Apple Computer Inc. v. Burst.com, Inc.

Filing 165

Declaration of Nicholas Brown in Support of 164 MOTION to Strike 146 Declaration in Opposition, 150 Declaration in Opposition, 147 Declaration in Opposition MOTION to Strike 146 Declaration in Opposition, 150 Declaration in Opposition, 147 Declaration in Opposition filed byApple Inc.. (Attachments: # 1 Exhibit A# 2 Exhibit B# 3 Exhibit C# 4 Exhibit D)(Related document(s) 164 ) (Brown, Nicholas) (Filed on 9/17/2007)

Apple Computer Inc. v. Burst.com, Inc.

Doc. 165 Att. 1

Case 3:06-cv-00019-MHP

Document 165-2

Filed 09/17/2007

Page 1 of 18

Exhibit A

Dockets.Justia.com

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2 3 4 5

9/11 Filed 09/17/2007/2007 Allen Gersho2 of 18 Page

UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF CALIFORNIA SAN FRANCISCO DIVISION APPLE COMPUTER, INC., Plaintiff/Counterdefendant, vs. BURST.COM, INC., Defendant/Counterclaimant. _________________________________/ DEPOSITION OF ALLEN GERSHO Tuesday, September 11, 2007 VOLUME I Pages 1 - 222 REPORTED BY: VALERIE J. EAMES, CSR NO. 9021, RPR, CRR Case No. C06-00019 MHP

APPEARANCES (Continued)

ALSO PRESENT: DAN MOTTAZ VIDEO PRODUCTIONS, LLC BY: CHRISTOPH GEMES

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

182 Second Street, Suite 202 San Francisco, California [phone] 94105

JAYNA WHITT, PATENT COUNSEL, APPLE COMPUTER (Via Telephone)

9/16/2007 3:14 PM

9/16/2007 3:14 PM

3

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

APPEARANCES

1 2 3 WITNESS

INDEX

EXAMINATION PAGE

For PLAINTIFF/COUNTERDEFENDANT: WEIL, GOTSHAL & MANGES LLP BY: GARLAND T. STEPHENS ATTORNEY AT LAW 700 Louisiana, Suite 1600 Houston, Texas 77002

4 5 6 7 8 9 10 11 12 13 252 EXHIBITS: 251 Declaration of Allen Gersho in Opposition to Apple's Second Motion for Summary Judgment of Invalidity Declaration of Allen Gersho in Support of Burst's Opposition to Apple's Motion for Summary Judgment of Invalidity Based on Kramer and Kepley Patents 253 254 CompuSonics document CompuSonics Corporation document with a "received" stamp date of 6/17/87 ALLEN GERSHO By Mr. Stephens

8

PAGE 54

[phone]; Fax [phone] [e-mail]

57

For DEFENDANT/COUNTERCLAIMANT: HEIM, PAYNE & CHORUSH LLP BY: LESLIE V. PAYNE ATTORNEY AT LAW 6710 Chase Tower 600 Travis Street Houston, Texas 77002

14 15 16 17 18 19 20 21 22 23 24 25

97 101

[phone]; Fax [phone] [e-mail]

EXHIBITS MARKED PREVIOUSLY AND APPENDED Exhibit No. 109, 364

9/16/2007 3:14 PM

2

9/16/2007 3:14 PM

4

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2 3

9/11 Filed 09/17/2007/2007 Allen Gersho3 of 18 Page

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

INDEX (continued)

BE IT REMEMBERED that pursuant to Notice, and on Tuesday, September 11, 2007, commencing at the hour of 8:50 a.m., thereof, at 1901 Avenue of the Stars, Suite 950, Los Angeles, California, before me, VALERIE J. EAMES, a Certified Shorthand Reporter, the following proceedings were had: PROCEEDINGS THE VIDEOGRAPHER: Good morning. This marks the

SECTION OF TRANSCRIPT REQUESTED MARKED BY MR. STEPHENS PAGE 47 LINE 14

4 5 6 7 8

QUESTIONS NOT ANSWERED BY THE WITNESS PAGE 12 28 29 43 43 46 47 50 51 55 74 77 77 79 83 LINE 12 10 3 4 24 8 24 7 20 12 14 1 12 14 20

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

beginning of Volume I, videotape number 1 in the deposition of Mr.-- of Dr. Allen Gersho in the matter of Apple Computer, Incorporated versus Burst.com. And the

case has been filed with the United States District Court Northern California of District -- Northern California District Court of San Francisco Division, Case No. C06-00019 MHP. Today's date, September 11th, year 2007, and the time on the video monitor is approximately 8:50 a.m. The

location of the deposition is 1901 Avenue of the Stars, here -- Suite 950, Los Angeles, California. the law offices of Susman & Godfrey. And if we could please ask the attorneys to introduce themselves for the record. MR. STEPHENS: Garland Stephens of Weil, Gotshal And we're at

& Manges representing Apple. MR. PAYNE: Les Payne for defendant Burst.com.

9/16/2007 3:14 PM

5

9/16/2007 3:14 PM

7

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

QUESTIONS NOT ANSWERED BY THE WITNESS (continued) PAGE 85 86 90 97 101 102 102 103 104 105 105 123 125 132 155 156 160 214 215 215 215 216 LINE 24 7 25 7 15 10 16 15 20 10 25 16 20 2 7 7 25 21 9 16 23 3

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

THE VIDEOGRAPHER:

My name is Christoph Gemes.

I am a video specialist, and I am employed by Dan Mottaz Video Productions, LLC, 182 Second Street, Suite 202, San Francisco, California 94105. And if we could please ask the court reporter to state her name for the record. THE REPORTER: Valerie Eames. Very well. And at this time

THE VIDEOGRAPHER:

if we could please ask you to swear in the witness. ALLEN GERSHO, called as a witness by the Plaintiff/Counterdefendant, and who, having been administered the oath by me, was examined and testified as hereinafter set forth: THE VIDEOGRAPHER: Counselor. EXAMINATION BY MR. STEPHENS: Q. A. Q. Good morning, Dr. Gersho. Good morning. Could you please state and spell your name for Okay. At this time,

the record. A. Q. A. Q. Allen Gersho, A-l-l-e-n G-e-r-s-h-o. And what's your home address? 4604 Via Gennita, Santa Barbara, California. You've submitted a couple of declarations in

9/16/2007 3:14 PM

6

9/16/2007 3:14 PM

8

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2 Can 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

9/11 Filed 09/17/2007/2007 Allen Gersho4 of 18 Page

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

A.

No.

I -- nothing in my personal life would be

A.

In some context some might say so, but I'm just I can't be specific here. So you think that it did have hardware Is that right?

connected. Q. How about a particular graduate student?

speculating. Q. Okay.

you recall a particular graduate student that worked with the Sun workstations and the time roughly when they were under your supervision? A. No. I had many students and visiting It's difficult to remember the time periods

for digitizing speech. A. Q. Yes.

Did you ever have hardware for digitizing audio

with a wider bandwidth than just speech? A. At some time we -- I believe we may have had

researchers.

when each was there. Q. Sun workstations were available by the Right?

such a capability in my laboratory. Q. A. Q. How about with the PDP11-24? No. Do you recall when you got the capability of

mid-1980s, though. A. Q.

I don't know that. You're just not sure. Okay. Well, let's talk a little bit about what How did you use that in the

digitizing speech on a PDP11-24? A. I can't say with certainty. I think it came

you did with the PDP11-24.

with -- we may have bought the audio equipment together with the computer at the same time. Q. A. Q. A. remember. Q. Can you think of any other kind it might have Okay. So that would have been in roughly 1981?

laboratory in connection with the development and research on speech, audio, and video compression? A. The students would write software to evaluate

If I'm correct about that recollection, yes. Did the PDP11-24 have magnetic disk storage? I assume it had some form of storage. I don't

algorithms using this computer. Q. audio? A. Q. I think so. I think so. Did it include hardware that could digitize

22 23 24 25

Would you agree that by the mid-1980s, that

had instead of disk storage? A. Q. It had floppy disk drives. And floppies are a form of magnetic disk.

people of skill in the art in this case -- you understand what I mean by that -- right? -- people of skill in the

9/16/2007 3:14 PM

17

9/16/2007 3:14 PM

19

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

relevant art to this case? A. Q. Yes. Would you agree that by the mid-1980s people of

1 2 3 4 5 6 I think there's a 7 8 9 10 11 12 13 14 15 Right? 16 17 18 19 20 21 22 23 24 25

Right? A. Q. A. Q. A. computer. That's true. Did it have a high-speed magnetic disk? I don't recall. It had the random access memory. Right?

skill in the relevant art to this case were quite familiar with hardware for digitizing audio and storing it in computer systems? A. Some were and some weren't.

I had no awareness of the internal design of the Now, I can make an assumption, but I don't

variety of backgrounds of people that still qualify as skilled in the art. I'd like to correct something I said before -Q. A. Okay. -- that our equipment may have had the ability When you said "digitize audio," I

specifically know that. Q. Well, virtually all general purpose computers in Right? I know it

the '80s had random access memory. A.

I'm not an expert on computers.

was -- it's common to have random access memory in digital computers. Q. And it was common in the mid-'80s for digital Right?

to digitize voice.

don't mean -- I wasn't agreeing to digitizing any form of audio, such as wideband audio. Q. A. But speech is a form of audio. It depends -MR. PAYNE: Go ahead. Object as to form. You can answer. The term "audio" does not have a Some will say it does

computers to have random access memory. A. Q. I suppose so.

And it was common for them to have magnetic disk Right? By "them," I

drives in the mid-1980s as well. mean digital computers. A. Q.

THE WITNESS:

I think that's probably correct. Would you agree, then, that it's likely that

precise definition out of context.

not include speech, depending on the context. BY MR. STEPHENS: Q. But some would say it does include speech, Right?

your PDP11-24 had a magnetic disk drive? MR. PAYNE: BY MR. STEPHENS: Q. By that I mean a high-speed one, not a floppy. Objection. Form.

again, depending on the context.

9/16/2007 3:14 PM

18

9/16/2007 3:14 PM

20

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2 3 4

9/11 Filed 09/17/2007/2007 Allen Gersho5 of 18 Page

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 it had.

MR. PAYNE: THE WITNESS:

Same objection. I'm not sure what kind of storage

Q. A.

Yes.

Thank you.

I don't recall if we really digitized our own If so it was -- it would have been very

source material. infrequent.

BY MR. STEPHENS: Q. Okay. Now, you said that your graduate students

Mostly it was -- there was interest in

5 6 7 8 9 10 11 12 13 14 15 16 17 18

listening to speech files. Q. But your system certainly had the capability of Right?

used the PDP11 to write software to evaluate algorithms. Is that right? A. Q. Yes. And those algorithms included audio compression Is that right?

recording your own source material. A.

I believe it had the capability of recording

speech from a microphone input. Q. And you're just not sure whether people used Is that right?

algorithms. A. Q.

Where audio means voice. Okay. Now, the -- during your tenure from 1980

that in their research. A. Q. That's right.

to 1998, you also worked on wideband audio compression as well. A. Q. Is that right? I did. Was there work in the signal compression

Now, it certainly would have been within the

normal skills of a graduate student working in your laboratory to record some speech using the analog-to-digital facility, and then compress it using a compression algorithm on the PDP11. A. A couple things here. Right?

laboratory in the '80s on wideband audio compression as well? A. I'm not certain, but my best recollection is --

The -- certainly some --

19 20 21 22 23 24

some, perhaps not all, of my students would have known how to digitize speech from -- coming in from a microphone and use it as input to a speech compression algorithm. Q. Okay. In fact, that was one of the main reasons

began -- it may have begun in the early '90s. Q. A. Q. kind? A. I don't think so. I don't recall, but I don't Was the PDP11 connected to any other computers? No. Was it connected to external storage of any

for developing speech compression algorithms is to compress speech that you encode using an

25

9/16/2007 3:14 PM

21

9/16/2007 3:14 PM

23

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

think so. Q. Now, in the process of evaluating algorithms 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 Analog to digital? 25

analog-to-digital encoder. A. No. MR. PAYNE: THE WITNESS:

Right?

speech compression, did your graduate students in the mid-'80s use speech compression algorithms to compress speech that was digitized into the PDP11? A. As I recall, most of the source material for

Objection. No.

Form.

The main reason was to study

speech algorithms and evaluate how they work, not the -the purpose was not to compress speech files. study compression algorithms. BY MR. STEPHENS: Q. But the reason to study compression algorithms It was to

experimenting with speech compression may have been provided to us by an outside organization. Q. A. What was the name of that outside organization? In different time periods, I've had some

is to compress speech that has been encoded using an analog-to-digital encoder. A. Q. A. No. How would you do it, then? No. The reason for studying compression of Correct?

relationships with a few companies, and one or another company may have -- you know, may have provided us with some files that are used for demonstrating speech compression. time. And so I'm not sure which company at which

But Bell Laboratories definitely at some time gave

speech was help students get Ph.D.s to publish papers, and perhaps to achieve results that might be useful in the real world. Q. world? A. Engineers in industry reading our publications How would those results be useful in the real

me some source files. Q. A. Q. Any others that you're aware of? I don't recall specifically. Did your graduate students ever use the speech

digitization facility that your PDP11 had? A. They used the digital-to-analog facility

may find it useful to apply one of the algorithms to some real world product. Q. And what kind of real world product might they

frequently. Q. me -A. What about the analog to "dility" -- excuse

apply it to? A. One of the main applications of voice

9/16/2007 3:14 PM

22

9/16/2007 3:14 PM

24

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2 3 4 5 6 7 8

9/11 Filed 09/17/2007/2007 Allen Gersho6 of 18 Page

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 here.

MR. PAYNE: THE WITNESS:

Objection.

Form.

Q.

Okay.

Is it possible to use a digital-to-analog

It seems there are many components

encoder and a digital DPCM encoder to get the same output substantially -- from substantially the same input as an analog DPCM encoder without regard to the complexity of the circuitry or cost? MR. PAYNE: THE WITNESS: Objection. Form.

I want to be sure I give a correct answer as I

don't want to make a sweeping statement. BY MR. STEPHENS: Q. A. clear. Q. Okay. Well, I want to make sure you're clear. Okay. You put in too many pieces here for me to be

You said "the same output," and

the answer would be no to the same output. BY MR. STEPHENS: Q. A. But I said "substantially." Well, you said -- "substantially the same

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

So if there's -- you need some clarification, just let me know, and I'm happy to add some clarification. A. I've already said that you can have -- it's

input," you said. Q. Well, I meant to say both substantially the same I think that's

possible to have improved efficiency using some form of DPCM compared to just using a PCM. Okay?

input and substantially the same output. what I said.

So I'm not sure beyond that what you're trying to find out. Q. I'm trying to find out if it's possible to use a

But that's fine if it wasn't clear.

So with that in mind, would you please answer the question. MR. PAYNE: THE WITNESS: Objection. Form.

digital DPCM encoder with an analog-to-digital encoder in a way that's substantially equivalent to using analog DPCM. That's my question. MR. PAYNE: Is it possible to use -Form.

I believe that it is possible to

get a similar quality audio output -- sorry -- a similar quality digital representation of audio by the two alternative methods: one of analog-to-digital

Objection.

THE REPORTER:

And I didn't get -- "I'm trying

to find out if it's possible to use a digital DPCM encoder" -///

conversion, followed by a digital DPCM circuit, as with a -- a DPCM encoding system, depending on the -- if one suitably chooses the number of bits and assigns these

9/16/2007 3:14 PM

37

9/16/2007 3:14 PM

39

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

BY MR. STEPHENS: Q. -- to get substantially equivalent results to

1 2 3 Form. 4 5 6 7 Form. 8 9 10 11 12 13 14 15 16 Is there a question? Yes. 17 18 19 20 If you don't 21 22 23 24 25

correctly. BY MR. STEPHENS: Q. So the audio quality would be equivalent. Is

using an analog DPCM encoder -MR. PAYNE: BY MR. STEPHENS: Q. Using an appropriate analog-to-digital converter Objection.

that right? MR. PAYNE: THE WITNESS: Objection. Form.

The audio quality represented by

with the DPCM encoder. MR. PAYNE: THE WITNESS: Objection.

the bit stream coming from either of the two options might be similar. BY MR. STEPHENS: Q. And what about the number of bits required to

The term "substantially equivalent There may -- the

results" is a little vague here.

results may -- if the results include the complexity, the practicality of implementing it -BY MR. STEPHENS: Q. Specifically excluding the complexity and I'm only asking you to

represent that audio signal? A. It is possible to have fewer bits coming from Oh, sorry. Wait a minute. I

the circuit with DPCM. withdraw that.

practicality of implementing it. focus on the output. MR. PAYNE:

Both cases we're talking about -- about starting with an analog input. Q. Yeah. In fact, we can say it's the identical

MR. STEPHENS: MR. PAYNE: BY MR. STEPHENS: Q.

analog input. A. Okay. It is possible to have some saving in But

I don't know the question.

bits by using DPCM rather than not using DPCM. that's not the issue right now.

Go ahead and answer, sir.

You're just saying

understand it, please let me know, and I'll be happy to clarify it. A. It might help if you repeat the question so I

that -- oh, the bit rate of the output is what you're trying to get. Q. So what I'm after is whether or not a person

make sure I'm answering the correct question.

having the appropriate knowledge of analog and digital

9/16/2007 3:14 PM

38

9/16/2007 3:14 PM

40

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2 3

9/11 Filed 09/17/2007/2007 Allen Gersho7 of 18 Page

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

computer to computer.

I have done some copying of files

1990s." Q. A. answer. Q. I guess what I'm asking is is that sentence What did you mean by that sentence? I think it's pretty clear. I'm not sure how to

to a flash drive and then from the flash drive to another computer. Q. Okay. And you typically have to wait while the Right?

4 5 6 7 8 9

copy to the flash drive is made. A. Q. Yes.

intended to mean only that it wouldn't have been obvious to send audio in a burst mode in view of Kramer or something more general? A. At the time I was thinking about Kramer. So I

And all other things being equal, it's desirable

for that process to happen faster rather than slower. Right? MR. PAYNE: THE WITNESS: Objection. Form.

10 11 12 13 14 15 16

don't know if -- I don't remember if -- at the time I wrote this, if I was thinking more broadly or not. Q. Would it have been obvious to a person of

I usually have other things to do So it doesn't bother

and -- after I give a copy order. me.

But I suppose some people may be impatient.

ordinary skill in the art to transmit audio in a burst mode if they had the Kepley patent in their possession in 1987? MR. PAYNE: Same objection. Don't answer that.

BY MR. STEPHENS: Q. UUCP? A. using it. Q. Were your Sun workstations that you used at your I've heard the term, but I don't recall ever Did you ever use UNIX to UNIX Copy protocol,

17 18 19 20

Beyond the scope of the declaration. BY MR. STEPHENS: Q. So your declaration is not intended to convey

research center networked? A. Certainly in the '90s. I don't recall when

the view that Mr. Lang's alleged inventions are not obvious in view of Kepley. A. Is that right? I don't recall right

21 22 23 24 25

they -- when we started to have network workstations. Q. And those Sun workstations are UNIX Right?

I need to think about it.

now what my intent was, if it was broader or not. MR. PAYNE: please. Can you read back the question,

workstations. A.

They were a version of UNIX offered by Sun

9/16/2007 3:14 PM

53

9/16/2007 3:14 PM

55

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

Microsystems. Q. A. sure. Q. Okay. Was compression used in -- well, I have Do you know if they supported UUCP? I don't want to speculate. I don't know for

1 2 3 4 5 6 I think the 7 8 9 10 11 12 13 14 15 16 Could you just 17 18 19 20 21 22 23 24 25

(The record was read as follows: "Q So your declaration is not intended to

convey the view that Mr. Lang's alleged inventions are not obvious in view of Kepley. Is that right?") THE WITNESS: I do not believe I was thinking

something else. If you could mark this, please. next exhibit number is 251. (Exhibit No. 251 was marked.) BY MR. STEPHENS: Q. A. Q. Do you recognize your declaration? Yes. And this is, to be clear, your Declaration in

about Kepley in this context -BY MR. STEPHENS: Q. A. Q. Okay. -- one way or the other. Have you expressed a view in connection with

this litigation whether or not Mr. Lang's alleged inventions are obvious in view of Kepley? A. I don't recall. You know, I've had a number of

Opposition to Apple's Second Motion for Summary Judgment of Invalidity. A. Q. Yes. If you turn to paragraph 38. Right?

phone conversations -MR. PAYNE: conversations. declaration -BY MR. STEPHENS: Q. My intention is to ask you about your That's what I meant. Don't talk about phone

If you're going to limit it to the

read the first sentence out loud? A. "Not only does Kramer not suggest faster

than realtime transmission, it would have been entirely surprising and unexpected for anyone to conceive of or propose the idea of transmitting audio in a burst mode given the statement of technology and the consumer product market in the late 1980s and early

declarations. A. Q.

I don't -- no, I don't recall right now. So you're not sure whether you did or not.

You're not saying you didn't. A. That's correct.

9/16/2007 3:14 PM

54

9/16/2007 3:14 PM

56

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2

9/11 Filed 09/17/2007/2007 Allen Gersho8 of 18 Page

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

Q.

Can you think of any reason other than

Q. A. Q.

Including the Audix system.

Right?

forwarding it? MR. PAYNE: Objection. Form. I'm going to

I understand that that was developed by AT&T. And the next sentence there says,

3 4 5 6 7

instruct him not to answer. compression.

This has nothing to do with

"Advantageously, the transmission facilities are high-speed digital facilities of the type used for computer data file transfers." A. Q. That's what it says. Then it goes on to say, "The use of digital Right?

I'm instructing you not to answer. BY MR. STEPHENS: Q. If you'll turn now to column 12. Line 47,

8 9 10 11 12 13 14 15 16 17 18

there's a section that begins "Voice Mail Message Forwarding." A. Q. Yes. That shows that the system is designed for Right? Do you see that?

high-speed transmission facilities of speed greater than 9.6 kilobits per second enables the exchange of digitally encoded and compressed voice mail messages faster than realtime speech." A. Q. Do you see that?

forwarding voice mails. MR. PAYNE: declaration.

Yes, I see it. Do you have any reason to disagree with that? MR. PAYNE: I'm going to object and instruct him

You're beyond the scope of his

You're getting into issues of transmission.

I instruct him not to answer this line of questioning, unless you can explain to me why it's tied to compression. MR. STEPHENS: I've already explained the

not to answer.

It's beyond the scope of his declaration.

It's in transmission areas. BY MR. STEPHENS: Q. In column 14, around line 5, there's a sentence

19 20 21 22 23 24 25

relevance to his declaration, and I'm not going to continue. You can make your instructions, and we'll just

that begins, "At step 706, the destination voice mail service system transmit protocol identifiers to originating voice mail service system to indicate the signaling format." A. Q. Yes. Is that something that's referred to as a Do you see that?

take it up with the judge. MR. PAYNE: BY MR. STEPHENS: Q. Now, on the next column, 13, there's a Okay.

9/16/2007 3:14 PM

77

9/16/2007 3:14 PM

79

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

discussion of the data call connection to transmit voice mails. A. Q. A. Right? I don't know where -- where is it? So, for example, starting at line 29. Yes, there's the discussion of data call

1 2 3 4 5 6 7 8 9 10 11 12 13

protocol negotiation? A. Q. A. Q. I don't know. You don't know about protocol negotiation? No. Do you know if the Kepley system would support

connections in this patent. Q. And it says there at line 29, that "This data

more than one protocol for transmission of voice mail messages? A. Q. I don't know. In column 14, around line 58, there's a sentence

call connection can be established using various types of transmission facilities." A. Q. Yes. And in the 1980s, AT&T had various types of Do you see that?

that says, "This voice mail message illustrated in Figure 4 can be of any length." A. Q. Yes. Is there any inherent limit on the length of a Do you see that?

transmission facilities that it made available to its business customers. A. Right? I mean I know that they had

I don't know that.

14 15 16 17 18 19 20 21 22 23

voice mail message that can be stored using voice mail compression mechanisms of the kind that you're familiar with that existed in the mid-'80s? A. Well, there has to be a length limitation.

various transmission facilities, but I don't know what's made available to customers other than the operating companies which it owned at the time. weren't customers. Q. So you don't know whether what's described in It wasn't -- they

There's a limited -- disk storage space was expensive in those days. It was -- they had a limited disk storage.

here about making data call facilities of various types available to support voice mail systems is an accurate description of AT&T's business at the time. right? A. I believe that voice mail systems was a part of Is that

They're not going to let someone talk forever and leave a message of three hours. Q. Okay. But other than the limit imposed by the

amount of storage available, was there any other limit? A. There may be technical limits involving with

24 25

AT&T's business.

the -- involving the implementation of the compression.

9/16/2007 3:14 PM

78

9/16/2007 3:14 PM

80

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
I 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 I

9/11 Filed 09/17/2007/2007 Allen Gersho9 of 18 Page

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

I can't -- I don't know -- not just the compression. mean the interfacing and the circuitry and so on.

whole -- the sign of the system, the engineering would be very different if it was limited to one user. Q. A. Q. Okay. You could use an answering machine. If you'll look at column 16 now of the Kepley

don't know enough about the hardware to -- and how things are implemented in circuits to say one way or the other. Q. So you don't have any reason to disagree with

that statement -- is that right? -- that the voice mail message can be of any length, other than there has to be some limit to the amount of storage available? MR. PAYNE: THE WITNESS: Objection. Form.

patent -- actually, before we go there, at the bottom of column 15, at line 65, there's a sentence that says, "The transmission of the digitally encoded, compressed voice." Do you see that? A. Q. Column 15, line 65? Yes. And that sentence carries onto the top of Could you just read that sentence out

At the moment I can't think of

specific reasons other than the ones I've already referred to that -- for putting a limit on the length of the message. BY MR. STEPHENS: Q. drive? A. One, the disk storage space. And the other is Which ones did you refer to other than the disk

the next column. loud. A.

"The transmission of the digitally

encoded, compressed voice mail message over high-speed digital facilities also is timewise efficient compared to transmitting the analog version of the voice mail message." Q. Could you explain that, please? MR. PAYNE: Same objection. Same instruction,

17 18 19 20 21 22

that implementation constraints for interfacing circuitry, which I don't understand well enough that there could be issues there. Q. some? A. Q. A. That's true. Okay. I can think of another limitation. So you don't know of any, but there could be

beyond the scope of his declaration. BY MR. STEPHENS: Q. In column 16 there, at line 11, there's a

23 24 25

sentence that says, "Destination voice mail service

9/16/2007 3:14 PM

81

9/16/2007 3:14 PM

83

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

Q. A.

Okay. That the number of calls coming in -- the

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 The 25

system receives each 128 byte segment, and it goes on from there." A. Q. Yes. That's talking about the destination voice mail Do you see that?

computer processor has limited capability, and it just can't keep processing one call forever when there's -- it can be overloaded and it can only handle so many simultaneous calls. Q. But that's true regardless of how long a Right?

system receiving the digitally encoded, compressed voice mail message. A. Q. A. Q. Right?

Perhaps I could read the context a bit. Sure. Yes. So that's a description of the destination voice Go ahead.

particular message is. A. Q. A.

Well, it's just -How many are happening at the same time. No. The longer it is, the more it ties up some

mail saving -- voice mail service system receiving the digitally encoded, compressed mail message and then storing it on a magnetic disk. MR. PAYNE: THE WITNESS: Objection. Right? Form.

of the resources of the system, and therefore it reduces the ability of the system to handle as many other messages as it could. Q. So that's a reason to make things happen quickly Right?

It's -- it doesn't mention

and short. A.

compressed, but it is -- it's receiving a segment of the message. voice. BY MR. STEPHENS: Q. Right? A. Q. That's what it says. And that database processor we saw had a Right? Then it stores it in the database processor. So presumably that's the digital compressed

That would be a reason to put a limit on the

duration of the message. Q. Okay. But that's not an engineering limitation.

That's just a decision about how to manage a bunch of simultaneous calls. A. Q. Is that right?

It is an engineering limitation, I think. If there was only one user for the system, it Right?

wouldn't be a limitation. A.

magnetic disk for storage. A. I don't know.

The system would be defined differently.

9/16/2007 3:14 PM

82

9/16/2007 3:14 PM

84

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2 3 4 5 6 7 8 9 10 11

9/1 Allen Gersho Filed 09/17/20071/2007 Page 10 of 18

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

Q.

Look back at Figure 2. MR. PAYNE: THE WITNESS: Objection. Form.

(Recess taken.) THE VIDEOGRAPHER: We're back on the record. The time is

Figure 2 shows an element number

This is the beginning of tape 2. approximately 10:58 a.m. Counselor. BY MR. STEPHENS: Q.

203 that looks like a symbol for a magnetic disk storage. BY MR. STEPHENS: Q. So just to be clear, then, what's happening in

the patent, with reference to Figure 1, we have the system 110 receiving a voice mail message from a handset where somebody's leaving a voice mail message. with me now? A. Q. A. Q. Figure 2? Figure 1. Oh, yes. -- in Figure 1 the voice mail service system 110 So you have -Are you

Dr. Gersho, looking back at your second

declaration, which I think was 251, if I remember right. That's the Declaration of Allen Gersho in Opposition to Apple's Second Motion. A. Q. A. Q. Yes. Do I have the number right, 251, on the exhibit? Yes. In paragraph 38 -- actually, before we get Do you have that?

12 13 14 15 16 17 18 19 20 21 22 23 24 25

receives a voice mail message from somebody using a telephone handset. A. Q. Right?

there, let me ask another question or two about Kramer -excuse me -- about Kepley. Now, you worked for AT&T for a number of years. Right? A. Q. Yes, I did. Did you file any patent applications while you

Yes, that's one option here. Okay. And then the voice storage processor

digitally compresses it and stores it in the database processor. A. Right?

The voice storage processor does compress it

were working for AT&T? A. Q. Yes. So there are some AT&T-issued patents that

and -- yes, I believe it's stored in the database processor. Q. Then the database processor transmits it over

you're named as an inventor on? A. That's correct.

high-speed digital facilities in a manner that's timewise

9/16/2007 3:14 PM

85

9/16/2007 3:14 PM

87

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

efficient, compared to sending the analog version to the voice mail system 150. MR. PAYNE: Right? It's beyond

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Q.

Is it your experience that when AT&T filed for a

patent, AT&T as a company did so because they believed there was an invention there worth patenting? A. I couldn't say that. I think they wanted to That was my impression.

I'm going to object.

the scope of the declaration. answer. BY MR. STEPHENS: Q.

Instruct him not to

patent everything they could. Q. Okay.

But it -- was it your view or your

So then the database processor transmits that to Right? Same instruction.

understanding that AT&T's policy was to fulfill all the requirements of the patent office, laws and regulations, in filing an application? A. I don't know their policy. They never discussed

the voice mail service system 150. MR. PAYNE: BY MR. STEPHENS: Q. Same objection.

Somehow the voice mail gets from database Right?

it with me. Q. Do you think AT&T would file patent applications

processor 113 to the database processor 153. A. Q. A. Q. Where's 153?

for things that didn't work? MR. PAYNE: THE WITNESS: Objection. Form.

That's in the voice mail service system 150. Yes, I believe so. And then there, that digitally compressed voice

I don't think they filed any

perpetual motion machines, but I know one patent that they filed -- that was filed and issued of mine that was totally useless. BY MR. STEPHENS: Q. "Useless" is different than "not working." Do you think they filed patent applications for things that didn't work? A. I have no reason to assume that, but -- no, I I don't know. Who knows. There

mail file is stored in database processor 153 on a magnetic disk. A. Yes. MR. STEPHENS: change the tape. THE VIDEOGRAPHER: approximately 1:48 a.m. THE WITNESS: We're off the record at Why don't we take a break to Right?

20 21 22 23 24 25

That concludes tape 1. Did you say 1:48? I'm sorry. 10:48. 10:48.

shouldn't speculate. may be a reason.

THE VIDEOGRAPHER:

9/16/2007 3:14 PM

86

9/16/2007 3:14 PM

88

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2 3 4 5 6

9/1 Allen Gersho Filed 09/17/20071/2007 Page 11 of 18

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

Q.

Okay.

Well, looking at the Kepley patent, that Right?

MR. PAYNE:

Objection.

Form.

I'm going to

was assigned to AT&T. A. Q. Uh-huh. Yeah.

instruct him not to answer. THE WITNESS: MR. PAYNE: about Kepley. declaration.

It's beyond the scope.

May I answer? He's asking a specific question

And it was published no later than Right? That's the issue date on the

December 1988. patent? A. Q. A. Q.

So that's beyond the scope of your So I'm going to instruct you not to answer.

If you say so. That's the issue date anyway? Uh-huh. Yeah.

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

BY MR. STEPHENS: Q. Well, Kepley was part of the state of the Right?

technology at the time. A.

Would you agree that that reflects the state of

I gather Kepley was involved in digital I have no reason to believe he was involved

the art in voice mail systems at the time? A. I don't have any opinion on that, because I I don't know how many other

switching.

in compression. Q. Well, the device described in there uses Right?

don't know the field.

patents were filed prior to this. And to clarify my previous answer, they may indeed have filed for patents that don't work. Because

compression. A.

I believe he just assumed that the people who

know compression would provide him with a compression algorithm to fit into their box. Q. A. And that was perfectly possible. I don't know. Right?

sometimes patents are drawing-board things without actually implementing and it may turn out it doesn't work, and we may not know until later. Q. Right? A. I have no reason to believe they would do it Okay. But they didn't do it intentionally.

Depends what the constraints are

and what the objectives are. Q. Well, there were many voice codecs, as you point

22 23 24 25

out, that were known to people of skill in the art at the time. A. Q. Right? Yes, there were. Okay. And some of them would have been suitable

intentionally. Q. Right? AT&T was pretty good at making things work.

9/16/2007 3:14 PM

89

9/16/2007 3:14 PM

91

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

A.

We had a lot of talented engineers at Bell

1 2 3 4 5 6 7 8 9 10 11 12 13

for use in a voice mail system described by Kepley. Right? A. I'm not sure -- I'm not sure I could make a

Laboratories, AT&T Bell Laboratories. Q. And they were obviously very good at high-speed Right?

data transmission. A.

sweeping statement about -- depends what his constraints and objectives were. Q. Is it your testimony that there was no codec

Most of the areas that I was involved in were So I don't know

not high-speed data transmission. specifically, you know.

There were some -- I know there They put in a lot of money

known to people of skill in the art that could have been used in Kepley in 1988? A. Q. A. No, that's not my testimony. Was there any? I can't say, unless we specifically define the

were some notable screwups.

into something that -- a lot of time in developing something that turned out not to be useful in digital transmission. Q. Right? A. Yes. There was a lot of expertise in the area Okay. That was in the research area. And they were good at compression.

constraints of what they need to achieve in that context. Q. patent. I'm just asking about what's describe in the So if a person of ordinary skill read the patent

14 15 16 17 18 19 20 21 22 23 24

of voice compression. Q. Now, looking back at your declaration, we And this is, again, the

in 1988, would they have been able to find a compression algorithm that would work in that system that's described in there? A. See, if I recollect -- recollect correctly, that

started to look at paragraph 38.

paragraph where in the first sentence you say that "it would have been entirely surprising and unexpected for anyone to conceive of or propose the idea of transmitting audio in a burst mode given the state of the technology and the consumer product market in the late '80s and early '90s." A. Q. Right?

he was describing a 16-kilobit-per-second rate or -- is that -- I don't remember. question more precisely. If he was seeking a compression method that produced 16-kilobit-per-second speech, I will definitely agree that there were compression algorithms for voice speech that could achieve 16 kilobit rate of -But I'm trying to define this

Yes, I said that. But Kepley did conceive of it. Right?

25

9/16/2007 3:14 PM

90

9/16/2007 3:14 PM

92

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2 3 4 5 6 7 8 9 10 11

9/1 Allen Gersho Filed 09/17/20071/2007 Page 12 of 18

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

Q.

Okay.

And could have been used in the system Right?

of or propose the idea of transmitting audio in a burst mode is wrong. A. Correct?

that's described there. A.

Again, that depends on the whole family of

Could you give me the first part of that

constraints, engineering constraints, but whether -- what would be suitable. right bit rate. Q. I'm just asking about the constraints that are Is there anything in there that It's not sufficient that it had the

question again? Q. Yes. If you assume that Kepley, the patent, was

part of the state of the technology in the late '80s, then your statement that it is -- it would have been entirely surprising and unexpected for anyone to conceive of or propose the idea of transmitting audio in a burst mode is wrong? A. I can't say that. First of all, you want to

described in the patent.

you can point to that says, no, there's no codec that would allow you to build the system that's described here? A. Well, I don't know the patent well enough to

12 13

assume that Kepley was part of the state of technology. That's -- under that hypothesis -- hypothetically, because I don't know for sure what role he played, whether he was -- you know, a significant role there, whether he was -Q. I'm asking about a person of ordinary skill in I'm not asking about

have the totality -Q. A. patent. Q. yourself. I want you to do what you need to satisfy Tell me if there's a reason in there that you You have it right there. Take a look.

14 15 16 17 18 19 You're 20 21 22 23 24 25

I don't know if you want me to read the whole

the art, not about Kepley as a man. his role in the state of the art.

couldn't build the system that's described. MR. PAYNE: I'm going to object to form.

I'm saying if a person of skill in the art knew about Kepley. Right? And by "Kepley," I'm referring to

talking about the compression or the whole system? MR. STEPHENS: I'm talking about whether there's

the patent, not the man. A. Q. A. Yeah. If they had in their hands Kepley -Right.

any reason that he knows of that there was -- that you could not use a known compression algorithm to build the system that's described.

9/16/2007 3:14 PM

93

9/16/2007 3:14 PM

95

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

THE WITNESS:

If you say that I know of, if

1 2 3 4 5 6 7 8 9 10 11

Q.

-- so that that's part of the state of the

there's any reason I know of, at this point I don't know of a specific reason why you cannot find some algorithm that would be suitable for this application. BY MR. STEPHENS: Q. So, now, would you agree that if Kepley was

technology -- this is known technology in 1988 -- then your statement that it would have been entirely surprising and unexpected for anyone to conceive of or propose the idea of transmitting audio in a burst mode is wrong, because Kepley, the document, proposes that. Right? MR. PAYNE: THE WITNESS: Objection. Form.

considered to be part of the state of the technology at the time, that your statement in paragraph 38 about it being entirely surprising and unexpected for anyone to conceive of or propose the idea of transmitting audio in a burst mode is wrong? A. You know, I don't know Kepley. I never met him,

No, I don't agree with that. And I have

First of all, my statement refers to audio. been assuming for -- in the context of Kramer

12 13 14 15

specifically, that audio refers to wideband music for entertaining the consumer. BY MR. STEPHENS: Q. So you weren't suggesting that it would not have

and I don't know if he was part of this -- how -- to what extent he represents the state of the technology, you know. Q. Okay. So you don't know whether Kepley would be Right? And I mean

16 17 18 19 20 21 22 23 24 25

been -- that it would have been entirely surprising and unexpected for anyone to conceive of or propose the idea of transmitting voice/audio in a burst mode. A. My statement was not addressing voice So I'm not -- I didn't give an opinion Right?

part of the state of the technology.

by -- when I say "Kepley," I mean the patent. A. I assume the patent examiners found the patent So under that assumption, the patent

worthy of issuance.

specifically.

was presumed to have -- to have validity. Q. So if you assume that what's described in Kepley

about that in the statement. Q. Do you have an opinion about whether it would

is part of the state of the technology in the late '80s and early '90s, then your statement that it would be entirely surprising and unexpected for anyone to conceive

have been obvious to conceive of or propose the idea of transmitting voice/audio in a burst mode in the late '80s?

9/16/2007 3:14 PM

94

9/16/2007 3:14 PM

96

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2

9/1 Allen Gersho Filed 09/17/20071/2007 Page 13 of 18

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

MR. PAYNE:

That's beyond the scope of his

"Say goodbye to endless loading or unloading of tape reels or carts. With the

declaration as stated. MR. STEPHENS: to answer? MR. PAYNE: BY MR. STEPHENS: Q. Do you have an opinion about whether or not it What's the question again? Are you going to instruct him not

3 4 5 6 7 8 9 10 11 12

DSP-2002's disk-based storage system, your sound effects or music can now be located by high-speed random access." A. Q. Yes. That shows that the digitally encoded wideband

would have been surprising or unexpected for anyone to conceive of or propose the idea of transmitting voice audio in a burst mode in the late '80s? MR. PAYNE: to answer. MR. STEPHENS: If you could mark this, please. Yeah, I'm going to instruct him not

audio is stored on a high-speed random access disk. Right? MR. PAYNE: THE WITNESS: Objection. Form.

It shows that high-speed random I have to

access storage is being used for the audio.

13 14 15 16 17 18 19 20 21 22 23

put the pieces together to see if it's -- that the encoded version is so stored, in fact. BY MR. STEPHENS: Q. Well, it has to be digitally encoded in order to Right?

(Exhibit No. 253 was marked.) BY MR. STEPHENS: Q. Dr. Gersho, the court reporter has handed you This is a CompuSonics document. Have you

Exhibit 253.

be stored on a disk. A. Q.

seen this before? A. Q. A. Q. A. Yes, I believe I've seen it. What's the system described in it? Some kind of audio processing workstation. And it's an audio editing system. I -MR. PAYNE: THE WITNESS: Go ahead. I believe they do mention that it Right?

It has to be in digital form, yes. And then if you'll look at the end of the I

document, there's a sort of specification section. think that's the name of it.

And then near the bottom, Do you see that? Is that

there's a section "Encoding Format." A.

I see "Audio Storage and CSX Encoding."

24 25

the section you mean? Q. Yes, that's the section I'm talking about.

9/16/2007 3:14 PM

97

9/16/2007 3:14 PM

99

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

does editing.

As I recall -- I haven't seen this in a I don't know for

1 2 3 4 Right? A. Q.

And that shows a variety of sample rates.

while, but it -- I shouldn't speculate. sure, but I -BY MR. STEPHENS: Q. Okay.

Yes. And those sample rates are sufficient to store Right? There

And it stores digitally compressed Right?

5 6 7 8 9 10 11 12

wideband audio. A.

wideband audio in random access storage. MR. PAYNE: THE WITNESS: BY MR. STEPHENS: Q. A. Objection. Form.

Depending on the quality that you want.

are different versions of audio. Q. Well, 50 kilohertz is wide enough to store the

I have to refresh my memory.

entire frequency band that human hearing is capable of sensing. A. Q. A. Right? Probably, yes. It's not probable; it's certain. Well, there's some people that believe that --

Take a moment to look it over, if you need to. The system does store audio in digital form that

has been encoded. Q. A. Q. And that's wideband audio. I believe so. And it stores it in random access memory? MR. PAYNE: THE WITNESS: Objection. Form. Right?

13 14 15 16 17 18 19 20 21 22 23 24 25

that they're useful -- people with good hearing may believe they can hear beyond the usual range that we assume. Q. But the usual range that's assumed by

I believe they do mention random

access memory, but I don't know if that's the storage means. BY MR. STEPHENS: Q. Just, for example, if you look at the page with

professionals working in the field is 20 Hertz to 20 kilohertz. A. Q. Right?

Something like that. And 50-kilohertz sampling rate is plenty to Right?

production number ending 699, there's a section entitled "Instant-Access Storage." A. Q. Yes. And it says: Do you see that?

store that full range. A. Q. Yes. Okay.

So the system was capable of storing Right?

wideband audio by any normal definition.

9/16/2007 3:14 PM

98

9/16/2007 3:14 PM

100

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
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

9/1 Allen Gersho Filed 09/17/20071/2007 Page 14 of 18

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

And let's just try and do this as quickly as we can so I can avoid Dr. Gersho having to make his car. THE VIDEOGRAPHER: We are off the record at

not fresh in my mind, though. Q. invented? A.

It's been awhile.

What is your understanding of what Mr. Lang

approximately 11:57 a.m., tape 2. (Whereupon, a lunch recess was taken.)

Well, first of all, I guess every claim is a So I don't know if I could say in

separate invention.

general, but I could address specific -- like claim 1 of '995 is the one that comes to mind most. Q. A. Okay. Let's go with that one.

And his invention has four major components:

the input means -- let's see -- random access storage means, and compression means, and output means, if I recall. Q. Those are -Okay. And is it your testimony that Mr. Lang is

the first person ever to have combined those? A. I didn't comment on that in my depositions. I

don't recall testifying to that specifically. Q. Do you have a view? MR. PAYNE: declaration. This is beyond the scope of his

I guess, yeah, you can answer in the

context of your declarations. THE WITNESS: BY MR. STEPHENS: Q. A. And what's the basis for that belief? Well, I haven't seen -- I haven't seen these I believe so, yes.

four elements combined in -- in any of the prior art

9/16/2007 3:14 PM

129

9/16/2007 3:14 PM

131

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

AFTERNOON PROCEEDINGS THE VIDEOGRAPHER: We are back on the record.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

references that I've seen before. Q. Right? MR. PAYNE: Objection. Form. Of course you did see them combined in Kepley.

The time is approximately 12:45 p.m., tape 2. Counselor. BY MR. STEPHENS: Q. Dr. Gersho, we've looked at a number of

MR. STEPHENS: to answer? MR. PAYNE: declaration. BY MR. STEPHENS: Q.

Are you going to instruct him not

different documents today, and those are all documents you had reviewed and understood in preparation for executing your declarations. A. To varying degrees. Right? Not all of them have I

Yeah.

It's beyond the scope of the

Now, going back to your declaration, No. 251,

studied all parts. Q. How did you determine which parts to study when

Exhibit 251, paragraph 38, you say: "It would have been entirely surprising and unexpected for anyone to conceive of or propose the idea of transmitting audio in a burst mode given the state of technology and the consumer product market in the late 1980s and early 1990s." Do you see that? A. Q. Yes. What is the idea of transmitting audio in a

you were reviewing them? A. I guess I looked for the parts that was most And knowing that there

relevant to my area of expertise.

was another expert that was covering a lot of material related to the transmission communication, and I mainly oriented to compression. Q. Okay. So your area of expertise primarily is Is that right?

compression. A. Q. Right? A. Yeah.

burst mode? A. Well, the idea is focused on a specific quantity

You also read and understood Mr. Lang's patents.

of audio that is in digital form and sending it in a faster time than realtime. The faster than the playback

I did read all four patents at the very But they're

24 25

beginning, and I believe I understood them.

time, the time it takes to get it from one place to

9/16/2007 3:14 PM

130

9/16/2007 3:14 PM

132

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2 3 4

9/1 Allen Gersho Filed 09/17/20071/2007 Page 15 of 18

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

A. Q. A.

I bought CDs at some point, but I didn't -No professional involvement? -- get the -- in professional involvement in the

location. Q. And the way you do that is by transmitting the

electronic data representing the file from one disk to another. A. that. Q. And the computer industry, including the Right? I suppose so. I don't see anything wrong with

CD industry. Q. Okay. Do you know how, for example, audio

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

cassettes were manufactured? A. Q. No, I don't. So you don't know whether people used high-speed

personal computer industry and other parts of it, have spent huge sums of money over the last four or five decades to make that process happen faster. MR. PAYNE: Objection. Form. Right?

tape duplicators to make copies of audio tapes to distribute and sell? A. Q. A. I -- this is all analog. For audio tapes, that's right? I mean the phrase "high-speed tape duplicator"

You're way beyond

the scope of his declaration. MR. STEPHENS: to answer? MR. PAYNE: BY MR. STEPHENS: Q. If you have an audio file on a computer -Yeah. Are you going to direct him not

makes sense to me, but I don't have experience with it. I never -- I never looked at what goes into such devices. Q. But you would agree that when you're making

copies and you're not actually listening to the copy that you're making, if you're doing it to distribute music, it makes sense to do that quickly. A. Q. better. A. Q. I'm not sure. Right?

computer work -- a Sun workstation's hard drive and you copy that audio file to an external hard drive, would that copy happen faster than realtime? MR. PAYNE: THE WITNESS: Objection. Form.

I could see pros and cons.

Well, all other things being equal, faster is Right? Well, I don't think all other things are equal. But clearly if you're trying to make a million

I'm not an expert on the insides

of what goes on in computer processing or computer communication. But, first of all, the realtime issue Is

CDs, then you want to be able to manufacture them

would only come up if the file happened -- is audio.

9/16/2007 3:14 PM

153

9/16/2007 3:14 PM

155

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

quickly.

Right? MR. PAYNE: THE WITNESS: Objection. Yeah. Form.

1 2 3 4 5 6 7 Because if you're 8 9 10 11 12 13 14 15 16 And that's how you make a copy of a file Right? 17 18 19 I'm 20 21 22 23 24 25

that part of your question? BY MR. STEPHENS: Q. A. Let's say it was an audio file. Yeah. So if it's an audio file, I guess it

But, I mean, if a printing

press can print thousands of pages per minute, it's just stamping out material. I don't know if it has anything

depends -- I can't say it's not possible that it would be faster than realtime. Q. Okay. In fact, let's assume that it's a PCM

to do with sending data faster than realtime. BY MR. STEPHENS: Q. Well, of course it does.

encoded 44.1 kilohertz 16-bit stereo file -- let's say mono file, and you copy that from one hard drive to another on the Sun workstations that your laboratory used in the mid to late '80s, would that copy happen faster than realtime? MR. PAYNE: Don't answer that question. Garland, come on. This man has put out a statement Assumes

sending data faster than realtime from one disk to another, you're doing it to make a copy. MR. PAYNE: THE WITNESS: Objection. Form. Right?

In your hypothetical situation, if

you are sending data from one disk to another and if that is needed to make a copy, then digital transmission is being involved internally here. BY MR. STEPHENS: Q. Okay.

facts not in evidence. MR. STEPHENS:

about technology that he's admitted depends on the workstations that were in his office. figure out how they work. And I'm trying to

from one disk to another. A. Q.

Direct him not to answer and

Well, not with audio cassette. I'm not asking about audio cassettes now.

we'll go on, if that's what you're intending to do. MR. PAYNE: Okay. Are you directing him not to

asking about a file on a disk. A. Well, there are many ways of creating a file,

MR. STEPHENS: answer? MR. PAYNE: critical -MR. STEPHENS:

but if you're -- if you're transferring a file from one place to another, a digital file from one place to another, then it would get stored on the destination

Yeah, yeah, because you've been

You don't have to explain it.

9/16/2007 3:14 PM

154

9/16/2007 3:14 PM

156

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2 3 4 5 6 7 8 9 10

9/1 Allen Gersho Filed 09/17/20071/2007 Page 16 of 18

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

times. A. Q. A. Q. A.

And on the next pass, you take the next -Yeah. -- the second bit for that sub-band. Yeah. What's wrong with that? But as I noted, that it seems to either

Q.

So when we're talking about the -- taking only

one out of a hundred bits, you're talking about the hundred bits being on the line leading into the demultiplexer. A. Q. Yes. Okay. And that's what you were using for all of Right? Right?

repeatedly send the entire data from the memory to feed it out to the demultiplexer and decoders a hundred times repeatedly is -- doesn't make any sense and isn't feasible. Q. Right? A. I don't know, because I -- to talk about If you did it, though, the numbers work out.

the counting that appears in your declaration. A. In that illustrative example, yes.

I could

construct another example based on your hypothesis that would also lead to nonsensical results. Q. Right now I'm only asking about what's actually And that's based on the

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

in your declaration.

demultiplexer seeing the 100 bits, not the decoder. Right? A. Q. Well -Let me be just be a little clearer. "Seeing" is

reproduction of sound, I don't think it would reproduce intelligible sound. So I don't know if the -- if it

makes sense to assess -- the hundred times reproduction of what sound? Q. Of garbage?

the wrong word. The 100 bits you're talking about are on the line leading into the demultiplexer, 24, in Figure 1. Right? A. Q. Yes. And they're not 100 bits on the line leading Right?

So you're basing your rejection, even of

exploring whether the arithmetic works out, on your conclusion that it wouldn't result in an audible signal. Is that right? A. Q. That question is a little confusing. I'm just trying to understand why, when you do

from the demultiplexer to the decoder. A. Q. Yes.

your counting, r1, b1, g1, et cetera -A. Right.

And you have not worked out an example where the

9/16/2007 3:14 PM

205

9/16/2007 3:14 PM

207

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

Q.

-- in an attempt to show that the numbers don't

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

100 bits, of which only one is taken by the decoder, are present on the line leading from the demultiplexer to the decoder. A. Is that right? I think -- I have worked out that example. I

work out, you're actually counting from the multiplexer rejecting 99 out of a hundred and not the decoder, as the reference actually says. A. taking. Q. A. What the reference says is the decoder is It doesn't say reject -Taking one out of a hundred bits presented to -Yeah, taking one out of a hundred. And it is,

think it's just the taking. you interpret the statement.

It's just a matter of how The way I understand

Kramer, this is not the same thing. Q. I'm not asking about the interpretation now of I'm asking about how you

but not by choice.

Because the demultiplexer is only

the language in Kramer. actually worked it out.

letting it take one out of a hundred. Q. Right? A. The demultiplexer is what does the presenting. That's how the circuits show. Well, it's -- and it's only -- but it is doing In order to deinterleave, it

So the 100 bits, though, were not present on the line leading from the demultiplexer to, for example, the rightmost decoder, 26. A. Q. A. Q. Right?

the one out of a hundred.

The hundred bits is what is coming in, yes. Into the demultiplexer? Into the demultiplexer. And they're not present on the line leading from Right?

has to -- it makes sense the demultiplexer has to be deinterleaving. Q. Well, is it true, sir, that the counting that

you did in an attempt to show that the numbers don't work out as the signal is presented to the decoders is based on the demultiplexer doing the rejecting of 99 out of a hundred bits? A. That's right, isn't it?

the demultiplexer to the decoder. A. Q.

To one specific decoder, right. Okay. So if you were going to correct the

language in column 4 that we were looking at, about the demultiplexer starting line 45, what would you have it say? A. Well, there are a lot of ways to correct it.

Well, it's equivalent to the decoder taking one The demultiplexer is --

out of a hundred. Q. A.

I just want a clear answer. Yeah.

One way would be by taking "e.g. only one out of every

9/16/2007 3:14 PM

206

9/16/2007 3:14 PM

208

9/11/2007 Allen Gersho Case 3:06-cv-00019-MHP

Document 165-2
1 2 3 4 5 So

9/1 Allen Gersho Filed 09/17/20071/2007 Page 17 of 18

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

hundred bits of information presented to the decoder at a time" -- sorry -- "presented to the demultiplexer." it would be -Q. it -A. Yeah. "Presented to" -- instead of "presented So just to be clear, the way you would correct

Q.

Just have a counter that counts up to a hundred When the counter reaches a Wouldn't that work?

and then takes the next bit.

hundred, then it starts over again. A. Q. A. No. Why not?

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Because it's not counting up to a hundred.

to it," I would say "presented to the demultiplexer." Q. Now, looking at your declaration in Exhibit 251,

You'd have to wait for the entire -- the entire data to pass through, and then -- in other words -Q. You could interleave it on the multiple And it does get somewhat more complicated

in column -- excuse me -- in paragraph 19, you're talking about the black box argument? A. Q. A. Q. Yes. And you put that in both your declarations? Uh-huh. And you make the argument that if the data is

sub-bands. there.

But if you look at it as if it was only one band, Right?

it's actually quite simple. A. simpler.

If there's only one band, it would be much But you really have to hold data to the right

sent a hundred times faster than the time it takes to play it back, that you'd have to store it; otherwise, you wouldn't be able to play it back. And the way you put that is "as long as the data is transmitted only once from memory to the box." A. Yes. Right?

time in order to get the right -- the right bit for the right sub-band. the next pass. Q. You could do this -MR. PAYNE: THE WITNESS: Whoa, whoa. It's kind of tricky. I'd have to If you don't get one pass, you get it

Under that condition, it's impossible

to -- to have -- it would not make sense of faster-than-realtime input if the output is realtime. Q. Okay. But if you did transmit it a hundred

spend a little bit of time thinking about whether or not it is possible. Although what I am acknowledging is --

and that you are correct, that the argument I gave by itself does not apply and does not show the impossibility of faster-than-realtime input with a realtime output if

times and only took 1/100 of the data each time it was sent, then that wouldn't be an issue. Right?

9/16/2007 3:14 PM

209

9/16/2007 3:14 PM

211

9/11/2007 Allen Gersho

9/11/2007 Allen Gersho

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

A.

Well, I think it would be a practical

1 2 3 4 5 6

you had more than one pass. BY MR. STEPHENS: Q. Okay. And you could arrange the kind of pass,

impossibility to transmit it a hundred times. Q. I understand you have reasons that you think But if

that you couldn't transmit it