`
`Exhibit 29
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`Case 1:14-cv-02396-PGG-SN Document 239-11 Filed 11/12/20 Page 2 of 7
`
`Page 1
`
` UNITED STATES DISTRICT COURT
`
` SOUTHERN DISTRICT OF NEW YORK
`
` - - - - - - - - - - - - - - - x
`
` NETWORK-1 TECHNOLOGIES, INC., :
`
` Plaintiff, :
`
` vs. : 14 Civ. 2396 (PGG)
`
` GOOGLE, LLC and YOUTUBE, LLC, : 14 Civ. 9558 (PGG)
`
` Defendants. :
`
` - - - - - - - - - - - - - - - x
`
` VIRTUAL VIDEOTAPED DEPOSITION OF: TREVOR DARRELL
`
` DATE: Friday, July 10, 2020
`
` TIME: 1:11 p.m. EDT
`
` LOCATION: Remote Proceedings
`
` REPORTED BY: Denise M. Brunet, RPR
`
` Reporter/Notary
`
` Veritext Legal Solutions
`
` 1250 Eye Street, N.W., Suite 350
`
` Washington, D.C. 20005
`
` Job No. CS4168610
`
`800-567-8658
`
`973-410-4098
`
`Veritext Legal Solutions
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 239-11 Filed 11/12/20 Page 3 of 7
`
`Page 102
`1 aware of that, for example, on page 97 and on
`2 page 108 and the following two pages. And I also
`3 cite code on pages 115, 117, and so on and so
`4 forth.
`5 Q All right. Well, let's look at the code
`6 that you referenced on page 97. Is this code that
`7 appears on page 97 of your report code that was
`8 compiled into the Clango alpha release?
`9 A I believe it has, and I believe I relied
`10 on the testimony of the creators of this system
`11 for that opinion.
`12 Q So you did not have a separate
`13 compiled -- or directory of the code that was
`14 itself actually compiled into the alpha. You're
`15 relying on the testimony that this is code that
`16 was in the alpha?
`17 A Whatever my opinions are, I believe my
`18 basis for them in these paragraphs are the
`19 testimony from the creator's depositions.
`20 Q And that's it, the testimony, right?
`21 A That is it.
`22 Q There was no other document, for example,
`
`Page 103
`
`1 that indicated that this particular code on
`2 page 97 of your report was actually part of the
`3 Clango alpha release, correct?
`4 MR. HARDY: Objection to form.
`5 THE WITNESS: I don't recall seeing
`6 anything like that.
`7 BY MR. LEDAHL:
`8 Q Okay. And you also mention code that you
`9 excerpted starting at page 108 of your report; is
`10 that right?
`11 A Yeah.
`12 Q Now, this code on page 108 is an include
`13 file?
`14 A Well, I believe it's literally an SCCS
`15 file. But it's an SCCS file that represents an
`16 underlying include file.
`17 Q When you say an SCCS file, what do you
`18 mean by that?
`19 A As described in paragraph 189 of my
`20 report, it's the file from a source code control
`21 system. The name of that -- of the particular
`22 source code control system was, unsurprisingly,
`
`Page 104
`
`1 SCCS.
`2 Q And what is include file?
`3 A It's a file that was used -- in
`4 particular, this is an include file in the C
`5 programming language, and it is part of the source
`6 code for a computer application or a computer
`7 program.
`8 Q And an include file identifies other
`9 source code files to be included in the
`10 definitional and functional sense of the code. Is
`11 that fair?
`12 A I actually think it might be slightly
`13 more fair to reverse it and say an include file is
`14 a file that another source code file includes to
`15 access routinely defined or included elements.
`16 Q Okay. Was the source code that you
`17 excerpt on page 108 of your report included in the
`18 Clango alpha release?
`19 A I can't recall which release it was, as I
`20 sit here today. I would have to refresh my memory
`21 from the report.
`22 Q Well, let me direct you to the text you
`
`Page 105
`1 have in paragraph 189. You offer the assertion
`2 that this source code file was created on
`3 July 26th, 2000. Do you see that?
`4 A Yes.
`5 Q And that's three weeks after the date you
`6 say the Clango alpha release took place, right?
`7 A That's correct.
`8 Q So unless people at Audible Magic were
`9 engaged in time travel, this file would never have
`10 been a part of any alpha release of the Audible
`11 Magic Clango system, correct?
`12 A No.
`13 Q So it's your testimony that this file
`14 created on July 26th was included in a release
`15 created at least three weeks earlier, according to
`16 your testimony?
`17 A I believe my testimony is this is
`18 representative and is consistent with Mr. Wold's
`19 testimony. And as you know from other discussions
`20 in this matter, the source code control system,
`21 this just shows the latest date this file could
`22 have been created. This file may have existed for
`
`800-567-8658
`
`Veritext Legal Solutions
`
`27 (Pages 102 - 105)
`
`973-410-4098
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 239-11 Filed 11/12/20 Page 4 of 7
`
`Page 106
`1 months or years before. And indeed this file was,
`2 according to the testimony, most likely obtained
`3 from a external open source repository that had
`4 existed for years and reflected source code and
`5 algorithmic knowledge that one of ordinary skill
`6 in the art would be very familiar with it.
`7 So I believe my testimony in
`8 paragraph 189 is clear, which is that it's
`9 consistent with the testimony regarding the alpha
`10 release.
`11 Q So as to the functioning of the alpha
`12 release, am I understanding correctly that you're
`13 relying on testimony and that the code that you
`14 reviewed was not actually the code from that alpha
`15 release?
`16 MR. HARDY: Objection to form.
`17 THE WITNESS: I don't believe I've give
`18 an opinion on that one way or the other. It's
`19 very consistent with the testimony. And the
`20 testimony was that they incorporated available
`21 implementations of approximate nearest neighbor
`22 searching, and that this file is consistent with
`
`Page 107
`1 what they did at the time even though the date on
`2 this -- the date that they have on the record
`3 was -- perhaps it had been moved -- in fact, if
`4 you look at the source code control metadata, it
`5 appears that they cleaned up or relabeled or
`6 renamed the files.
`7 This file is clearly based on an
`8 implementation of algorithms that, you know, date
`9 from 1997 to 1993 that were, according to the
`10 testimony, not written by Muscle Fish, but they
`11 were downloaded and used by Muscle Fish. So this
`12 file is just one of many copies that had likely
`13 been around the web that you could get.
`14 And so the fact that this file has a date
`15 on it of July of 2000 is not evidence that they
`16 didn't have access to the source code earlier;
`17 it's just this happens to be the date that the --
`18 whatever archive process had produced.
`19 I find the source file useful in showing
`20 an illustration of a contemporaneous example of
`21 what the testimony was talking about.
`22 BY MR. LEDAHL:
`
`Page 108
`
`1 Q What's the first date that this source
`2 file was in Audible Magic Clango software?
`3 A I'm not sure.
`4 MR. HARDY: Objection to form. Sorry.
`5 BY MR. LEDAHL:
`6 Q You don't know?
`7 A I don't know a specific date. I only
`8 know the testimony of the creators.
`9 Q And you wrote that Mr. Wold created
`10 kd-tree code in July of 2000 in paragraph 189 of
`11 your report, correct?
`12 A Correct.
`13 Q So do you have any basis to suggest that
`14 that file was created before July 5th, 2000?
`15 A Yeah. The file he created -- I believe
`16 his testimony is not that he wrote this, but that
`17 he had downloaded it and used it. His testimony
`18 is that they had a version that was running prior
`19 to this. Obviously, there's no evidentiary record
`20 of a version earlier than this, but that's not
`21 evidence that it didn't exist before.
`22 So I'm relying on the testimony of the
`
`Page 109
`
`1 creators for the dates and I'm relying on the
`2 source code to show an example of what they were
`3 talking about.
`4 Q So just to be clear, you've seen no
`5 evidence of an actual file that implemented a
`6 kd-tree structure that's dated earlier than
`7 July 26th, 2000; is that correct?
`8 A Well, I certainly have in the world.
`9 These -- this software release was well-known and
`10 available to anyone who wanted to download it for
`11 years prior to 2000.
`12 Q Sorry. Let me be clear. You've not seen
`13 any evidentiary record that Audible Magic or the
`14 Clango system had a file that implemented a
`15 kd-tree structure prior to July 26th, 2000,
`16 correct?
`17 A The evidentiary record that I have seen
`18 is that they express the desire and interest to
`19 find an approximate nearest neighbor search
`20 algorithm in 1999 and early 2000, well before
`21 July, and it seems they went and found one of the
`22 standard open source repositories that had been
`
`800-567-8658
`
`Veritext Legal Solutions
`
`28 (Pages 106 - 109)
`
`973-410-4098
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 239-11 Filed 11/12/20 Page 5 of 7
`
`Page 110
`1 available for years to anyone who was interested
`2 in this matter.
`3 And the specific file that I reproduced
`4 on 108 was, I assume, the earliest record they
`5 could have which has a date of July 26th, 2000.
`6 And that doesn't prove or disprove what dates they
`7 may have earlier had this file.
`8 Q And you -- my question to you, sir, is
`9 the simple question: You've seen no evidence of
`10 the existence of this file at Audible Magic
`11 earlier than July 26th, 2000, correct?
`12 A I mean, I'd say no. I have seen evidence
`13 that this file was available to anybody who wanted
`14 on the Internet in the late '90s. And I've seen
`15 evidence in their testimony that they wanted to go
`16 out and get something like that and they got it.
`17 So to me, that's evidence.
`18 Q What evidence have you seen that this
`19 file was implemented in any version of the Clango
`20 system prior to July 26th, 2000, sir?
`21 A The deposition testimony of the creators
`22 of the system.
`
`Page 112
`1 we've been discussing on page 108, that Mr. Wold
`2 testified that that was in the Clango system in
`3 July of 2000.
`4 And separately, on page 113, in
`5 paragraph 189, I then proceed to discuss software
`6 related to the Clango August beta application.
`7 Q Which file are you talking about in
`8 connection with the August Clango beta application
`9 that you're discussing on page 113?
`10 A I'm confused by your question, because I
`11 didn't refer to a file in the August beta
`12 application.
`13 Q Okay. So you're not discussing an actual
`14 source code file of the August beta application in
`15 paragraph 113; is that fair? Or on page 113.
`16 A No.
`17 Q Well, what file are you talking about,
`18 then?
`19 A Well, in paragraph 196, I say that I have
`20 examined the source code files reflected in
`21 Exhibit 18 from the Audible Magic deposition. And
`22 then I reproduced a graphical folder depiction of
`
`Page 111
`
`Page 113
`
`1 Q And that's all?
`2 A That's quite enough.
`3 Q Have you seen evidence that that file
`4 that you excerpted on page 108 was incorporated
`5 into a beta release of the Clango system?
`6 A In paragraph 195, I refer to the
`7 testimony of Mr. Wold who discusses the source
`8 code of the Clango August beta application.
`9 Q Is that this file or a different file?
`10 A I'm sorry, could you repeat that
`11 question? I'm not sure I understand what you mean
`12 by "this file."
`13 Q Well, we've been talking about a file
`14 that you've reproduced, or part of one, on
`15 page 108 of your report. And you just mentioned
`16 testimony that you referred to from Mr. Wold and
`17 discussed in paragraph 195 of your report about
`18 the Clango beta release, you say.
`19 Is that testimony about this file that is
`20 on page 108 of your report?
`21 A I believe my testimony in paragraph 189
`22 is that the source code and include file that
`
`1 the set of those files that I'm referring to in
`2 paragraph 198 -- excuse me, 196.
`3 Q Now, the modification date of all the
`4 files that you referred to there on page 114, in
`5 paragraph 196, those are all -- with the exception
`6 of the first two files, those are all listing
`7 modification dates in 2013, correct?
`8 A Yeah.
`9 Q Do you have any understanding of what
`10 modifications were made to those files prior to
`11 that date in 2013?
`12 A Could you repeat the question?
`13 Q Sure. The files that are listed on
`14 page 114 of your report from this directory all
`15 reflect a modification date in 2013. Do you have
`16 any understanding as to what modifications were
`17 made to those files at any time between 2000 and
`18 2013?
`19 A My understanding is there were none
`20 because the testimony of Mr. Wold is that he
`21 represents this was the source code from the
`22 period in question, from the beta release.
`
`800-567-8658
`
`Veritext Legal Solutions
`
`29 (Pages 110 - 113)
`
`973-410-4098
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 239-11 Filed 11/12/20 Page 6 of 7
`
`Page 114
`1 Q Did you review any source code control
`2 system information that confirmed that
`3 representation?
`4 A I relied on Mr. Wold for that opinion.
`5 Q Is it common in a source code control
`6 system to have versioning control that will
`7 recreate, for example, a release of a software
`8 product as a snapshot in time?
`9 A Could you repeat the question?
`10 Q Yes. Is it common in a source code
`11 control system to have a versioning control that
`12 would allow you to recreate, for example, a
`13 particular release of a software product at a
`14 snapshot in time?
`15 A That's a typical function.
`16 Q Did you see any snapshot in time code for
`17 the Clango alpha release in the source code
`18 control system that you reviewed?
`19 A I don't recall one way or the other.
`20 Q Did you see any snapshot in time code for
`21 the beta release of the Clango system in the
`22 source code control system that you reviewed?
`
`Page 116
`
`1 A So let me reiterate. The software that
`2 we're talking about here, especially the kd-tree,
`3 was publicly available software that had been
`4 widely distributed, and any person of ordinary
`5 skill in the art would known it existed and would
`6 go get it whenever they wanted to implement
`7 approximate nearest neighbor kd-tree search, as
`8 did these inventors.
`9 So, you know, even the notion of version
`10 control is kind of complicated. Are we talking
`11 about the version inside Muscle Fish, inside
`12 Audible Magic, in the repository where the
`13 creators of the Audible Magic and Muscle Fish
`14 system went to go get the base implementations of
`15 these algorithms that they then modified or used
`16 in their application? I don't see that version
`17 numbers are a material issue here. So I don't see
`18 any version numbers, but -- that I recall, as I
`19 sit here today, but it doesn't seem to be an issue
`20 to me.
`21 Q Sir, I'm going to ask you to try to limit
`22 yourself to the questions I actually ask you. Is
`
`Page 115
`
`1 A My memory is imperfect. I feel like
`2 there may have been, but I can't remember where in
`3 my report it is.
`4 Q Is there some other section of your
`5 report we should be looking at for that
`6 information?
`7 A I may be confusing myself between the
`8 FreeAmp software and the Clango software. There's
`9 something in my report about version control, but
`10 I can't remember where it is, to be honest with
`11 you. So if you want to guide me to something,
`12 I'll be happy to explain it.
`13 Q Was there some versioning number that was
`14 associated with the Clango alpha release, to your
`15 understanding?
`16 A I don't recall. If it's in my report,
`17 it's true, but I -- I don't recall it being in my
`18 report.
`19 Q And is the same true for the beta
`20 release, that you don't recall one way or the
`21 other whether there was any kind of versioning
`22 number for that beta release?
`
`Page 117
`1 it correct to say that the only basis on which you
`2 rely in order to assert that the Clango alpha
`3 software utilized a kd-tree search is testimony
`4 from Mr. Wold?
`5 MR. HARDY: Objection. Misstates the
`6 testimony.
`7 THE WITNESS: Could you repeat the
`8 question?
`9 BY MR. LEDAHL:
`10 Q Yes. Am I correct that it is your
`11 assertion that the Clango alpha software system
`12 utilized a kd-tree-based search?
`13 A I think that's a different question.
`14 Q I realize that. I'm breaking it up a
`15 little bit. So I'll restate it. Am I correct
`16 that you assert that the Clango alpha system used
`17 a kd-tree search?
`18 A My testimony is that the testimony of
`19 Dr. Wold was that there was a kd search algorithm
`20 in the Clango system at that time.
`21 Q Okay. Let me break it up. So your
`22 assertion is that the search algorithm used by the
`
`800-567-8658
`
`Veritext Legal Solutions
`
`30 (Pages 114 - 117)
`
`973-410-4098
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 239-11 Filed 11/12/20 Page 7 of 7
`
`Page 118
`1 Clango alpha system was a kd-tree-based search,
`2 correct?
`3 A Yes.
`4 Q And the complete basis for your opinion
`5 that the Clango alpha system used a kd-tree-based
`6 search is Dr. Wold's testimony?
`7 MR. HARDY: Objection. Misstates the
`8 record.
`9 BY MR. LEDAHL:
`10 Q I'm sorry, I didn't hear your answer,
`11 sir.
`12 A No.
`13 Q What else do you base that opinion on?
`14 A Well, as is testified to in my report, I
`15 also show that the source code they had
`16 contemporaneous with that release was well-known,
`17 open source software that, according to their
`18 testimony and -- according to their testimony was
`19 available, that it's very consistent that they had
`20 that implementation and they used it, and I have
`21 into reason to doubt that testimony.
`22 Q So you're relying on the testimony that
`
`Page 120
`1 that file in the possession of Audible Magic with
`2 a date earlier than July 26th, 2000?
`3 MR. HARDY: Objection to form.
`4 THE WITNESS: I think I did.
`5 BY MR. LEDAHL:
`6 Q Where do you discuss that in your report,
`7 sir?
`8 A In the testimony of the inventors, the
`9 creators.
`10 Q Is the sole evidence that you're relying
`11 on that Audible Magic was in possession of a copy
`12 of that file prior to July 26th, 2000 the
`13 testimony of the inventors?
`14 MR. HARDY: Objection to form.
`15 THE WITNESS: I believe it's
`16 corroborative that -- whether it's a week or two,
`17 you know -- it's corroborative that they had been
`18 using a famous widely available open source
`19 package. So I do believe this is useful evidence
`20 to know what they represented the package that
`21 they were using to have been.
`22 BY MR. LEDAHL:
`
`Page 119
`
`Page 121
`
`1 that's what was used in the alpha, correct?
`2 MR. HARDY: Objection to form.
`3 THE WITNESS: No. No. I'm also relying
`4 on the other evidence that I cite in my report.
`5 BY MR. LEDAHL:
`6 Q Which other evidence are you talking
`7 about?
`8 A The one we just referred to. The file
`9 that we were talking about on page 108.
`10 Q The file that you say has a creation date
`11 of July 26th, 2000?
`12 A That's a copy of the file. That copy had
`13 that creation date.
`14 Q Did you see any evidence of a copy of
`15 that file in the possession of Audible Magic that
`16 had an earlier date?
`17 A Well, I'm aware that this file was
`18 available for access for years prior --
`19 Q Listen to my question, please. That's
`20 not what I asked you, and I think you know that's
`21 not what I asked you.
`22 Did you see any evidence of a copy of
`
`1 Q I don't think I understand any of what
`2 you just said. Can you clarify -- what's the
`3 evidence that that file existed at Audible Magic
`4 on a date prior to July 26th, 2000?
`5 MR. HARDY: Objection. Asked and
`6 answered.
`7 THE WITNESS: I'll repeat -- and I hope
`8 it's clear enough -- is that it's determinative to
`9 me to know which particular software library they
`10 had been relying on as they developed this
`11 software. And so irrespective of the date on this
`12 particular copy of the software, the fact that
`13 they produced evidence that in July of 2000, they
`14 were well aware and had been routinely using this
`15 software to me is useful evidence to corroborate
`16 the testimony of Mr. Wold.
`17 So you've been asking me over and over
`18 and over again this question of, is the only thing
`19 I relied on Mr. Wold's testimony? And I'd like to
`20 just respectfully offer the answer no.
`21 BY MR. LEDAHL:
`22 Q And I'm trying to understand, sir, what
`
`800-567-8658
`
`Veritext Legal Solutions
`
`31 (Pages 118 - 121)
`
`973-410-4098
`
`