR. HANNAH:
`Q.
`Sir, I believe you were just describing the high-level
`function versus the low-level function.
`Let's turn to the Court's claim construction here.
`Can you explain what is being shown here?
`And we can focus on the '344 patent just for brevity
`
`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
`know, what sort of steps are necessary.
`Q.
`So, if we look at -- let's focus -- well, for the record, if
`I talk about the Column 5 algorithm, you understand that I'm
`talking about the algorithm at Column 5, Lines 33 through 55?
`A.
`Yes.
`Q.
`And if I talk about the Column 18 algorithm, that I'm
`talking about the first algorithm that is found in the Court's
`claim construction here?
`A.
`Yes, I would understand that.
`Q.
`So let's talk about the Column 5 algorithm.
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:23:48
`
`10:24:01
`
`10:24:20
`
`10:24:41
`
`10:24:50
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 18 of 79 PageID #: 24612
`
`
`18
`
`And do you have a copy of the '344 patent with you?
`Yes, I do.
`A.
`So looking at the Column 5 algorithm, can you please explain
`Q.
`how one of skill in the art would understand the algorithm that
`is described there?
`A.
`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 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:25:07
`
`10:25:27
`
`10:25:42
`
`10:26:01
`
`10:26:18
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 19 of 79 PageID #: 24613
`
`
`19
`
`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
`specification.
`Q.
`Can you explain specifically why this disclosed algorithm is
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:26:38
`
`10:26:54
`
`10:27:10
`
`10:27:32
`
`10:27:48
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 20 of 79 PageID #: 24614
`
`
`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
`further needs to be made in order to have the connection.
`Q.
`Can you explain how these steps of the algorithm correspond
`to the identified function that is being shown on this Slide 11?
`A.
`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
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:28:03
`
`10:28:22
`
`10:28:38
`
`10:28:53
`
`10:29:11
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 21 of 79 PageID #: 24615
`
`
`21
`
`various connections and working through the portal computer.
`But these are high-level descriptions that, again, one
`of skill in the art would readily understand.
`Q.
`Why don't you need any additional details other than those
`as described in the Column 5 algorithm to -- to make it
`understandable?
`A.
`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
`on this description.
`Q.
`So is it fair to say that your students would understand the
`bounds of this algorithm and the bounds of the claims in light
`of this algorithm on Column 5?
`A.
`Yes, they would both understand how to do it and they
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:29:28
`
`10:29:46
`
`10:30:02
`
`10:30:17
`
`10:30:33
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 22 of 79 PageID #: 24616
`
`
`22
`
`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
`channel.
`Q.
`Let's turn to the next algorithm that we've identified as
`the Column 18 algorithm.
`And you've got the '344 patent up there. You can refer
`
`to it.
`
`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.
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:30:53
`
`10:31:08
`
`10:31:34
`
`10:31:49
`
`10:32:11
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 23 of 79 PageID #: 24617
`
`
`23
`
`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
`the embodiment.
`Q.
`Can you please explain -- well, you said this actually talks
`about a series of algorithms -- can you explain what you mean in
`relation to Figure 8?
`A.
`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
`
`1 2 3 4 5 6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`10:32:28
`
`10:32:44
`
`10:32:57
`
`10:33:19
`
`10:33:42
`
`
`
`Case 1:16-cv-00455-RGA Document 384 Filed 02/01/18 Page 24 of 79 PageID #: 24618
`
`
`24
`
`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
`there are various ways to accomplish those goals.
`Q.
`So in looking at the Column 18 algorithms, is it fair to say
`that it describes the connecting to the identified broadcast
`function in addition to other functions?
`A.
`Yes, it's describing the connecting funct