throbber
Case 1:14-cv-02396-PGG-MHD Document 153-12 Filed 06/28/19 Page 1 of 96
`Case 1:14-cv-02396—PGG-MHD Document 153-12 Filed 06/28/19 Page 1 of 96
`
`EXHIBIT K
`
`EXHIBIT K
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 153-12 Filed 06/28/19 Page 2 of 96
`
`Deposition of:
`Michael Mitzenmacher , Ph.D.
`June 24, 2019
`
`In the Matter of:
`Network-1 Technologies, Inc. Vs.
`Google LLC And Youtube, LLC
`
`Veritext Legal Solutions
`800.808.4958 | calendar-dmv@veritext.com |
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 153-12 Filed 06/28/19 Page 3 of 96
`
` UNITED STATES DISTRICT COURT
`
` SOUTHERN DISTRICT OF NEW YORK
`
`Page 1
`
`- - - - - - - - - - - - - - - X
`
`NETWORK-1 TECHNOLOGIES, INC.,
`
` Plaintiff,
`
` v. 14 Civ. 2396 (PGG)
`
`GOOGLE LLC and YOUTUBE, LLC, 14 Civ. 9558 (PGG)
`
` Defendants.
`
`- - - - - - - - - - - - - - - X
`
` VOLUME I Pages 1-208
`
` VIDEO DEPOSITION OF MICHAEL D. MITZENMACHER, Ph.D.
`
` Monday, June 24, 2019, 9:19 a.m.
`
` Finnegan, Henderson, Farabow, Garrett & Dunner, LLP
`
` Two Seaport Lane
`
` Boston, Massachusetts 02210
`
` --- Reporter: Kimberly A. Smith, CRR, CRC, RDR ---
`
` Realtime Systems Administrator
`
` Veritext Legal Solutions
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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 1:14-cv-02396-PGG-MHD Document 153-12 Filed 06/28/19 Page 4 of 96
`
`Page 2
`
`Page 4
`
`1 APPEARANCES:
`
`23
`
` Russ August & Kabat
`4 By: Brian D. Ledahl, Esq.
`5 12424 Wilshire Boulevard, 12th Floor
`6 Los Angeles, CA 90025
`7 (310) 826-7474
`8 bledahl@raklaw.com
`9 for the Plaintiff;
`10
`11 Williams & Connolly LLP
`12 By: Samuel Bryant Davidoff, Esq.
`13 725 Twelfth Street, N.W.
`14 Washington, D.C. 20005
`15 (202) 434-5000
`16 sdavidoff@wc.com
`17 for the Defendants.
`18
`19 Also Present: Gayle Ashton, Video Operator
`20
`21
`22
`23
`24
`25
`
`1 THE VIDEO OPERATOR: Good morning.
`2 We are going on the record at 9:19 a.m. on June 24,
`3 2019. This is Media Unit 1 of the video recorded
`4 deposition of Michael Mitzenmacher, Ph.D., taken by
`5 counsel for defendant in the matter of Network-1
`6 Technologies, Inc. vs. Google LLC and YouTube, LLC.
`7 This deposition is being held at
`8 Finnegan Henderson, located at Two Seaport Lane,
`9 Boston, Massachusetts. My name is Gayle Ashton,
`10 and I'm the videographer from the firm Veritext.
`11 The court reporter is Kimberly Smith, also from
`12 Veritext.
`13 Will counsel please identify their
`14 appearances.
`15 MR. DAVIDOFF: Sam Davidoff on behalf --
`16 from Williams & Connolly on behalf of defendants
`17 Google and YouTube.
`18 MR. LEDAHL: And Brian Ledahl from Russ
`19 August & Kabat on behalf of the plaintiff.
`20 THE VIDEO OPERATOR: Will the court
`21 reporter please swear in the witness.
`22 MICHAEL D. MITZENMACHER, Ph.D.,
`23 having been satisfactorily identified by the
`24 production of his driver's license, and
`25 duly sworn by the court reporter, was deposed
`
`Page 5
`
`1 and testified as follows:
`2 EXAMINATION
`3 BY MR. DAVIDOFF:
`4 Q. Good morning, professor.
`5 A. Good morning.
`6 MR. DAVIDOFF: I would like to mark
`7 actually an exhibit, which I think we'll mark as
`8 Exhibit 1 to your deposition.
`9 (Mitzenmacher Exhibit 1 was marked
`10 for identification.)
`11 Q. And this is just a list of your prior
`12 litigation experience over the last five years that
`13 was provided to us.
`14 Is that familiar to you?
`15 A. Yes.
`16 Q. And really I thought this might just help
`17 refresh your memory because it looks like you've
`18 done a bit of testifying.
`19 A. Um-hum.
`20 Q. My question is if you could just tell me
`21 which of the -- not just these, which cases you've
`22 testified at trial in. And I thought maybe it would
`23 help to look at the list.
`24 A. Okay. My recollection is the France
`25 Telecom case, the Juniper vs. Palo Alto Networks
`2 (Pages 2 - 5)
`
`1 I N D E X
`
`Page 3
`
` WITNESS: Michael D. Mitzenmacher, Ph.D.
`
`23
`
`45
`
` EXAMINATION Page
`6 By Mr. Davidoff 5
`7 AFTERNOON SESSION
`8 By Mr. Davidoff 112
`9
`10 EXHIBITS FOR IDENTIFICATION:
`11 Mitzenmacher Description Page
`12 Exhibit 1 List of witness's litigation 5
`13 testimony
`14 Exhibit 2 Witness's 5/29/19 declaration 43
`15 Exhibit 3 U.S. Patent 7,831,438 58
`16 Exhibit 4 U.S. Patent 8,065,733 58
`17 Exhibit 5 3/27/15 Karypis declaration 63
`18 Exhibit 6 U.S. Patent 8,010,988 93
`19 Exhibit 7 U.S. Patent 8,205,237 93
`20 Exhibit 8 U.S. Patent 8,904,464 93
`21
`22
`23
`24 Original exhibits retained by reporter to be
`25 returned to Williams & Connolly
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 153-12 Filed 06/28/19 Page 5 of 96
`
`Page 6
`
`1 case. I think in the JobDiva, it wasn't at trial.
`2 There was a hearing. But I don't think . . .
`3 Q. Was that a Markman hearing?
`4 A. Yeah. I mean the Microsoft vs. Parallel
`5 Networks, those were IPRs. I don't know the legal
`6 thing, like the deposition counts as testimony.
`7 I mean it wasn't a trial. It was --
`8 Q. It was a deposition like this.
`9 A. Yeah. But I somehow understand for IPRs
`10 they're treated a little differently, but it wasn't
`11 a trial. I believe Finjan vs. Blue Coat, Finjan vs.
`12 Sophos. I guess one of the Realtime cases.
`13 Q. Had trial testimony?
`14 A. Yeah. I think the second Finjan vs. Blue
`15 Coat. And I think -- yeah. I think the second
`16 Finjan vs. Blue Coat and Finjan vs. Symantec --
`17 I have to look up -- I think those might have been
`18 combined at some later point because I think
`19 Symantec was buying Blue Coat, so . . .
`20 And, again, maybe the Apple vs. Caltech
`21 for the IPRs --
`22 Q. Sure.
`23 A. -- I was deposed, but . . .
`24 Oh, and Centripetal vs. Keysight.
`25 Q. Is that it?
`
`Page 7
`
`1 A. I think so.
`2 Q. And just let me make sure I have it. So
`3 you have France Telecom. And was that testimony in
`4 New York or California?
`5 A. Um --
`6 Q. Do you recall where the trial was?
`7 A. The trial was in California.
`8 Q. Okay.
`9 A. Yeah.
`10 Q. That's what I thought. I think the case
`11 got transferred at some point.
`12 A. Yeah. Sorry.
`13 Q. And Juniper v. Palo Alto, that was in
`14 Delaware?
`15 A. Yes.
`16 Q. JobDiva was in New York?
`17 A. Yes.
`18 Q. And the Finjan vs. Blue Coat, you recall
`19 testifying at two trials there?
`20 A. Yeah.
`21 Q. And both in San Francisco?
`22 A. Yeah.
`23 Q. And the Finjan v. Sophos, was that also in
`24 San Francisco?
`25 A. Yeah. Or that area. Might have been
`
`Page 8
`
`1 San Jose or . . .
`2 Q. Fair enough. Northern District of
`3 California?
`4 A. Yeah.
`5 Q. So three trials where you were testifying
`6 on behalf of Finjan?
`7 A. Yes. I guess -- yeah, I think there's been
`8 at least three -- well, the two Blue Coat and the
`9 Sophos/Symantec, yeah, I think that's three.
`10 Q. And Realtime, that was in the Eastern
`11 District of Texas?
`12 A. Yes.
`13 Q. When was that testimony, roughly?
`14 A. It was a few years ago. I'd have to look
`15 it up.
`16 Q. That's fine.
`17 A. 2016 or '17.
`18 Q. I'm sure it's written down somewhere.
`19 A. Yeah.
`20 Q. And Centripetal vs. Keysight, that was in
`21 Eastern District of Virginia?
`22 A. Yes.
`23 Q. And when was that testimony?
`24 A. That was late last year, probably -- might
`25 have been November, October/November last year.
`Page 9
`
`1 Q. That was a jury trial?
`2 A. Yes.
`3 Q. What was the result of that, do you recall?
`4 A. The result was -- I guess my understanding
`5 was, after I testified but before the trial was
`6 complete, the parties settled.
`7 Q. And do you have any cases that are
`8 scheduled for trial in the next year that you know
`9 of?
`10 A. Yes, possibly. I mean, like I think
`11 they've been scheduled. Whether they'll actually
`12 stay scheduled, I don't know.
`13 Q. Sure. Can you tell me what they are.
`14 A. I think the Finjan vs. ESET one here, and
`15 then -- let's see -- I think the Finjan vs. Juniper
`16 that's listed there.
`17 Q. So those two cases on this list are
`18 obviously still active. Are there any other cases
`19 on this list that are active? And I know maybe
`20 there's something on appeal or something, but that
`21 you know of is active?
`22 A. Yes. So as far as I know, the Caltech vs.
`23 Apple is still active.
`24 Q. Okay.
`25 A. I think some of the Acceleration Bay cases
`3 (Pages 6 - 9)
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 153-12 Filed 06/28/19 Page 6 of 96
`
`Page 10
`1 are still active, though I don't know their status.
`2 I guess something that's not on this
`3 list -- or I don't think it's on this list, which
`4 is, I guess, new, and so maybe that's why it's not
`5 on the list -- I don't remember when this was sent
`6 to you -- is I have some cases -- and I have to look
`7 up the name. It's for Brew House.
`8 It's a case involving the Telephone
`9 Consumer Protection Act, so it's not a patent case.
`10 And I don't know what the status of that is, if
`11 that's going to be planned to go to trial or what
`12 the timeline of that would be.
`13 Q. That was going to be my next question.
`14 And I understand these things are kind of moving
`15 targets.
`16 A. Yeah.
`17 Q. Is there anything else you're working on --
`18 any other cases you're working on that aren't listed
`19 on there?
`20 A. Let me think for a second. So there is
`21 that one. So not -- nothing else that I've given
`22 testimony on.
`23 Q. Is there anything else that you've been
`24 disclosed as an expert?
`25 A. I believe I might have already been
`
`Page 11
`1 disclosed for like an additional Finjan case versus,
`2 I think it's Bitdefender.
`3 Q. Are you also working on another Realtime
`4 case now?
`5 A. Yeah. So I think -- or they've asked me to
`6 do stuff, but I'm not sure I've been disclosed or I
`7 don't actually know what the status of that is.
`8 I haven't . . .
`9 Q. I probably wouldn't have heard of it if you
`10 haven't been disclosed. Anything else, any other
`11 cases that you're working on?
`12 MR. LEDAHL: Just to be clear, you're
`13 asking about cases where he's been disclosed to the
`14 opposing party so it's not a confidential consulting
`15 relationship or something like that?
`16 MR. DAVIDOFF: Right.
`17 A. I believe that's it.
`18 Q. And then without -- just "yes" or "no," I
`19 guess -- are there other cases where you're working
`20 on a confidential consulting basis?
`21 A. Yes.
`22 Q. And how many of those are you working on
`23 now?
`24 A. I think just one.
`25 Q. And is that one case with Russ August?
`
`Page 12
`1 A. Well, actually there's that one and there's
`2 another one that I guess has been in limbo that I
`3 just was recently contacted about. So that might be
`4 two.
`5 Q. Is one of those or both of those with Russ
`6 August?
`7 A. No.
`8 Q. What about Amster Rothstein?
`9 A. What about?
`10 Q. Amster Rothstein, is that the . . .
`11 A. No, I don't think so.
`12 Q. The Brew House case, who have you been
`13 retained by?
`14 A. Well, I was retained, I guess, by Brew
`15 House. I'd have to look up the attorneys. That was
`16 work -- I sometimes do work with a consulting firm
`17 called the Analysis Group.
`18 Q. And do you know who the opposing party on
`19 that case is?
`20 A. I'd have to look up the names. It's like
`21 it's an individual who I think is trying to set up a
`22 class action.
`23 Q. I see. Brew House is the defendant?
`24 A. Yes.
`25 Q. And do you know what court that's pending
`Page 13
`
`1 in?
`2 A. Um . . .
`3 Q. Even just geographically.
`4 A. Yeah. It was someplace odd. I think it
`5 may be Missouri.
`6 Q. In all of your prior work -- obviously we
`7 haven't covered other places where you may have just
`8 given deposition testimony or expert reports --
`9 but I now want to ask a question that sort of
`10 encompasses all of that, if that makes sense.
`11 A. Um-hum.
`12 Q. In any of that -- the work that you've done
`13 related to litigation, have you ever had any of your
`14 opinions excluded by a court?
`15 A. Not like in the Daubert sense --
`16 Q. Okay.
`17 A. -- if that's what you mean.
`18 Q. In some other sense?
`19 A. Well, my understanding is that like at
`20 various times during trial, judges may say like, Oh,
`21 well, like this patent is out, so there's going to
`22 be no testimony about that patent if I'd put in
`23 testimony, like in the deposition or the report or
`24 things like that.
`25 So I think there have been times where
`4 (Pages 10 - 13)
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 153-12 Filed 06/28/19 Page 7 of 96
`
`Page 14
`1 the judge has sort of said, Okay, the testimony is
`2 limited to the subjects in these areas.
`3 Q. Understood. Any other kind of situations
`4 where you -- and of course, you've maybe offered
`5 opinions and cases have settled.
`6 A. Um-hum.
`7 Q. But any other situations other than the
`8 patents dropped out or the case settled where your
`9 opinions were precluded?
`10 A. The only one that really sticks to mind is
`11 there was a case where there was a lot of code
`12 involved, and so some of the code I discussed in the
`13 report and then we had like some large appendices
`14 where we were citing to related code, I think the
`15 judge said like, You can only talk about code that
`16 he specifically talked about like in the main body
`17 of the report.
`18 Q. Do you remember which case that was?
`19 A. It was one of the Finjan cases, but I don't
`20 remember which one.
`21 Q. Even if your testimony wasn't excluded,
`22 has, to your knowledge, a court ever found any of
`23 your opinions unreliable?
`24 A. I'd have to go back and look. I mean, I
`25 don't recall that specific phrasing. But, you know,
`Page 15
`
`1 I don't recall everything.
`2 Q. I don't mean to be wedded to the particular
`3 phrasing. Do you recall a court ever criticizing
`4 your opinions or your testimony?
`5 A. Not offhand, but I'd have to go back and
`6 look.
`7 Q. So as you sit here today, you can't remember
`8 a judge in the courtroom, for example, saying
`9 something negative about your testimony?
`10 A. Oh, I mean, I can imagine a judge saying
`11 some negative things.
`12 Q. I'm not --
`13 A. I've had cases where the judge has said
`14 like, you know, Stop interrupting the lawyer or, you
`15 know, or like said like, That's not relevant; let's
`16 stay on topic -- or I'm sure -- because I've been in
`17 some number of trials, I've certainly had different
`18 interactions with judges where, you know, they've
`19 tried to put me on the track that I guess they
`20 wanted for their trial.
`21 Q. Can you remember any of those specifically?
`22 A. Let's see. I'd have to go back and recall.
`23 I know it's happened. I just might be mistaken if I
`24 tried to relate a specific incident.
`25 Q. What about written criticism? Has a judge
`
`Page 16
`1 ever written an opinion or order that criticized
`2 your testimony?
`3 A. I don't know. They may have. I mean, I
`4 don't recall offhand.
`5 Q. As you sit here, you're not aware of any
`6 judge writing an opinion that criticized your
`7 testimony?
`8 A. Again, I just say like I can't recall.
`9 Q. So you think it might have happened?
`10 A. Yeah. I mean, it might have. I mean, I
`11 think like there's certainly times where I've gone
`12 back and forth with judges and --
`13 Q. I'm asking about in writing.
`14 A. What?
`15 Q. I'm asking about in writing. Has a judge
`16 written an opinion or an order?
`17 A. I'd say I think I don't recall in many
`18 cases, like, you know, I might not have seen all
`19 their writings with regard to my testimony.
`20 Q. I'm just asking you -- I'm just asking you
`21 what you're aware of. Can you think of any instance
`22 where a judge in writing criticized you?
`23 A. I'd have to go back and look to remember.
`24 Q. So right now you can't remember any?
`25 A. Yeah. I'd have to go back and look.
`
`Page 17
`
`1 Q. So it's possible that a judge has written
`2 an opinion criticizing your testimony; you just
`3 don't recall it right now?
`4 MR. LEDAHL: Objection, speculation.
`5 A. It's certainly possible, yeah.
`6 Q. When were you first retained to work on
`7 this matter for Network-1?
`8 A. I think it was just a couple months ago.
`9 Q. This year?
`10 A. I think so, yeah.
`11 Q. Is there some ambiguity about when you were
`12 retained? You're hesitating.
`13 A. No. I just don't recall. My recollection
`14 is -- I mean, I checked before the deposition, but
`15 my recollection is that it was very recently.
`16 Actually, yeah, now that I think about it, I think
`17 it was last month and a half, two months ago that --
`18 Q. Certainly in 2019?
`19 A. Yeah.
`20 Q. And that's the first time you did any work
`21 for this case, I assume?
`22 A. Yeah.
`23 Q. You've done work -- been retained by -- or
`24 on behalf of clients of Russ August before, correct?
`25 A. Yes.
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`5 (Pages 14 - 17)
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 153-12 Filed 06/28/19 Page 8 of 96
`
`Page 18
`1 Q. And can you tell me which matters you've
`2 been retained by Russ August in connection with?
`3 A. I believe it was the set of Realtime cases.
`4 I think we've discussed or come out that there's
`5 some other cases where -- or that may come up that I
`6 don't think I've been disclosed.
`7 Q. In addition to the work you're doing for
`8 Network-1, you're currently working with Russ
`9 August on other cases as well, correct?
`10 A. Yes. I believe there's been one other case
`11 I've worked with them recently.
`12 Q. You're working with them on one other case
`13 in addition to Network-1?
`14 A. I believe so, yes.
`15 Q. I just want to make sure I'm not . . .
`16 It's at least one other case; is that
`17 what you're saying?
`18 A. I think right now it's just one.
`19 Q. So at least one other case?
`20 A. Well --
`21 MR. LEDAHL: Objection, misstates the
`22 testimony.
`23 Q. I'm just trying to understand what your
`24 "I believe so" -- I just -- you're working with them
`25 on Network-1, correct?
`
`Page 20
`1 Q. But you're aware that you've been retained
`2 by Russ August in connection with a case that has a
`3 party with the name "Realtime" in their name?
`4 A. That's what I recall, yes.
`5 Q. Is there any ambiguity about your
`6 recollection there?
`7 A. I don't think so, but I would, again, have
`8 to go back and look. Like I said, I haven't billed
`9 or done anything with that case that I can recall in
`10 a few months, so I would just have to look back and
`11 check its status.
`12 Q. You don't recall that relationship being
`13 terminated?
`14 A. No.
`15 Q. Other than that case, are there any other
`16 cases in which you're currently under retention by
`17 Russ August? Other than that case and this case, of
`18 course.
`19 A. I don't believe so.
`20 Q. Have you ever been -- worked with the firm
`21 Amster Rothstein in connection with a case?
`22 A. I don't remember the name. I don't think
`23 so.
`24 Q. They're local counsel in this case. I'm
`25 not trying to hide the ball on that. You didn't
`
`Page 19
`
`1 A. Yes.
`2 Q. And you're working with them on a Realtime
`3 case, correct?
`4 A. I believe -- I'd have to check the
`5 paperwork. I believe that's been signed, so I think
`6 so, yes.
`7 Q. Well, I mean, I haven't heard an objection
`8 so I'm just asking you if you remember. You've been
`9 retained by them for a Realtime case that you're
`10 currently working on?
`11 A. I'm not clear. I think we might have just
`12 had some back-and-forth conversations to begin with,
`13 but I don't think I've done any active work on it
`14 for a few months. So I don't know that I'm working
`15 on it at the moment.
`16 Q. You're currently retained by Russ August in
`17 connection with a Realtime matter, correct?
`18 A. I believe so, yes. That's my recollection.
`19 Q. Well, I guess I'm just trying to
`20 understand. When you say you believe so, is there
`21 some ambiguity?
`22 A. So I don't remember what the actual name of
`23 the party was, like there's been a Realtime Data,
`24 but there was some other name. Like I would just
`25 have to check my record.
`
`Page 21
`
`1 know that?
`2 A. No.
`3 Q. And you don't recall working with them on
`4 any other matter?
`5 A. No.
`6 Q. Have you -- You've obviously submitted a
`7 declaration relating to claim construction in this
`8 case, correct?
`9 A. Yes.
`10 Q. Have you been retained to look at any
`11 issues of infringement in connection with this case?
`12 A. I mean, I believe my retainer says that
`13 that may come up -- or that like I may be involved
`14 in other parts of the case, but I haven't done any
`15 work on infringement at this point.
`16 Q. Your retention's not limited to claim
`17 construction issues, correct?
`18 A. No, I don't believe so.
`19 Q. You signed a protective order in this case;
`20 is that right?
`21 A. Yes, I believe so.
`22 Q. Why did you do that?
`23 A. Because the attorneys asked me to.
`24 Q. Have you reviewed any material covered by
`25 the protective order in this case?
`
`6 (Pages 18 - 21)
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 153-12 Filed 06/28/19 Page 9 of 96
`
`Page 22
`
`Page 24
`
`1 A. I don't believe so.
`2 Q. Have you reviewed any Google or YouTube
`3 nonpublic material in connection with this case?
`4 A. No, not at this point.
`5 Q. Have you spoken with any other experts who
`6 have been retained in connection with this case?
`7 A. No.
`8 Q. Have you gained in your work on this case
`9 so far any understanding of how the YouTube content
`10 ID system works?
`11 A. No.
`12 Q. You've not had any conversations about that
`13 with anyone?
`14 A. No.
`15 Q. Including counsel?
`16 A. No. I believe that just in terms of the
`17 background of the case, I believe it came up that,
`18 you know, they think that YouTube or Google used
`19 some form of locality-sensitive hashing, which I
`20 said, you know, matched my background, but no
`21 details, I think, beyond that general description.
`22 Q. Can you recall what you were told about
`23 Google or YouTube's use of locality-sensitive
`24 hashing?
`25 A. I believe simply that they used it.
`
`1 the area of -- read a variety of papers that relate
`2 to Google and Google systems or to systems generally
`3 that deal with video and audio files.
`4 And so my recollection is that I've
`5 heard that content ID uses locality-sensitive
`6 hashing, but again just in a general sense. I don't
`7 know the details of that functionality.
`8 Q. You've read papers that describe how Google
`9 sensitive ID works?
`10 A. I don't recall any level of detail of that
`11 locality-sensitive hashing. I mean -- and again,
`12 this is sort of vague recollection -- I've read lots
`13 of papers that talk about systems in general, but in
`14 particular, I've read some papers that relate to
`15 YouTube systems. I believe they relate to locality-
`16 sensitive hashing.
`17 Q. Do you recall the authors of any of those
`18 papers?
`19 A. No.
`20 Q. Do you recall what journals you read those
`21 papers in?
`22 A. Not offhand. And it might not have been
`23 for journals. It might have been for conferences or
`24 things I reviewed.
`25 Q. Can you recall any identifying information
`
`Page 23
`1 This was just making sure that I'd have background
`2 suitable for this sort of case.
`3 Q. Were you told anything else about the
`4 operations of YouTube?
`5 A. No, not by counsel or . . .
`6 Q. Have you read any depositions in this case?
`7 A. No. I don't think so. I can check my
`8 report, but I don't -- my declaration, but I don't
`9 recall any.
`10 Q. There's nothing cited in your declaration,
`11 I don't think.
`12 A. Then no.
`13 Q. What, if anything, is your understanding of
`14 how locality-sensitive hashing is used by content ID?
`15 MR. LEDAHL: Vague and ambiguous.
`16 A. Can I -- I would say I don't know at any
`17 detailed level. My understanding is they may use
`18 some form of locality-sensitive hashing to provide
`19 some sort of, you know, I guess, sketch or form of
`20 identifier for the content.
`21 Again, I think that's things I think
`22 I've also heard outside of the context of counsel,
`23 but just at that high level.
`24 Q. Where else do you think you've heard that?
`25 A. Well, so, again, I didn't sort of work in
`
`Page 25
`
`1 about those papers?
`2 A. No. I don't recall or know how the --
`3 currently how the system functions at that level of
`4 detail.
`5 Q. I'm just asking about the papers. Can you
`6 recall the dates of publication of any of those
`7 papers?
`8 A. Nope.
`9 Q. Can you recall when you read them, roughly?
`10 A. Not with any specificity. Again, I do
`11 reviews and things for a lot of systems-types
`12 conferences, so I would expect maybe over the last
`13 five years, but I can't recall with any specificity.
`14 Q. Other than what you've read in papers and
`15 what counsel has told you, is there any other source
`16 of your understanding that content ID may use
`17 locality-sensitive hashing?
`18 A. Not that I can recall.
`19 Q. Have you ever spoken with any employees of
`20 Google or YouTube about content ID?
`21 A. Not that I can recall.
`22 Q. When you say not that you can recall, do
`23 you think it's a possibility that you've done that?
`24 A. I mean, I have friends or colleagues that
`25 work at Google, and I'm sure at times I've talked
`7 (Pages 22 - 25)
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 153-12 Filed 06/28/19 Page 10 of 96
`
`Page 26
`
`1 with them generally about things. I don't recall
`2 any specific conversations about content ID.
`3 Q. Who are your friends who work at Google?
`4 A. Andrei Broder, I believe, still works
`5 there. He was a previous mentor of mine.
`6 Let's see. Adam Kirsch, K-i-r-s-c-h.
`7 He was a student of mine that went to work at
`8 Google.
`9 Let's see. I think Serge Vassilvitskii
`10 works at Google. I talk with him about research
`11 problems. I don't think I would have ever -- I
`12 don't think he knows about or I would have talked
`13 with him about content ID. But I'm sure there are
`14 plenty of other people I know at Google. I think --
`15 Q. Can you think of anyone else?
`16 A. I'd have to go check and look just because
`17 I -- even if I -- I wouldn't be sure that they
`18 worked at Google.
`19 But, yeah, I don't think any of these
`20 people work on -- specifically on content ID, so I'm
`21 not sure what they would have known, but probably in
`22 terms of conversations just talked about general
`23 things with them, since I'm interested in hashing
`24 generally, topics related to hashing might have come
`25 up in talking to them.
`
`Page 27
`
`1 Q. Just a point of clarity. You said your
`2 friends and colleagues that work -- or worked at
`3 Google. And I asked you about friends and you gave
`4 me these three names. Were you distinguishing or
`5 were you giving me friends and colleagues?
`6 A. Friends and colleagues. You know, I guess
`7 Andrei I would consider a friend. And Adam was my
`8 student. I think I would consider him a friend.
`9 Serge is more of a colleague.
`10 Q. Any other colleagues who either work or
`11 have worked at Google that you can think of?
`12 A. So -- Let's see. So again, to be clear,
`13 these are not people who I recall having any sort of
`14 specific conversation about Google products. But if
`15 you're asking just at a general level, are there
`16 colleagues I have known who work at Google, I know
`17 David Anderson. He's a professor at CMU. I think
`18 he recently spent some sabbatical time at Google.
`19 Another colleague that I think is
`20 currently spending time at Google is Rasmus Pagh,
`21 P-a-g-h. Again, I should make clear that, you know,
`22 I don't recall particularly, you know, with David or
`23 Rasmus, who are just professors and colleagues of
`24 mine who I've written papers with and who I know
`25 spent time at Google.
`
`Page 28
`
`1 And in particular, I don't think I've
`2 had any Google-specific conversations, certainly not
`3 with regard to technology, you know, with -- and
`4 even with the others, I don't recall any
`5 conversations or anything specific about content ID.
`6 They're just colleagues of mine that I'm sure I've
`7 talked about work things with at various times.
`8 Q. But with respect to the first three --
`9 Adam, Andrei, and Serge -- you think it's possible
`10 you could have talked about locality-sensitive
`11 hashing with them generally?
`12 A. Possibly, I mean, just because I'm
`13 interested in hashing in my research, and I've
`14 talked with them about research topics, so . . .
`15 Q. And so you might have talked with them
`16 about hashing work they did at Google?
`17 A. At a very general level.
`18 Q. You mentioned with respect to Serge that
`19 you were talking with him about a particular
`20 research problem or problems. Did you have
`21 something more specific in mind?
`22 A. It doesn't really relate to hashing.
`23 I mean, we've been sort of -- he's written some
`24 stuff about how to use machine learning in
`25 conjunction with what I'd call more traditional
`Page 29
`1 algorithms to improve them, and I've also done some
`2 work in that area.
`3 So we've had some back-and-forth in
`4 letting each other know about our work in that area
`5 just because we're both sort of working on the
`6 same -- or not the same set of problems but sort of
`7 a similar theme.
`8 Q. What's your understanding of what he's
`9 working on?
`10 A. I believe he works in the Google research
`11 division. So I guess I think he's like a professor
`12 at Google, but not at a university. So all I know
`13 is, you know, the research he publishes or puts out.
`14 Q. Do you know what projects he works on at
`15 Google?
`16 A. No.
`17 Q. Do you know whether or not Google uses
`18 machine learning in its content ID system?
`19 A. No, not specifically. I mean, I might
`20 guess that it does in places, but I honestly haven't
`21 looked at the system.
`22 Q. What's your guess on where it would use it?
`23 MR. LEDAHL: Objection, calls for
`24 speculation.
`25 A. Just a guess, I know Google invests heavily
`8 (Pages 2

This document is available on Docket Alarm but you must sign up to view it.


Or .

Accessing this document will incur an additional charge of $.

After purchase, you can access this document again without charge.

Accept $ Charge
throbber

Still Working On It

This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.

Give it another minute or two to complete, and then try the refresh button.

throbber

A few More Minutes ... Still Working

It can take up to 5 minutes for us to download a document if the court servers are running slowly.

Thank you for your continued patience.

This document could not be displayed.

We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.

Your account does not support viewing this document.

You need a Paid Account to view this document. Click here to change your account type.

Your account does not support viewing this document.

Set your membership status to view this document.

With a Docket Alarm membership, you'll get a whole lot more, including:

  • Up-to-date information for this case.
  • Email alerts whenever there is an update.
  • Full text search for other cases.
  • Get email alerts whenever a new case matches your search.

Become a Member

One Moment Please

The filing “” is large (MB) and is being downloaded.

Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.

Your document is on its way!

If you do not receive the document in five minutes, contact support at support@docketalarm.com.

Sealed Document

We are unable to display this document, it may be under a court ordered seal.

If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.


Access Government Site

We are redirecting you
to a mobile optimized page.





Document Unreadable or Corrupt

Refresh this Document
Go to the Docket

We are unable to display this document.

Refresh this Document
Go to the Docket