Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 1 of 14
`
`
`
`
`
`
` Exhibit 6
`
`
`
`
`
`
`
`
`
`
` Exhibit 6
`
`
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 2 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 1
`
` UNITED STATES DISTRICT COURT
` SOUTHERN DISTRICT OF NEW YORK
`___________________________________
` )
`NETWORK-1 TECHNOLOGIES, INC., )
` )
` Plaintiff, )
` )
`vs. ) No. 14 Civ. 2396
` ) 14 Civ. 9558
` ) (PGG)
`GOOGLE LLC and YOUTUBE, LLC, )
` )
` Defendants. )
`___________________________________)
`
` CONFIDENTIAL ATTORNEYS' EYES ONLY
` VIDEOTAPED DEPOSITION OF AUDIBLE MAGIC 30(b)(6)
` ERLING WOLD
` September 4, 2019 at 10:01 a.m.
` Three Embarcadero Center, 26th Floor
` San Francisco, California
`
`REPORTED BY: LANA L. LOPER,
` RMR, CRR, CCP, CME, CLR, CSR No. 9667
`____________________________________________________
` DIGITAL EVIDENCE GROUP
` 1730 M Street, NW, Suite 812
` Washington, D.C. 20036
` (202) 232-0646
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 3 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 14
`
`C-l-a-n-g-o?
` A Yes.
` Q And in general terms, what was Clango?
` A Clango was intended for people to be listening
`to, say, an Internet radio station. And when they were
`listening to an Internet radio station, they can run
`this other operation called Clango on their desktop.
` And if they heard a tune that they liked, they
`could press a button, I think, and they would identify
`that tune that was playing and then allow you to, you
`know, purchase it, for example, if you were interested.
` It showed you the metadata, and then it gave
`you a URL to an e-commerce site.
` Q Is it fair to say that Clango identified songs
`by their content?
` A Yes.
` Q Why try to identify songs by -- by their
`content?
` A Well, at the time there wasn't -- there wasn't
`always metadata. So if -- if someone was just listening
`to a radio station on the Internet, it might just be a
`stream from a terrestrial radio station, and so it
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 4 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 29
`
`simple idea.
` It's basically, at the time we were using
`10 MFCCs per segment of sound. And a segment of sound,
`you know, could be 25 milliseconds, or it could be a
`second, for example, something in that -- but it's
`something short, in that range.
` And the -- the way those -- so they're just
`floating by numbers.
` Let me think.
` The way that they are computed is, you do some
`kind of spectral analysis of the sound. So that's,
`like, what the ear does. You look for low frequencies,
`high frequencies, so on. And then you rectify that.
`And then you take another spectral analysis of that.
` And -- and what you end up with is a set of
`features that really describes the spectral shape of the
`sound. The lower-order features are kind of the gross
`shape of the spectrum, and the higher-order features are
`more the detail, the -- yeah, the higher features on the
`spectrum, and over time.
` Q And once the Clango application had extracted
`those features, what would it do next in the process of
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 5 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 30
`
`identifying a song?
` MR. LEDAHL: Lacks foundation.
` THE WITNESS: So in the actual application, as
`it was released in mid-2000, there would be the Clango
`application, itself, would just extract the features.
`And so it would -- when the person pressed the button,
`there would be, say, a circular buffer of the last -- or
`not necessarily a circular buffer. There would be a
`buffer containing the last so many seconds of the sound
`of what they had been listening to.
` That would all be fingerprinted at the Clango
`application. So this set of MFCCs would be extracted.
` Then that would -- that package would be sent
`over the network to a server at -- under Audible Magic's
`control. I think they actually had it in a third-party
`site at the time. I'm not sure.
` And at that -- and on -- on that end, there
`would be a reference database of fingerprints. And then
`some kind of lookup algorithm would be done to compare
`the fingerprint coming in with that set of references
`and -- to see if any of them were close enough that we
`would report a match.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 6 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 93
` Q And how were those fingerprints indexed for
`lookup as of the July 2000 alpha Clango release?
` A So the -- in the July release, we had moved
`from the -- the tree that I had mentioned earlier in my
`deposition, to a k-d tree, which is just another way of
`segmenting up the reference space, future space. It's
`a -- it's a binary tree.
` Q And how did searches of query fingerprints in
`the July 2000 k-d tree proceed?
` A So you would first -- um, well, a tree would
`have to be built in the -- in the database. So that --
`that was a step that was taken.
` So, you know, you would read in this database,
`and you would do an indexing step. The tree would be
`constructed. So that was before any queries were done.
` But when a query was done, then it would follow
`that path in a similar way to the way I mentioned
`before. Since this was the binary tree, you would
`just -- at each layer of the tree, you would be
`splitting -- you would be choosing one half of the
`remaining section of the possibilities, until you got to
`a leaf. And then you would compare it to everything
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 7 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 94
`
`there.
` Q Did searches of a query fingerprint in the July
`2000 k-d tree compare that fingerprint to every
`reference fingerprint in the k-d tree?
` A No, not typically.
` Q Did the search of the k-d tree only identify
`exact matches of the reference -- of the query
`fingerprint?
` A Well, again, I -- it depends what you mean by
`exact.
` No, the -- typically, you would -- you would
`find -- I mean, actually, I should just say no.
` It would not only find exact matches; it would
`find similar enough matches.
` Q How did the search algorithm decide whether a
`match was similar enough?
` A So the -- after you followed this tree down to
`the bottom, then you would compare the query to every
`reference in that leaf. And that was a Euclidean
`distance at the time.
` So we were just taking the set of however many
`numbers, 200 number, so 200 dimensional feature vector,
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 8 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 98
`linearly, so less than order of N, where N is the
`database size.
`BY MR. DANG:
` Q So to clarify, it scaled less than linearly
`with the size of the database?
` A Correct.
` Q Was that true even if the computer processing
`power was held constant?
` A Yes.
` Q As of the August 2000 beta release, how were
`the reference fingerprints indexed in the back-end
`database?
` A Sorry, say that once more time. I was just
`thinking to the answer to the question before, so
`please?
` Q No problem.
` As of the August 2000 beta release --
` A Yes.
` Q -- how were the reference fingerprints indexed
`in the back-end database?
` A Yeah, so that was using this k-d tree that I
`mentioned.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 9 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 204
`
`ID server at the back end.
` Q Now, the actual search, if you will, of the
`database, that was run on Audible Magic servers,
`correct?
` A Yes.
` It's -- it's possible that, even at that time,
`there may have been third-party server companies or
`something like that. But they were our servers, yes.
` Q Well -- well, let me -- let me see if I can
`differentiate a little more clearly.
` So Clango was essentially a client application
`that would be downloaded onto a user's computer, right?
` A That's correct.
` Q But the Clango application that a client would
`download, that didn't actually run the search, did it?
` A That is correct.
` Q The search was on something else that the user
`didn't have access to?
` A That's correct, except through the sending of
`the package to it, yes.
` Q But it was a black box the user had no
`understanding of, correct?
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 10 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 205
`
` A Right.
` MR. DANG: Object to the form.
` THE WITNESS: I mean, some users may have known
`something, but yes.
`BY MR. LEDAHL:
` Q But it wasn't -- it's not as though code for
`running the search was downloaded onto a user's
`computer. Is that right?
` A Correct.
` Q And it's not --
` A Actually, you know what, I'm not sure about
`that. I actually wonder if we didn't distinguish,
`because we -- we gave them one piece of code. It's --
`it's possible that that code existed on the Clango
`application. I -- I would have to check into that.
` Q You just don't know?
` A But it wasn't called directly, if that's the
`distinction you're trying to make.
` Q Well, was there -- I -- I guess users didn't
`have -- I mean, for one thing, they didn't have the
`database, right?
` A Correct.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 11 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 212
` a high hit rate on the tier 1 database.
` Does that help you understand the distinction?
` A That -- that makes me a little bit unclear what
`it is, because the daily updates makes it sound like
`it's the first thing I said; that the tier 2 database
`was to make -- now, maybe -- actually, now that I'm
`rereading it...
` Ah, now that I'm rereading it, actually, I
`think now it makes more sense.
` I think that the tier 2 database was then
`incorporated into the tier 1 database on a daily basis.
` Q Was -- was there ever a -- a database of what I
`will call more frequently hit references?
` A Later, but not at this time.
` Q Now, what was the first time that you or
`anybody at Audible Magic made any public disclosure
`about any of the indexing techniques you use there?
` MR. RAMSEY: Objection. Vague.
` THE WITNESS: Yeah, I mean, that's a pretty
`broad question.
` I mean, in some ways, we've never made public
`disclosure. We -- except I -- I -- I mean, it depends
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 12 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 213
`
`on what you mean.
` I mean, obviously, the -- the product exists in
`a sense that it's a public disclosure. And I'm sure
`there have been casual mentions to customers or
`important suppliers or other people, but I -- I don't
`know. I mean, that's a really big question.
`BY MR. LEDAHL:
` Q Now, earlier, you talked about when -- the idea
`that a potential user could download the -- the
`demonstration or the -- the alpha version of the Clango
`application client, right?
` A Right.
` Q And I think you said -- but I want to make sure
`I understood correctly -- that there was some kind of
`end-user license agreement, but you didn't know what the
`terms of that were?
` A This is a vague memory that -- in the initial
`version, there was some kind of end-user license
`agreement. And, no, I do not.
` If -- if that's true, I don't remember the
`terms.
` Q And is the same true for the beta, that you
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 13 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 297
` MR. RAMSEY: I want to designate the entire
`thing confidential, attorney eyes only. And we'll all
`work in good faith as we proceed to de-designate, if
`necessary.
` THE REPORTER: Read and sign?
` MR. RAMSEY: Yes.
` (END TIME: 4:39 p.m.)
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
`
`Case 1:14-cv-02396-PGG-SN Document 234-8 Filed 11/11/20 Page 14 of 14
`
`9/4/2019
`
`Network-1 Technologies, v. Google LLC and Youtube LLC
`Confidential - Attorneys' Eyes Only
`
`Erling Wold 30(b)(6)
`
`Page 298
`
`STATE OF CALIFORNIA )
`COUNTY OF SAN FRANCISCO )
`
` I, Lana L. Loper, a Certified Shorthand
`Reporter, do hereby certify:
` That prior to being examined, the witness in
`the foregoing proceedings was by me duly sworn to
`testify to the truth, the whole truth, and nothing but
`the truth;
` That said proceedings were taken before me at
`the time and place therein set forth and were taken down
`by me in shorthand and thereafter transcribed into
`typewriting under my direction and supervision;
` I further certify that, pursuant to FCRP Rule
`30(e)(1), the review and signature by the witness:
`
` __X_ was requested by the deponent or a
` party before the conclusion of the
` deposition;
` ____ was not requested by the deponent
` or a party before the conclusion of the
` deposition;
`
` I further certify that I am neither counsel
`for, nor related to, any party to said proceedings, nor
`in anywise interested in the outcome thereof.
` In witness whereof, I have hereunto subscribed
`my name.
`
`Dated: September 17, 2019
`
`___________________________________________________
`LANA L. LOPER, RMR, CRR, CCP, CME, CLR, CSR 9667
`
`1
`2
`
`3 4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`11
`
`12
`13
`
`14
`15
`16
`
`17
`18
`
`19
`20
`21
`
`22
`
`www.DigitalEvidenceGroup.com
`
`Digital Evidence Group C'rt 2019
`
`202-232-0646
`
`
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
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.
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.
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
