See related files:
http://www.eff.org/IP/Video (EFF Archive)
http://cryptome.org/cryptout.htm#DVD-DeCSS
(Cryptome Archive)
http://www.2600.com/dvd/docs (2600 Archive)
http://eon.law.harvard.edu/openlaw/dvd/ (Harvard DVD OpenLaw Project)
1
1
2 UNITED STATES DISTRICT COURT
3 SOUTHERN DISTRICT OF NEW YORK
4
UNIVERSAL STUDIOS, INC.; )
5 PARAMOUNT PICTURES CORPORATION;)
METRO-GOLDWYN-MAYER INC.; )
6 TRISTAR PICTURES, INC.; )
COLUMBIA PICTURES INDUSTRIES, )
7 INC.; TIME WARNER ENTERTAINMENT)
CO., L.P.; DISNEY ENTERPRISES, )
8 INC.; and TWENTIETH CENTURY )
FOX FILM CORPORATION, )
9 )
Plaintiffs, )
10 )
vs. ) No. 00 Civ. 277
11 ) (LAK)(RLE)
SHAWN C. REIMERDES, ERIC CORLEY)
12 a/k/a "EMMANUEL GOLDSTEIN"; )
ROMAN KAZANI; and 2600 )
13 ENTERPRISES, INC., )
)
14 Defendants. )
-------------------------------)
15
16
17
18 DEPOSITION OF PETER RAMADGE
19 New York, New York
20 Friday, July 14, 2000
21
22 Reported by:
23 SHAUNA STOLTZ-LAURIE
24 CSR NO. 810490
25 JOB NO. 111054
2
1
2
3
4
5 July 14, 2000
6 11:15 a.m.
7
8 Deposition of PETER RAMADGE, held at
9 the offices of Proskauer Rose LLP,
10 1585 Broadway, New York, New York, pursuant
11 to agreement, before Shauna Stoltz-Laurie, a
12 Notary Public of the State of New York.
13
14
15
16
17
18
19
20
21
22
23
24
25
3
1
2 A P P E A R A N C E S:
3
4 PROSKAUER ROSE LLP
5 Attorneys for Plaintiffs
6 1585 Broadway
7 New York, New York 10036-8299
8 BY: WILLIAM M. HART, ESQ.
9
10 FRANKFURT GARBUS KLEIN & SELZ, PC
11 Attorneys for Defendants
12 488 Madison Avenue
13 New York, New York 10022
14 BY: MARTIN GARBUS, ESQ.
15 DAVID Y. ATLAS, ESQ.
16
17
18
19
20
21
22
23
24
25
4
1
2
3
4
5 IT IS HEREBY STIPULATED AND AGREED, by
6 and between the attorneys for the respective
7 parties herein, that filing and sealing be
8 and the same are hereby waived.
9 IT IS FURTHER STIPULATED AND AGREED
10 that all objections, except as to the form
11 of the question, shall be reserved to the
12 time of the trial.
13 IT IS FURTHER STIPULATED AND AGREED
14 that the within deposition may be sworn to
15 and signed before any officer authorized to
16 administer an oath, with the same force and
17 effect as if signed and sworn to before the
18 Court.
19
20
21
22
23
24
25
5
1
2 ([Ramadge] Exhibit 1, Mr. Ramadge
3 declaration, marked for identification, as
4 of this date.)
5 P E T E R R A M A D G E , called as a witness,
6 having been duly sworn by a Notary Public,
7 was examined and testified as follows:
8 EXAMINATION BY
9 MR. HART:
10 Q. How do you want me to address you, as
11 professor, doctor, Pete?
12 A. Whatever is most convenient is fine.
13 Q. Thank you.
14 Thank you for coming today.
15 How did you get involved in this case?
16 A. I think I was -- initially received an
17 e-mail from Andrew Appel asking if I would be
18 interested in testifying, and with some information
19 directing me how to find out what the case was
20 about.
21 And I was contacted by Ed Hernstadt from
22 the defense's law firm, and we had a discussion
23 about my research and how I did my research and how
24 I used video, and he asked me if I'd be able to
25 testify.
6
1
2 Q. And can you put a time frame on it?
3 A. Gee, I don't know when that was. I
4 could look up the e-mail date. I don't have it in
5 my head, though.
6 And the contact with Ed Hernstadt was
7 maybe two weeks later.
8 Q. The e-mail, again, was that from
9 Dr. Appel?
10 A. Yes.
11 Q. And Dr. Appel is at Princeton?
12 A. Yes.
13 Q. And you're at Princeton?
14 A. Correct.
15 Q. And how long have you been at Princeton?
16 A. About 16 years.
17 Q. And has your job or area of practice, if
18 you will, changed in any way over the course of
19 16 years at Princeton?
20 A. My -- maybe you could be a little bit --
21 Q. I'll rephrase the question.
22 You teach at Princeton full time?
23 A. I do.
24 Q. And do you teach in any specialized
25 areas?
7
1 Ramadge
2 A. Electrical engineering is the general --
3 is my department.
4 Q. Right.
5 And within electrical engineering are
6 there any areas in which you specialize?
7 A. Yes.
8 Q. What are those?
9 A. I'm in a subdivision of the department
10 that goes under the name of Information Science and
11 Systems.
12 Q. And if you were to be forced to explain
13 that in 25 words or less in English, could you just
14 tell us what you do currently at Princeton?
15 A. Currently at Princeton my research is in
16 the area of signal processing particularly applied
17 to video analysis, digital libraries, video
18 compression.
19 Q. And when you say "signal processing,"
20 are you talking about means by which a video signal
21 is transmitted and/or converted and/or used for
22 viewing?
23 MR. ATLAS: Object as to form.
24 A. Those are very broad areas. My research
25 is only in specialized sub areas of those areas.
8
1 Ramadge
2 Q. Could you just help me understand what
3 areas you specialize in.
4 A. For the past five years I've been doing
5 research in the analysis of digital video.
6 Q. And by that what do you mean?
7 A. We start with digital video usually in
8 compressed form, such as MPEG form, and then we
9 would write algorithms to read the data from the
10 MPEG file, process that data, perhaps find
11 interesting pieces of that compressed video.
12 Q. And by interesting pieces, what are you
13 referring to?
14 A. That's determined by the application at
15 hand. So if someone might be interested, for
16 example, in finding scene changes, because scene
17 changes can help you browse through a video. Other
18 people might be interested in finding certain types
19 of camera motion, because that may give you some
20 information about what is happening in the video.
21 Q. OK, let me try and simplify it. And
22 please correct me if I'm wrong, all right?
23 Are you saying that you specialize in
24 developing electronic search techniques to search
25 and analyze video content by some defined feature
9
1 Ramadge
2 or aspect of that content?
3 MR. ATLAS: Object as to form.
4 If that's your understanding.
5 A. What you've just described in general
6 terms is perhaps more than what I do.
7 Q. Ah, OK.
8 A. I am at the -- this research area is
9 only at its very beginnings, and so at the moment
10 we're only working on very fundamental algorithms
11 to investigate what can you do in these contexts.
12 We haven't yet got to the level of being able to
13 perhaps interact with an actual user and be able to
14 take what the user wants and produce it
15 necessarily, because we don't know how to do all
16 those things just yet.
17 Q. When you say "interact with the user,"
18 do you mean to frame a search based on a request
19 like please scan for every instance in which a
20 given actor, for example, Humphrey Bogart riding a
21 motorcycle?
22 A. That's something we couldn't do at the
23 moment. Nor do we try and do that. I concentrate
24 on much more fundamental parts of -- which might go
25 into produce building blocks to solve such
10
1 Ramadge
2 problems.
3 Q. Is there a way you could generally
4 describe for us what it is that you are
5 concentrating on?
6 A. If I give you an example --
7 Q. Please.
8 A. -- one thing we've worked on is
9 estimating camera motion from compressed video, so
10 without uncompressing the video we read through the
11 compressed video file, use the information which
12 can be pulled out very quickly from the compressed
13 video, and use that to estimate the motion of the
14 camera during the video.
15 Q. And what is the value of discerning
16 camera motion as you've just described it?
17 A. Well, that's an elementary building
18 block which could then be used later on in a more
19 complex search where -- in situations where camera
20 motion may tell you something interesting is
21 happening.
22 Q. For example, ultimately leading toward a
23 request such as the example I posited a moment ago,
24 like searching for Humphrey Bogart riding on a
25 motorcycle?
11
1 Ramadge
2 A. That's not specifically referencing a
3 camera motion. It would have to be something that
4 makes some context where camera motion is
5 important. So let me give you an example.
6 Searching for a fast break in a basketball game
7 would be something where camera motion can be quite
8 important, because typically the camera tracks the
9 ball very closely.
10 Q. And just so I get this straight -- then
11 we'll leave this topic -- is that because there is
12 some content-related interest in discerning the
13 events taking place in the game, that is, you know,
14 a particular basket was shot within a game, and,
15 you know, based on the camera motion, that there
16 would have been some necessary camera motion
17 incident to filming that, and therefore you're
18 using the camera motion as a cue for the search?
19 MR. ATLAS: Object as to form.
20 If you follow that, you could answer
21 it.
22 A. At the moment we don't know the ultimate
23 application for which this may be put. That's why
24 I say we're developing basic building blocks.
25 After we got a set of basic building blocks in
12
1 Ramadge
2 place we try to investigate and see what those may
3 be used on, and that may be context dependent.
4 Q. Thank you. I'm sorry for the lack of
5 clarity in my questioning.
6 Apart from camera motion, are there any
7 other types of cues that you're working with now?
8 A. I've mentioned scene change detection.
9 I haven't actually been researching that, but other
10 people have been researching screen detection.
11 I've also been researching search, for
12 example, where you provide a small clip of video
13 and you say such through the archive to find pieces
14 similar to this clip.
15 Q. Very good.
16 You said a moment ago that these
17 searches are done using compressed video, typically
18 MPEG?
19 A. Ideally they would be in the compressed
20 domain to save time. You want the search to be as
21 fast as possible.
22 Q. I see.
23 Is that the only reason you're working
24 with compressed video?
25 A. No, it's not the only reason. The
13
1 Ramadge
2 compressed video, the encoder used to produce the
3 decompressed video has done a lot of work, and
4 sometimes you can use that material. It actually
5 has some information which is valuable to you. So
6 it's not just it's in the compressed domain. It
7 saves you time from compressing it, but you can
8 exploit the point that the software and hardware
9 that did the compression has extracted useful
10 information and has placed that into the compressed
11 file, and you can use that information.
12 Q. When you said you were working with MPEG
13 type video compression, are you referring to one
14 particular type of compression algorithm or
15 process, or are there multiple video compression
16 algorithms or processes that you're working with?
17 MR. ATLAS: Object as to form.
18 A. We've been working with a variety of
19 different compression techniques.
20 Q. Can you identify those for me?
21 A. We've worked with video in MPEG I
22 format.
23 Q. Right.
24 A. MPEG II format.
25 Often when we are in the process of
14
1 Ramadge
2 gathering video or transforming video we pass it
3 through other formats.
4 Q. I'm sorry, compression formats?
5 A. Other compression formats which might be
6 hardware specific. So if we're using a machine
7 from Silicon Graphics, we might use a compression
8 technology or representation format for the Silicon
9 Graphics machine.
10 Q. Gotya. OK.
11 Are any of these processes better than
12 any of the others, in your experience?
13 MR. ATLAS: Better as to what?
14 Object.
15 A. These different compression technologies
16 have been designed for different purposes.
17 Q. What are the different purposes?
18 A. Normally, because video contains a lot
19 of information, many, many bits, you have to
20 compress the video in order to transmit it or store
21 it, transmit it over a bandwidth limited channel or
22 to store it on some storage device which has
23 limited space, so there's many variables you can
24 trade off one against the other, so different
25 compression techniques and standards do different
15
1 Ramadge
2 trade-offs to achieve different goals.
3 Q. Is it fair to say that there's some
4 relationship between the degree of compression and
5 the degree of viewing quality, if you will?
6 A. There are other variables in addition to
7 viewing quality.
8 Q. And what are those?
9 A. Well, viewing quality is a very
10 subjective measure, and it contains many other
11 subcomponents, size of the image, depth of the
12 colors, things like that.
13 Q. Have you had any experience in using the
14 compression process referred to as DIVX?
15 A. Can you spell that, please.
16 Q. D-I-V-X.
17 A. Yes, I have.
18 Q. And what has been your experience in
19 using DIVX?
20 A. In connection with this case I have
21 looked on the web and used a search engine to look
22 for references to this compression technology, and
23 I have downloaded several files which claim to code
24 and decode in this compression technology.
25 Q. Have you used this compression
16
1 Ramadge
2 technology?
3 A. I'm in the process of attempting to get
4 those things downloaded to function and investigate
5 how well they work.
6 Q. When did you start doing this?
7 A. Over the previous two weeks.
8 Q. So in last two weeks you have not
9 actually used DIVX?
10 MR. ATLAS: Object. I don't think
11 that was his testimony, but --
12 A. I don't know how to answer that
13 question, because I don't think DIVX is a very
14 defined object. I haven't seen a precise
15 definition of what DIVX is. I've used several
16 codecs. I've downloaded from the web. One or
17 another of those could be the one referred to as
18 DIVX.
19 Q. Have you made any evaluation of the
20 differences between the codecs that you downloaded
21 under the name of the DIVX?
22 A. I'm in the process of doing such an
23 evaluation. I haven't reached conclusions yet.
24 Q. I see.
25 Are there any particular objectives to
17
1 Ramadge
2 your research, and if so, what are they?
3 A. My first objective is simply to
4 investigate what these codecs do, what trade-offs
5 they make in the compression, and to do some sort
6 of evaluation of the resultant quality of the image
7 or video.
8 Q. And how are you evaluating quality, by
9 what criteria?
10 A. At the moment I'm looking at various
11 criteria, the size of the image, the bit rate of
12 the video, frame rate of the video, size of the
13 compressed file, and the subjective search of the
14 resultant viewing quality.
15 Q. And how is the subjective measure being
16 evaluated, or how will it be evaluated?
17 MR. ATLAS: Object to the form.
18 A. I haven't got to that point yet.
19 THE WITNESS: Sorry.
20 MR. ATLAS: That's OK.
21 Q. With respect to the other video
22 compression processes that you've worked with, have
23 you ever evaluated those processes in the same way
24 that you're now describing in the process of what
25 you're doing with DIVX?
18
1 Ramadge
2 A. I'm sorry, could you repeat that?
3 MR. HART: I'll have her read it back.
4 (Record read.)
5 (Discussion off the record.)
6 Q. You mentioned certain evaluations that
7 you're now undertaking with respect to what's
8 called DIVX, right?
9 A. Correct.
10 Q. Have you ever performed similar
11 evaluations for any of the other video compression
12 technologies that you worked with?
13 A. I have made comparison for my own
14 benefit between compression technology, such as
15 MPEG I and -- MPEG I and MPEG II and AVI, a
16 Microsoft format.
17 Q. And when you say you've made comparison
18 of these video compression processes for your own
19 benefit, what criteria did you use in those
20 evaluations?
21 A. We needed to set a -- choose a format
22 for my research and my graduate students' research
23 which would enable us to work with video in
24 compressed form, tackle the essential problems that
25 such video poses, but without necessarily having to
19
1 Ramadge
2 deal with all of the bells and whistles that the
3 consumer might want to have. So we had to evaluate
4 -- we worked with MPEG I ,we work with MPEG II, we
5 work with AVI.
6 Q. Did you ultimately come to a conclusion
7 for which to use?
8 A. For the purposes of the research I'm
9 involved with, MPEG I involves all the essential
10 pieces for the first pass, through some of the
11 things we're working through. After MPEG I, we
12 might see -- if there is a change, we need to work
13 with MPEG II.
14 Q. Does it compress at a higher rate?
15 MR. ATLAS: MPEG II?
16 MR. HART: Yes.
17 Q. (Continuing) Than MPEG I.
18 A. MPEG II has a higher bit rate than
19 MPEG I.
20 Q. Right. And does it yield a more highly
21 compressed file?
22 A. That -- I can't answer that question,
23 because there are other variables. It has a
24 greater resolution. It has more pixels, has a
25 higher bit rate. It has essentially the same frame
20
1 Ramadge
2 rate, but it uses interlaced format. So they're
3 really intended for different applications, so --
4 Q. Right. And MPEG II is for progressive
5 scan?
6 A. It could do it, but the new feature from
7 MPEG II is it does interlaced scan.
8 Q. MPEG II does both interlaced and
9 progressive scan, whereas MPEG I was only
10 interlaced?
11 A. MPEG is only progressed.
12 Q. Excuse me. I apologize.
13 Are the compression ratios of those
14 processes variable by the user?
15 A. You have to define who the user is,
16 because there -- video content produces --
17 Q. Fair enough. By the person that's
18 applying the process could compress the video.
19 A. Yes, they can choose the amount of
20 compression they wish to apply.
21 Q. Have you studied MPEG IV at all?
22 A. I'm aware of MPEG IV. It's not my area
23 of research.
24 Q. Is there a reason why MPEG IV is not
25 within your area of research?
21
1 Ramadge
2 A. The objectives of MPEG IV are different
3 from the objectives of MPEG I and MPEG II. MPEG IV
4 is a load bit rate coder intended for a different
5 suite of applications.
6 Q. And in plain English, what are those
7 different applications?
8 A. Wireless, multi-media, video phones,
9 video e-mail.
10 Q. Have you had occasion to see the
11 resulting screen display of any DIVX compressed
12 video?
13 MR. ATLAS: Objection.
14 Just so the record is clarified, I
15 think the witness testified he wasn't clear
16 that there was one --
17 MR. HART: Fine. Any DIVX -- I
18 understood that, but my question takes that
19 into account, I think.
20 A. I have downloaded coders which seem to
21 be DIVX coders. I downloaded several codecs. I
22 have been progressing my way through those codecs.
23 I'm not sure at this point if I have looked the
24 ones explicitly called DIVX.
25 Q. And the ones you've looked at that you
22
1 Ramadge
2 downloaded that you believed were called DIVX --
3 Right so far?
4 A. Yes.
5 Q. -- have you seen the result on a screen
6 display of processing using those codecs?
7 A. At this point I cannot be sure, because
8 the codecs were not labeled as DIVX coder in the
9 software, they were labeled by something else.
10 Q. What were they labeled by?
11 A. I think one as labeled "MPEG IV Codecs"
12 or some permutation of that. More than one was
13 labeled "MPEG IV High Bit Codec."
14 Q. And with respect to the two that you
15 just mentioned, have you processed video using
16 either of them and seen the resulting screen
17 display from them?
18 MR. ATLAS: Objection as to form. Go
19 ahead.
20 A. I have processed video using several of
21 those codecs, the ones just mentioned, and I have
22 seen the video on the screen.
23 Q. And what kind of screen were you seeing
24 it on?
25 A. I have played it on my laptop screen and
23
1 Ramadge
2 I have played it on a 20-inch monitor. Computer
3 monitor.
4 Q. And how did the video look to your
5 professional eye?
6 A. It had very noticeable artifacts.
7 Q. And by artifacts you're referring to
8 what specifically?
9 A. When you compress a video it's a lossy
10 process. Information is loss. That lost
11 information appears as distortions in the actual
12 video when you view it. Those are normally called
13 artifacts.
14 Q. How do those manifest themselves to the
15 user?
16 A. They may manifest themselves in various
17 ways. The one way is as a blockiness of the image,
18 because many of these codecs are block based. It
19 may manifest itself as a miscoloring of these
20 blocks, because the blocks are sometimes pulled
21 from prior frames --
22 (Telephone interruption.)
23 Q. Anything else to add to your answer?
24 A. There are other things, but they occur
25 less often.
24
1 Ramadge
2 MR. HART: I'm sorry. I apologize for
3 the interruption. The phone is ringing.
4 (Discussion off the record.)
5 MR. ATLAS: Can we just have the
6 answer read back. I don't know if he was
7 finished.
8 MR. HART: I apologize for that.
9 (Record read.)
10 MR. ATLAS: Wait. Are you done, or
11 are there other ways?
12 A. (Continuing) I believe -- as the
13 telephone rang I believe I was going to say that
14 there are other artifacts which can be produced,
15 such as, for example, a lost block.
16 Q. Right.
17 A. But they occur less frequently.
18 Q. What source material did you use for
19 these compression experiments?
20 A. I used a DVD.
21 Q. Any reason why you used a DVD?
22 A. High quality digital video, source of
23 high quality digital video.
24 Q. As compared to what?
25 A. NYSC videotape or television broadcast
25
1 Ramadge
2 signal.
3 Q. Could you tell me what the compression
4 ratio was that you used in these evaluations?
5 A. I haven't used any fixed compression
6 ratio.
7 Q. OK. Can you give me a range of what
8 kind of compression ratios you were dealing with in
9 these tests?
10 A. One objective is to examine what
11 compression ratio would be required in order to
12 reduce a video file down to 650 megabytes.
13 Q. And to reduce a video file to 650
14 megabytes what kind of compression ratio are we
15 dealing with?
16 A. That depends upon the size of the
17 original video.
18 Q. In the case of the original video that
19 you were using, how big was the uncompressed file,
20 if you will?
21 A. There were many -- there were many
22 compressed files on the DVD. All -- the video on
23 the DVD is already in compressed form.
24 Q. I understand that, but pre your applying
25 in the compression that you applied --
26
1 Ramadge
2 A. There are various files totaling over
3 seven gigabits on the DVD.
4 Q. And after applying the compression
5 processes you described you wound up with a 650
6 megabyte sized file; is that right?
7 A. At the current time I'm still
8 investigating what compression ratios would be
9 required to reach a file size of 650 megabytes.
10 Q. Let me just ask you this, as a matter of
11 ratios.
12 What is the ratio in size, if you will,
13 between a seven gigabyte file and a 650 megabytes
14 file?
15 A. You simply do the math, divide 650
16 megabytes into seven gigabytes with the appropriate
17 conversion for units and you get the required
18 compression ratio to take all of the information on
19 the DVD down to 650 megabytes.
20 Q. Tell me approximately what that ratio is
21 as you sit here today?
22 A. If you give me a calculator, I will look
23 it up for you. It's simply a math problem.
24 Q. Have you attempted to compress such
25 video files to a size smaller than 650 megabytes?
27
1 Ramadge
2 A. Because these operations are very
3 lengthy, time consuming operations, I have been
4 targeting my research first to investigate the
5 different codecs that are available on smaller size
6 files, so to save time. Then, when I narrow down
7 the fields what I think might be good codecs for
8 further investigation, I can then investigate some
9 of the questions you are raise ing.
10 Q. So you haven't gotten there yet.
11 A. That's right.
12 Q. When do you think you'll get there?
13 A. Within the next two weeks I hope.
14 Q. And this is purely a function of the
15 research that you've been asked to undertake in
16 connection with this case, or is this a function of
17 your normal work research?
18 MR. ATLAS: Objection as to form.
19 (Mr. Garbus joined the proceedings.)
20 A. These questions that I'm investigating
21 here are relevant to my normal research, but I've
22 been motivated to limit them because of this court
23 case.
24 Q. And prior to your involvement in this
25 court case you had not begun to undertake any of
28
1 Ramadge
2 these experiments, correct?
3 A. Correct.
4 Q. And how long have you been experimenting
5 with video and compression generally?
6 MR. ATLAS: Objection.
7 A. Over five years.
8 Q. I see.
9 Now, is there any magic to the 650
10 megabyte number?
11 A. That's the size of a file that you can
12 burn into a writable CD.
13 Q. And do you know what the cost of the
14 equipment that's necessary to burn such a file into
15 a writable CD is to the consumer?
16 MR. ATLAS: Objection.
17 A. Various elements are required.
18 Q. And what are those?
19 A. You need a blank CD.
20 Q. OK.
21 A. Cost about one dollar.
22 Q. Right.
23 A. You need a CD writable drive. Costs
24 about two to $300. Sometimes cheaper.
25 You need to input that drive into a
29
1 Ramadge
2 computer system, PC based typically. Cost roughly
3 a thousand to $2,000.
4 Q. For the computer system.
5 A. Yes.
6 Q. Like just a desktop PC.
7 A. Correct.
8 Q. Anything else?
9 A. You need some software to actually
10 coordinate the copying of the file to the CD, but
11 that usually comes with standard packages that you
12 would get when you purchase the CD and the
13 computer.
14 Q. And have you burned or written to a CD
15 650 megabyte video files?
16 MR. ATLAS: Objection as to form.
17 Go ahead.
18 A. I haven't done that myself personally,
19 but I have asked my graduate students to burn such
20 files on several occasions.
21 Q. And have you had occasion to play those
22 files back from a CD so burned?
23 A. The files that I asked my graduate
24 students to copy onto the CD writable disks were
25 actually software, not video.
30
1 Ramadge
2 Q. Ah.
3 Have you ever caused audiovisual
4 information to be burned to a CD in a file size of
5 650 megabytes?
6 A. We regularly do backup of our home
7 directories, and our current backup mechanism is to
8 copy them to a writable CD, and inside our home
9 directories there will be video content.
10 Q. When you say your home directories, what
11 are you referring to specifically?
12 A. We have a networked computer system in
13 my lab. Each faculty member and each student has a
14 home directory where they store the files. Those
15 home directories are backed up on a regular basis
16 in case of a computer crash or a disk crash.
17 Q. And the video that's on those home
18 directories, what does that consist of?
19 A. That's video used for research purposes,
20 usually small segments of video that we are testing
21 our algorithms on.
22 Q. And when you say "small segments," how
23 big are those files typically?
24 A. File size could be as small as one
25 frame, it could be as large maybe as three to five
31
1 Ramadge
2 minutes of video.
3 Q. And when you say "we," who are you
4 referring to?
5 A. Myself and my graduate students.
6 Q. And where, to your knowledge, was this
7 video content obtained from?
8 A. Various sources. We have our own
9 camera. Some of it was taken with our own camera.
10 Some of it was downloaded from the web. Some of it
11 has been given to us for research purposes by the
12 copyright holder.
13 Q. Ah.
14 And in those instances where it's given
15 to you by the copyright holder for research
16 purposes, how did you accomplish that, did you
17 request permission or what?
18 MR. ATLAS: Objection.
19 A. We were involved with a research
20 contract with an industrial partner, and the
21 industrial partner negotiated with the copyright
22 holder for permission to use a certain segment of
23 video content.
24 Q. And was there a value to using a
25 particular segment?
32
1 Ramadge
2 MR. ATLAS: Objection.
3 A. Some of the algorithms that we are
4 developing, as I said before, can be employed in
5 various context, so sometimes it's interesting to
6 restrict the video to a certain context so we can
7 see how our tools could be employed in that
8 context.
9 Q. And in the particular instance you were
10 referring to in a moment ago, when the industrial
11 partner obtained the right to use certain video,
12 what was the particular attribute of that video
13 that made it valuable for your research?
14 MR. ATLAS: Objection.
15 Are we only talking about one, or is
16 this on multiple occasions?
17 THE WITNESS: It's on two occasions.
18 A. In general terms, it was a type of
19 sports video.
20 Q. Is that true on both occasions?
21 A. Yes.
22 Q. What kind of sport?
23 A. I think now I'm getting into an area
24 which maybe I can't talk about because I have
25 signed an agreement with the industrial partner not
33
1 Ramadge
2 to disclose.
3 Q. All right. I understand it. I'm not
4 asking you to do anything you'd be uncomfortable
5 with.
6 Is there a reason why a sports video
7 presented a particular good example for your
8 testing, and if so, what was that?
9 A. There are several reasons. One of them
10 is, as I pointed out before, that camera motion
11 might be quite important in sports videos, and we
12 have developed tools to help in the analysis of
13 video through camera motion.
14 Q. Gotya.
15 Now, you said -- and I'll try and
16 restrain my gotya's in deference to Mr. Garbus, who
17 joined us.
18 MR. GARBUS: Excuse me. I have to
19 make a phone call.
20 Q. And that's an inside joke between us.
21 You said you also downloaded video
22 content from the web.
23 Do you recall what you downloaded and
24 from where you downloaded it?
25 MR. ATLAS: Objection.
34
1 Ramadge
2 Go ahead.
3 A. I can't recall all of the instances, but
4 movie studios sometimes put trailers or short clips
5 from their latest releases on the web.
6 Q. Are those encrypted?
7 A. Those are not encrypted.
8 Q. Those are in digital form on the web,
9 yes?
10 A. Yes.
11 Q. Have you personally had any experience
12 in obtaining permission to use digital video
13 materials from copyright holders?
14 A. My only experience in that regard
15 directly is in asking other professors if I could
16 use a video which I know they were in possession of
17 and to be told that no, I couldn't, because I
18 wasn't part of the agreement for the use of that
19 video.
20 Q. And by "that" did you understand these
21 professors to mean that they themselves had
22 obtained agreements from the copyright holders with
23 respect to the videos they had in their possession?
24 A. That was my understanding. But they may
25 not have obtained agreements from the copyright
35
1 Ramadge
2 holder, typically from industrial partners who have
3 obtained the agreement with the copyright holders.
4 Q. Thank you for clarifying that.
5 Let's take a look at your declaration
6 for a few minutes we've marked as Exhibit 1.
7 I'd like to take you through paragraph
8 two and the list of subjects that you provide in
9 paragraph two, starting with the word "including
10 regarding"; do you see where I am?
11 A. Yes.
12 Q. And we have "Video Image Processing,"
13 right?
14 A. Correct.
15 Q. And what is that?
16 A. That's reading in the video content,
17 usually in digital form, and then analyzing or
18 changing that video content into a new form and
19 outputting that and then viewing that on the
20 screen.
21 Q. When you say "reading in," do you mean
22 copying?
23 A. No. I mean reading from a file. So
24 this would be stored on the computer in digital
25 format. We would open a file and read that file
36
1 Ramadge
2 into a piece of software that we have written.
3 That piece of software would perform various
4 analysis, possibly processing, changing the
5 content, and either write that out to a new file
6 that we later view, or immediately display on the
7 screen.
8 Q. And where does the video content that's
9 on the computer come from in the first place, is it
10 copied into the computer?
11 MR. ATLAS: Objection.
12 A. It wouldn't need to be loaded onto or
13 into the computer.
14 Q. And the processing that you're talking
15 about, I just need to clarify a couple things on
16 that. First you had talked about different kinds
17 of search algorithms, I believe, earlier in your
18 testimony.
19 You include that in the process you're
20 referring to now?
21 A. Yes.
22 Q. Is there any other kind of processing
23 that you're talking about apart from that?
24 A. Yes.
25 Q. And what is that?
37
1 Ramadge
2 A. We have developed algorithms which will
3 allow you to view the stored video in novel ways.
4 Q. Can you tell me a little bit about that,
5 please.
6 A. One way is to create what's called a
7 mosaic from the video, where you combine the
8 sequential frames into one image so that you can
9 have a general overview of what's in the video by
10 looking at one image.
11 Q. Hm.
12 And what's the purpose of that?
13 A. It's a tool for browsing through the
14 video.
15 Q. I see.
16 Any other kinds of image processing?
17 A. Yes. We've had students work on
18 different ways of interpolating video frames,
19 rotating algorithms or rotating and lodging video
20 frames, dropping video frames, and various other
21 things.
22 Q. And the value of those kinds of
23 processes is what?
24 A. If I want to create a mosaic of a video,
25 show you one still image of a large size, which
38
1 Ramadge
2 encapsulates or summarizes what's in the video, I
3 need to line up the video frame on the same plane,
4 that requires some sort of nonlinear
5 transformation, such as a rotation of an image
6 would be an elementary example of that, and we
7 would like to be able to do these things very
8 quickly, to do these elementary pieces of the
9 puzzle.
10 Q. And what's the value of that kind of
11 video manipulation?
12 A. If you had -- well, it has several
13 applications. One is to give you a very quick
14 summary, snapshot of what's happening in this
15 video. Another one might be a different way of
16 presenting the video content just for the consumer.
17 Another one might be as a compression tool, if
18 nothing in the video is changing very rapidly, then
19 creating a mosaic might be a good basis for a
20 compression technology.
21 Q. Are all video compression technologies
22 to some degree lossy?
23 A. No. There are two types of compression
24 technology. That's lossless compression and lossy
25 compression.
39
1 Ramadge
2 Q. Is one better than the other?
3 MR. ATLAS: Objection.
4 A. It depends upon the application.
5 Q. Are there trade-offs between lossless
6 and lossy types of compression, and if so, what are
7 they?
8 MR. ATLAS: Objection to form.
9 A. There are applications where it's deemed
10 more suitable to do lossless compression and there
11 are applications where it's deemed more suitable to
12 do lossy compression.
13 Q. And can you just generally describe to
14 me what that difference is, where lossless
15 compression is preferable and where lossy
16 compression is preferable?
17 MR. ATLAS: Objection.
18 You can answer.
19 A. One example where lossless compression
20 is preferred is in medical imaging. Medical images
21 are very large, so you still have to store them in
22 a compressed format to save space on your computer
23 system, but you cannot or are unwilling to tolerate
24 any loss of information, so therefore lossless
25 compression is required.
40
1 Ramadge
2 In consumer oriented video it's
3 permissible or deemed permissible to use a lossy
4 compression, because you can drop perceptual
5 insignificant pieces of the video but still keep
6 the consumer happy.
7 Q. Thank you. That was very clear.
8 With respect to the processing you
9 described a moment ago in your answer, rotating
10 images, dropping images, what have you, what were
11 the source materials for those processes?
12 MR. ATLAS: If you know.
13 A. Whatever video we have available on our
14 computer system.
15 Q. And again going back to the directories
16 that I think you mentioned --
17 A. Yes.
18 Q. -- among you and your graduate
19 students --
20 A. Yes.
21 Q. -- you talked about sourcing materials
22 from the web, sourcing materials through
23 arrangements with industrial partners, who had made
24 arrangements with copyright owners, and are there
25 any other --
41
1 Ramadge
2 MR. ATLAS: And I think filming
3 himself.
4 A. We have our own camera.
5 (Discussion off the record.)
6 Q. Have you ever had occasion to decrypt
7 any encrypted digital video in connection with your
8 work?
9 A. In connection with this case I have.
10 Q. Prior to this case did you ever have
11 occasion to decrypt any encrypted video in
12 connection with your work?
13 A. That cannot recall. Someone may have
14 sent me a mail message in encrypted format and I
15 may have decrypted it, but I don't recall.
16 Q. When you say "a mail message," do you
17 mean an e-mail with text?
18 A. E-mail message.
19 Q. Are you talking about PGP?
20 A. Yes.
21 Q. But with respect to video content, have
22 you ever had occasion to decrypt encrypted video
23 content in connection with any of your work prior
24 to this case?
25 A. I don't recall doing so.
42
1 Ramadge
2 Q. Now, have you had occasion since your
3 involvement in this case to decrypt any encrypted
4 video?
5 A. For the purposes of preparation for this
6 case, yes.
7 Q. And what did you do in that regard, sir?
8 A. I went to the web and started a web
9 search under the search DeCSS. That came up with
10 many web pages. I selected one of those and went
11 there and downloaded some software material which
12 claimed to do decryption of DVDs.
13 Q. Do you recall what web page you
14 downloaded DeCSS from?
15 A. There were many web pages that resulted
16 from the search. I went through several until I
17 found the program, and took it off the first one I
18 found.
19 Q. Do you recall which web page was the
20 first one that you found in your search?
21 A. No, I don't.
22 Q. Was it 2600?
23 A. I don't know.
24 Q. Was it there any reason why you did not
25 go to 2600's web page?
43
1 Ramadge
2 A. At that time I was not familiar with
3 2600's web page.
4 Q. And this was after you got involved in
5 this case?
6 A. Yes.
7 Q. But you weren't familiar --
8 A. I had heard the name 2600, but I knew
9 nothing else about it.
10 Q. You didn't know they had a web page.
11 A. I knew that they had a web page.
12 Q. Did you know that they were essentially
13 accused of providing DeCSS through their web page
14 in this case?
15 A. I did.
16 Q. But you did not see fit to visit the
17 2600 web page in connection with your researches in
18 this case?
19 MR. ATLAS: Asked and answered.
20 A. My normal mode is to go straight to a
21 search engine and type what I want into the search
22 engine.
23 Q. I see.
24 Did anyone tell you not to go to 2600 in
25 connection with your researches?
44
1 Ramadge
2 A. No.
3 Q. So you downloaded DeCSS from a web page,
4 correct?
5 A. Correct.
6 Q. And when was this, approximately?
7 A. About three weeks ago.
8 Q. And then what did you do with respect to
9 DeCSS?
10 A. I examined the code. I ran the program.
11 It seemed to function and do what it said it was
12 claiming to do.
13 Q. How do you know that?
14 A. I ran it.
15 Q. You say you ran it. What do you mean?
16 A. I executed it. I doubled clicked on it
17 on my icon and it opened up. It was able to read
18 the contents of a DVD.
19 Q. Do you recall which DVD you used in
20 connection with DeCSS?
21 A. "Contact."
22 Q. Jodie Foster feature length film?
23 A. Correct.
24 Q. Did DeCSS cause an unencrypted digital
25 copy of the content of the movie "Contact" to be
45
1 Ramadge
2 sent to your hard drive?
3 A. In order to save time I didn't decrypt
4 the entire movie. But it did appear to decrypt
5 whatever I asked it to decrypt.
6 Q. And these were in the form of DVD files?
7 A. Yes.
8 Q. Do you recall how many you caused to be
9 copied to your drive?
10 A. Three or four.
11 Q. And then what did you do with DeCSS for
12 those .vob files?
13 MR. ATLAS: Note my objection.
14 (Record read.)
15 A. I did nothing further with the program.
16 I attempted to play the vob files using my standard
17 vob player that came with my laptop. It refused to
18 play them. But subsequently I was able to
19 transcode those files into a different format and
20 play them.
21 Q. What kind of media player do you have
22 installed in your computer?
23 A. I believe it's called DVD Media Express.
24 Q. And when you say that you were -- you
25 then transcoded those files and played them, what
46
1 Ramadge
2 did you transcode them into and what kind of player
3 did you ultimately play them back on?
4 MR. ATLAS: Objection.
5 THE WITNESS: I'm sorry, could you
6 repeat the question?
7 (Record read.)
8 THE WITNESS: OK.
9 (Discussion off the record.)
10 MR. HART: Mr. Garbus is making hand
11 motions again to me.
12 MR. ATLAS: I think he's trying to
13 find out how long we're going with this
14 witness.
15 MR. HART: Right. We're doing great,
16 so keep moving along.
17 MR. GARBUS: You're doing great if we
18 do it before Sunday night at 6:00.
19 MR. HART: I'm sorry, we had a
20 question pending?
21 A. I transcoded them into a AVI format and
22 played them with the Microsoft media player.
23 Q. Did you try and play the nontranscoded
24 .vob files using your Microsoft video player?
25 A. I did not.
47
1 Ramadge
2 Yet.
3 Q. Thank you.
4 Have you done anything else with those
5 .vob files or with DeCSS other than what you've
6 already described here today?
7 MR. ATLAS: Objection.
8 A. I have opened them up in other media
9 players and transcoders and experimented with
10 different ways of transcoding and different --
11 transcoding them to different sizes, different
12 compression rates.
13 Q. So when you say "transcoding," you're
14 including within that term compression processing;
15 is that correct?
16 A. Yes.
17 Q. What compression processes did you apply
18 to the .vob files that you extracted from the movie
19 "Contact"?
20 A. I used a program called Flask MPEG.
21 That program allows you to do transcoding, and you
22 can set various parameters, frame size, frame rate.
23 Q. And what was the result of your
24 processing the "Contact" vob files with Flask MPEG?
25 A. I produced AVI files and I was able to
48
1 Ramadge
2 play those AVI files.
3 Q. How did they look?
4 A. Sorry?
5 Q. How did they look?
6 A. They contained visible artifacts.
7 Q. And that was playing back on what, your
8 computer monitor?
9 A. Correct.
10 Q. And do you know what the effective
11 compression ratio you were using was?
12 A. I think we went over this already, but
13 no, I can't remember. There were various choices
14 one can make, and I had started with small files
15 just to save time.
16 Q. I understand that we went over the
17 general issue of compression ratios, but what I'm
18 asking specifically, with respect to your
19 processing of the vob from the movie "Contact" with
20 using the Flask MPEG process, do you know what the
21 effective compression ratio was with respect to
22 those particular experiments?
23 A. Not off the top of my head.
24 Q. Was it greater than ten to one?
25 A. I can't recall.
49
1 Ramadge
2 Q. What else did you do with the vob files
3 from the movie "Contact"?
4 A. Because these files take up a large
5 amount of disk space, after my experiments are
6 completed each day I delete them so I have room for
7 other business.
8 Q. Now, isn't it also possible for you to
9 transfer those files to an auxiliary hard drive?
10 A. Because my research lab deals with video
11 and multi-media, we're constantly short of disk
12 space.
13 Q. I see.
14 Do you have any idea what the market
15 cost is of accessory hard drive space on a gigabyte
16 basis or based on products you --
17 A. The hardware cost is inexpensive, but
18 maintenance is quite high.
19 Q. By maintenance, what --
20 A. We have to get someone to set it up and
21 make sure it's working correctly.
22 Q. Have we covered all the different things
23 you did with DeCSS and the vob files from the movie
24 "Contact"?
25 A. I believe so. But I also want to say
50
1 Ramadge
2 this is ongoing effort.
3 Q. I see.
4 Have you used DeCSS to decrypt any other
5 DVDs apart from "Contact"?
6 A. No.
7 MR. GARBUS: Excuse me, "apart
8 from" --
9 MR. HART: "Contact."
10 MR. ATLAS: That was the movie.
11 Q. Did you experiment in the electronic
12 transmission of any of those files from one
13 computer to another or via the Internet?
14 MR. ATLAS: Objection.
15 A. No.
16 MR. HART: Why don't we take five
17 minutes.
18 (Recess taken.)
19 Q. You said earlier in your testimony,
20 professor, that you've had occasion to download a
21 number of different codecs from the Internet that
22 are referred to as DIVX, although you can't be sure
23 it's one particular type of codec or another at
24 this point; is that a fair statement?
25 A. There's very little information about
51
1 Ramadge
2 what DIVX actually is.
3 Q. Do you have on your computers the
4 downloads of those codecs and where you got them
5 from?
6 A. I have the downloads. I don't believe
7 the source of the download is stored.
8 Q. Can you recall as we sit here today the
9 names of any of the pages, in rough terms, of where
10 you downloaded so-called DIVX codecs from on the
11 net?
12 A. I started a search engine and searched
13 under "DIVX." There were many hits. Many of them
14 referred to something completely different. It was
15 a video rental scheme. So it required some
16 searching through these to find an actual reference
17 to this codec. At this point I can't recall what
18 that site was.
19 Q. Do you recall looking at a page or
20 downloading any codecs from a page called FM 4?
21 A. It doesn't ring a bell.
22 Q. Do you recall if any of the pages from
23 which you downloaded some form of DIVX codec
24 contained any references to DeCSS?
25 A. I don't recall whether the actual page
52
1 Ramadge
2 that I downloaded these codecs from -- and I think
3 there were several such pages where I downloaded
4 codecs. I don't recall whether they also had any
5 reference to the decryption software.
6 Q. In order to use the codecs that we've
7 been referring to as DIVX codecs with the caveat
8 you made earlier in your testimony about what
9 precisely they are, is it necessary to decrypt the
10 digital video content before you process it using
11 such a codecs?
12 A. With the codecs that I've used to date,
13 they appear to be simple transcoders. They assume
14 you start with a source which is a valid format,
15 and they produce an output. They do not assume
16 that there's any encryption of the source, nor do
17 they encrypt the output.
18 Q. "Nor do they" --
19 A. -- encrypt the output.
20 Q. Let me just clarify your answer and make
21 it simple for my simple mind.
22 Is it necessary to use these so-called
23 DIVX codecs to work with unencrypted content, or
24 will they process encrypted -- will they compress
25 encrypted content?
53
1 Ramadge
2 MR. ATLAS: Objection.
3 A. I don't know.
4 Q. Do you have any understanding of why
5 these video codecs are made available on the
6 Internet?
7 A. They're very useful tools.
8 Q. For making big video files smaller?
9 A. No, for transforming video content in
10 one format into an alternative format.
11 MR. ATLAS: You've got to give me a
12 second to object before you answer.
13 MR. GARBUS: Could I have the last
14 question and answer, please.
15 (Record read.)
16 Q. And what is the value of transforming
17 the video content from one format to another?
18 A. If we were doing an experiment, and we
19 have video in MPEG I format, we may also wish to
20 run the same experiment on the same video but in a
21 different format, in, for example, AVI format. So
22 such a tool would enable us to use the same video
23 in two different formats, running the same
24 experiment.
25 Q. Do you have any views as to whether the
54
1 Ramadge
2 AVI format is preferable for consumer use, say,
3 compared to any of the MPEG formats that we talked
4 about?
5 MR. ATLAS: Objection.
6 A. It's a very subjective decision, and I
7 also believe that Microsoft is constantly updating
8 their video formats.
9 Q. And when you say Microsoft is updating
10 their video formats, which video formats are you
11 referring to by name?
12 A. I don't know the exact names that
13 Microsoft gives them, but I know Microsoft is
14 constantly developing new tools and new formats for
15 distribution of digital media.
16 Q. Who developed AVI?
17 A. I believe it's a Microsoft product.
18 Q. And what about the MPEG formats?
19 A. That's an ISO standard.
20 Q. Going back to your declaration,
21 paragraph two, where you say this list, video and
22 image processing, are there any other areas, apart
23 from what you've testified to here today, where you
24 have been involved in video and image processing?
25 A. Can you I ask for clarification of the
55
1 Ramadge
2 question?
3 Do you mean from this list here?
4 Q. Right.
5 Looking at paragraph number two, you
6 give a list of topics starting with video and image
7 processing and so on.
8 A. I understand.
9 Q. And you've given us some testimony here
10 today about what you've done in connection with
11 video or image processing, your experience with
12 that.
13 I'm asking apart from what you've
14 testified to here today, have you done anything
15 else in connection with video and image processing
16 that we haven't talked about.
17 A. I've had various projects with both
18 undergraduates at Princeton, graduate students at
19 Princeton, summer students at Princeton related to
20 different issues in video processing and image
21 processing that we haven't talked about here today.
22 Q. Anything relevant to this case as far as
23 you're concerned?
24 A. Only to the extent that they're in the
25 general video processing area.
56
1 Ramadge
2 Q. Now, you say that when you downloaded
3 DeCSS you observed it and ran it.
4 A. Correct.
5 Q. What form was it in when you downloaded
6 it?
7 A. I downloaded source code and executable
8 code.
9 Q. From the same site?
10 A. Yes, they were both on the same site.
11 Q. Now, also in paragraph two, after video
12 and image processing you say hybrid slash switching
13 system slash adaptive control.
14 Can you tell us what that is?
15 A. That's an area of research concerned
16 with developing the algorithms that process
17 measured signals and make decisions based on those
18 processed signals in order to control the future
19 evolution of the signals.
20 Q. Can you describe some of the
21 applications to which you put your work in the area
22 of hybrid/switching systems/adaptive controls?
23 MR. GARBUS: Excuse me, can I hear the
24 last answer again?
25 (Record read.)
57
1 Ramadge
2 A. It's a very broad area, so it covers
3 many different applications, all the way from
4 temperature control in buildings to flight control
5 of aircraft.
6 Q. With respect to video systems, what is
7 the application of hybrid/switching
8 systems/adaptive control?
9 A. My research in this area has been on
10 fundamental pieces of larger problems, so those
11 individual pieces may be transferred to different
12 problems, perhaps in the video area.
13 Q. Have you done any work with
14 hybrid/switching systems/adaptive control in
15 connection with video?
16 A. We wrote one paper, myself and
17 colleagues, on the stability of mosaicing
18 techniques, which used some of my prior knowledge
19 obtained from work in this other area.
20 Q. The other area being hybrid switching?
21 A. Yes, correct.
22 Q. And what I am trying to get at -- I
23 don't want to belabor this -- is simply what
24 relevance hybrid/switching systems/adaptive control
25 has, based on your understanding, based on what you
58
1 Ramadge
2 bring to the table here in this case.
3 A. It's my general background.
4 Q. Oh, I see.
5 Stochastic optimization, can you tell us
6 what that is?
7 (Discussion off the record.)
8 Q. The question is can you tell us what
9 that is.
10 A. Sometimes when you're trying to optimize
11 something you get noisy measurements, corrupted
12 measurements of either the quantity you're trying
13 to optimize or maybe some gradient or differential
14 of that object. Stochastic optimization describes
15 the general class of algorithms taking such
16 corrupted or noisy observations and still doing the
17 minimization or maximization of an objective
18 function.
19 Q. Does stochastic optimization relate in
20 any way to video codecs?
21 A. I haven't seen it applied to video
22 codecs.
23 Q. Is that an area of possible application?
24 A. Many things are possible. If we knew
25 all the answers ahead of time, half the game would
59
1 Ramadge
2 be over.
3 Q. How true.
4 What relevance does stochastic
5 optimization have to video or imaging processing or
6 anything that you're prepared to testify on in
7 connection with this case?
8 MR. ATLAS: Objection.
9 A. It has general relevance to some of our
10 work in video processing, because when we, for
11 example, do camera motion estimation or mosaicing,
12 it involves estimating certain parameters. Those
13 estimates are obtained by minimizing a performance
14 measure, so we're doing an optimization of a
15 performance measure.
16 Q. But as you sit here today none of the
17 work that you've done in the area of stochastic
18 optimization has been applied to video processing
19 or video compression?
20 MR. ATLAS: Objection.
21 A. My research has applied certain ideas
22 from optimization, where the stochastic
23 optimization, for example, for creating video
24 mosaics, which we've already discussed.
25 Q. And other than that?
60
1 Ramadge
2 A. That's been my application.
3 MR. HART: Sorry, Marty.
4 Q. Let's turn to paragraph three of your
5 declaration.
6 MR. GARBUS: You've impressed the hell
7 out of me, Bill, with your knowledge. But
8 that was serious, that wasn't a joke.
9 MR. HART: Well, thank you. That's
10 better than the insult you gave me the last
11 time.
12 Q. Paragraph number three. You say you'll
13 be testifying in this case with respect to, and I
14 quote, "the necessity for and uses of digital video
15 information such as that found on DVDs in
16 connection with my present research on video
17 processing and video recognition software," and
18 then the sentence continues.
19 You see where I am?
20 A. Yes.
21 Q. But is it fair to stop there only in
22 that after the word "and" we're talking about
23 something else? Can we divide this into two
24 subjects, that is, necessity for and uses of
25 digital video information such as that found on
61
1 Ramadge
2 DVDs in connection with your present research on
3 video processing and video recognition software,
4 and the other subject is video compression
5 technology and the effect of compression on the
6 quality of video images?
7 MR. ATLAS: Are you asking whether one
8 relates to the other?
9 Q. (Continuing) Can we deal with those
10 fairly as two subjects?
11 A. There is some overlap, because one of my
12 current projects is involved in a new form of video
13 compression.
14 Q. OK. I'm sorry for the awkwardness of
15 the question.
16 A. I understand.
17 Q. I'm trying to make it simpler.
18 Let's focus on the first in paragraph
19 three, if you would. You say you'll be testifying
20 with respect to the necessity for and uses of
21 digital video information such as that found on
22 DVDs in connection with your present research.
23 What is the gist of your testimony in
24 that regard?
25 A. Briefly that DVD is a digital form of
62
1 Ramadge
2 video. We are developing algorithms for indexes,
3 searching, and analysis of digital video. To
4 adequately test our algorithms we do need a large
5 supply of digital video subject to fair use. It
6 would be a very good source for us if we could use
7 the video available on DVDs.
8 Q. And in order to use the video available
9 on DVDs are you suggesting you would need to
10 decrypt it?
11 A. Yes.
12 Q. Are you aware if there are any DVDs not
13 encrypted with CSS, that are not commercially
14 available?
15 A. I haven't seen any, but it wouldn't
16 surprise me if there are unencrypted DeCSS.
17 Q. Would those be of use to your research
18 without the necessity of decrypting them?
19 MR. ATLAS: If they exist.
20 MR. HART: If they exist, sure.
21 A. Generally, if we want to test our
22 algorithms, we want to have video that is
23 representative of the video out there in the
24 consumer marketplace. To ensure that, we prefer
25 not to have to select a sub set of available
63
1 Ramadge
2 videos, since that isn't necessarily representative
3 of what's out there.
4 Q. Do you know what's, quote, out there in
5 terms of commercially available DeCSS as we sit
6 here today?
7 A. Only through visits to the video store.
8 Q. And other than the decryption of the
9 movie "Contact" that you mentioned earlier in your
10 testimony, have you had occasion to decrypt any
11 other CSS encrypted DeCSS in your researches or
12 experiments?
13 A. No.
14 Q. Now, it's possible, is it not, to take
15 an analog source of video information and to
16 digitize it? Isn't that true?
17 A. That is correct.
18 Q. And for the purposes of your research,
19 is there any difference between sourcing the
20 material from an original digital source as opposed
21 to one that was digitized from an analog source?
22 A. The source in digital format is usually
23 preferable. It's less noisy. It has fewer
24 artifacts introduced from the analog recording
25 process and the subsequent digitization. And the
64
1 Ramadge
2 digitized analog signal would have certain
3 characteristics peculiar to the particular coder or
4 digitizer that we used, so, again, we would be
5 restricting us to a particular class of video,
6 mainly the class produced by this particular
7 encoder.
8 Q. Is it your testimony that by converting
9 the signal from digital to analog you reduce
10 artifacts?
11 A. The artifacts are presently in the
12 analog because it's recorded on analog tape. The
13 digitization process would introduce other
14 artifacts, yes.
15 Q. Now, in all events we're also talking
16 about content that has been compressed to one
17 degree or another, correct?
18 A. Not the analogs.
19 Q. Fair enough.
20 With respect to all the signals that are
21 digitized, or if they are originally a digital
22 domain, we're talking about signal that has in all
23 circumstances been processed, I'm sorry, compressed
24 to one degree or another; is that correct?
25 MR. ATLAS: Objection.
65
1 Ramadge
2 A. In the case of large video files, that's
3 true simply because of space restrictions. Small
4 video files are sometimes transmitted in an
5 uncompressed format.
6 Q. And when you say "transmitted," what are
7 you referring to?
8 A. They could be posted on the web. They
9 could be burnt onto a CD.
10 Q. Isn't the same true with compressed
11 video files?
12 A. Compressed video files are usually, if
13 they're large, delivered on a CD or a DVD format.
14 Q. And how do you know that?
15 A. From our own experience with our
16 research.
17 Q. Could you elaborate, please?
18 A. Even compressed video files are very
19 large, so moving them around is not necessarily an
20 easy task.
21 Q. Now, when you talk about your present
22 research on video processing and video recognition
23 software, have you caused the video material that
24 you are analyzing in these researches to be copied
25 to or loaded onto a computer in order to analyze
66
1 Ramadge
2 it?
3 MR. ATLAS: Objection.
4 A. Because these files are large and
5 processing them takes a long time, we like to have
6 them on the hard drive to minimize time to read
7 them.
8 Q. And how much video information is
9 presently stored on the hard drive you're using, in
10 gigabytes?
11 A. I have no idea. We have about
12 36 gigabytes of disk space, but I don't know what
13 proportion of those, code, video, e-mail, other
14 programs, I don't know the breakdown.
15 Q. Now, are these hard drive files
16 accessible to the public or only to persons working
17 with you in your lab?
18 A. Only to registered users within my
19 research group.
20 Q. And why is that?
21 A. It's a standard procedure for security
22 reasons to always have user accounts with passwords
23 and group permissions on certain files to restrict
24 access, so you can be sure who has access to what
25 type of program, software, data.
67
1 Ramadge
2 Q. And what's the reason for restricting
3 access to those programs and that data?
4 A. In some of the agreements that we've
5 made with our industrial partners we sign an
6 agreement that we would not distribute digital
7 media given to us or for our research purpose. We
8 make a best effort to fulfill that by restricting
9 access to that media.
10 Q. Now, again focusing on your statement in
11 paragraph three about your present research on
12 video processing and video recognition software, is
13 that essentially what we were talking about earlier
14 in your testimony today about developing algorithms
15 that can search video content?
16 A. Correct.
17 Q. Is there anything else, any outside
18 indicator to add to that in the context of what you
19 may or will be testifying to in this case?
20 A. My present research is that, what you've
21 just mentioned as one component. The -- sorry.
22 The video mosaic, the estimating of camera motion.
23 So there are various analysis tools for compressed
24 video in digital format, and those would be used
25 for searching database creation, index creation in
68
1 Ramadge
2 a library of digital video.
3 We have other research efforts under way
4 in analysis of video for different purposes, for
5 recognition and tracking purposes, detecting
6 objects, tracking them, and possibly recognizing
7 certain attributes of those moving objects.
8 Q. What's the potential application of
9 that?
10 A. There are many potential applications.
11 One could think of applications in surveillance.
12 Q. Does that involve use of decrypting
13 DeCSS that are partially available?
14 A. It could if we need a source of video to
15 test our algorithms. DeCSS could be a very
16 fruitful source of video.
17 Q. Now, isn't it true that you could make
18 digital video using the digital camera and
19 equipment that you have in your lab?
20 A. That is correct. But as I've also
21 stated, if you want to test your algorithms, you
22 would like to have independently created video to
23 test your algorithm. That's just a fundamental
24 aspect of the scientific method.
25 Q. And just clarify one more thing and then
69
1 Ramadge
2 we'll move on, and that is what is the value of
3 using commercially released movies in testing
4 processes that involve, say, for example, security,
5 watching moving objects and the like?
6 MR. ATLAS: Asked and answered.
7 You can answer it again.
8 A. I gave one example of where that might
9 apply, surveillance, but there are many examples.
10 Q. Including watching the basketball?
11 A. One application which we have a
12 demonstration is to segment the moving players in a
13 sports video, create a mosaic and reinsert the
14 players back on the video mosaic.
15 Q. OK, thanks for clarifying that.
16 You say in paragraph three you'll be
17 testifying in this case about video compression
18 technology and the affect of compression on the
19 quality of video images.
20 A. Correct.
21 Q. What does that refer to?
22 A. Because video is very large file and
23 needs to be compressed in order to be stored and
24 transmitted either on DVD, CD or electronically,
25 there are various trade-offs involved in
70
1 Ramadge
2 compressing video. You can gain a smaller file
3 size. That's an advantage if that's your
4 objective, but you have to give up something else
5 in return. My testimony will be concerned with
6 what trade-offs you need to obtain your goals.
7 Q. Can you tell me what your testimony is?
8 A. We discussed tests that I'm undertaking
9 at the moment, taking DVD vob files, transcoding
10 them to different formats, different ultimate file
11 sizes, look at what rates can be achieved, what
12 frame rates, what frame sizes in terms of number of
13 pixels, and subjective measure of quality of the
14 resultant video.
15 Q. And beyond what you've testified to
16 earlier today, is there anything else that you
17 haven't already testified, that you contemplate may
18 be included in your testimony at the trial?
19 A. As I have said previously, this
20 investigation is in progress.
21 Q. Right.
22 A. If we discover anything else subsequent
23 to today, it may appear in my testimony, but it's
24 not known to me at this point in time.
25 Q. Do you have any idea how you're going to
71
1 Ramadge
2 measure this objective quality of the resultant
3 videos derived from the various compression
4 experiments that you said you're still in the
5 process of undertaking?
6 MR. ATLAS: Objection.
7 A. It appears from my initial
8 investigations that the quality of the videos is
9 visually quite different, so therefore a mere
10 visual inspection may be adequate to demonstrate
11 the significant difference in the quality.
12 Q. And what are we comparing? What is
13 being used in the context of the comparison you
14 just described?
15 A. So the vob file from a DVD and its
16 existing format as it exists on the DVD and the
17 quality therein compared to a transcoded file
18 targeted at 650 megabytes size, roughly.
19 Q. And you're saying that if you, Peter
20 Ramadge, were to look at the vob file on the one
21 hand and a compressed version of that vob file that
22 had been compressed to a 650 megabytes file size,
23 that it can be obvious to you the quality
24 difference between the two.
25 A. That has been my impression with the
72
1 Ramadge
2 experiments I've done to date.
3 Q. And what experiments have led you to
4 that conclusion so far?
5 A. I have taken vob files from a DVD,
6 decrypted them, transcoded them down to a smallest
7 frame size, slower frame rate, and then I have
8 played them, stored them in AVI format, and played
9 them using the Microsoft media player.
10 Q. That's using AVI?
11 A. Yes.
12 Q. You haven't yet done it with DIVX or
13 so-called DIVX, correct?
14 A. I've downloaded a bunch of codecs. I am
15 working my way through them.
16 Q. Right. OK.
17 Are you aware of any writable DVD
18 formats by which the vob files can be transferred
19 onto a readable disk without the necessity of
20 substantial compression or any additional
21 compression?
22 MR. ATLAS: Objection.
23 Q. (Continuing) Beyond that which is in the
24 DVD movie to begin with.
25 THE WITNESS: May I just hear the
73
1 Ramadge
2 question, please?
3 (Record read.)
4 MR. ATLAS: Objection to the whole
5 question.
6 A. Perhaps you could rephrase the question,
7 because I'm a little confused.
8 Q. Delighted to.
9 Are you aware of any writable DVD
10 formats that are currently in the marketplace or
11 are forthcoming to the marketplace?
12 MR. ATLAS: Objection.
13 A. I believe I've seen some commercial
14 advertisements -- I don't know if they're
15 prereleased advertising or release advertising --
16 for something called a DVD writable drive, but I do
17 not know anything beyond that.
18 Q. You've never worked with one.
19 A. No.
20 Q. Is it your understanding, based on what
21 you've read or know, that you could transfer the
22 unencrypted vob files from -- taken from a DVD
23 movie onto a writable DVD format without the
24 necessity of further reducing the file size through
25 a compression process?
74
1 Ramadge
2 MR. ATLAS: Before you answer, could
3 you just read that back? I'm sorry.
4 (Record read.)
5 MR. ATLAS: Objection.
6 If you know.
7 A. I'm still a little confused with the
8 question.
9 Q. I'm sorry.
10 MR. GARBUS: Can I hear the question
11 again?
12 MR. HART: I'll be happy to rephrase
13 it.
14 MR. GARBUS: Go ahead.
15 Q. Do you have any understanding of the
16 file size of writable DVD media based on what
17 you've read or know?
18 A. I believe it's in the range of four
19 gigabytes also.
20 Q. Given that file size, is it possible to
21 transfer decrypted vob files to writable DVD
22 without the necessity of additional compression?
23 A. Assuming the DVD drive works correctly,
24 then you could transfer a file of up to
25 approximately four gigabytes onto the writable DVD
75
1 Ramadge
2 disk.
3 Q. Do you have any professional opinion
4 about the average quality of NYSC television that
5 most consumers watch in their homes?
6 A. No.
7 (Laughter.)
8 MR. ATLAS: Substantive quality, what
9 they're watching or --
10 MR. GARBUS: You mean the Robin Byrd
11 show?
12 Q. Since, obviously, taking a poll in this
13 room yields some rather strange interpretations of
14 my question, I think I'm going to rephrase that
15 question, doctor.
16 Do you regard NTSC standard play-back
17 that's commonly available in everyone's home today
18 in the United States to be of high quality as far
19 as you're professionally concerned?
20 MR. ATLAS: Objection to the form.
21 A. It's a very subjective issue. Some
22 people find it adequate, other people find it
23 inadequate.
24 Q. What do you watch and own?
25 (Discussion off the record.)
76
1 Ramadge
2 MR. HART: Let me rephrase it.
3 Q. Can you describe the video system that
4 you have at home?
5 A. I have a television. A VCR.
6 Q. Do you have a DVD player?
7 A. I don't have a DVD player.
8 MR. GARBUS: I'll ask him if he's ever
9 watched "Survivors."
10 MR. HART: Come on, Marty.
11 Q. Would you agree with the statement that
12 the average quality of television as we know it
13 today by NYSC standard is pretty below par compared
14 to what the technology can give?
15 MR. ATLAS: Objection. He's not here
16 as an expert as to the quality of
17 television.
18 MR. HART: I understand.
19 MR. ATLAS: If you have sort of a --
20 MR. HART: Well, he is here to talk
21 about quality and compression, so I think it
22 is fair for me to ask, based on play-back
23 devices that are commonly available to
24 people, what your impression of the average
25 play-back available in the home is today.
77
1 Ramadge
2 MR. ATLAS: It's subjective.
3 MR. HART: Let the record reflect that
4 Marty Garbus is leaving.
5 MR. GARBUS: And giving Mr. Hart a
6 note.
7 MR. HART: Oh, excellent.
8 MR. GARBUS: Which I insist on an
9 answer.
10 MR. HART: "Were you going to shave
11 for Monday?" No.
12 THE WITNESS: Can you read back the
13 question, please?
14 (Record read.)
15 A. I think associated with any technology
16 is also a cost issue, and we can do marvelous
17 things with technology but at a certain cost.
18 Q. Right.
19 A. And I'm not an expert in the economics
20 and the technology combined side of this, but
21 people who decide what they think the consumer will
22 pay for and will desire and will be happy with.
23 Q. OK. That's a very politic answer,
24 doctor. Thank you for that.
25 I guess my next question is is the
78
1 Ramadge
2 equipment in your lab under which you and your
3 staff evaluate the quality of the video display
4 representative of what's in the average home or is
5 it of better quality?
6 A. I would hope it's of better quality.
7 Q. Why is that?
8 A. I think we have fairly high resolution
9 monitors, computer monitors in my lab. I believe
10 that to be of higher quality than the average
11 television.
12 Q. Because of the difference between
13 progressive and interlaced scanning rate or --
14 A. Just the number of pixels available on
15 the screen.
16 Q. Now, are you going to be testifying at
17 trial in this case?
18 A. Yes.
19 Q. And when did you first find out that you
20 would be doing so?
21 A. Approximately two to three weeks ago.
22 Q. Can you tell me what opinions, if any,
23 you are prepared to render in connection with the
24 case.
25 MR. ATLAS: Objection. I think he's
79
1 Ramadge
2 set forth his opinions today during this
3 deposition and also in his declaration.
4 Q. Apart from what you've testified to so
5 far, are there any other opinions that you
6 contemplate rendering in connection with this case?
7 A. As I've said, my experiments are
8 ongoing. Anything I discover subsequent to this
9 deposition would -- I -- would be discussed.
10 Q. Are there any other areas that you plan
11 to do experiments in, that we haven't discussed
12 here today?
13 A. Not at the current time, but in the
14 process of doing experiments one comes to things
15 one didn't think of, but --
16 MR. ATLAS: Can we take a break for a
17 second?
18 (Witness and counsel left the room for
19 a discussion off the record.)
20 MR. HART: Are we ready to go back on?
21 MR. ATLAS: Sure.
22 Q. What did you just discuss with
23 Mr. Hernstadt?
24 MR. ATLAS: He didn't discuss anything
25 with Mr. Hernstadt.
80
1 Ramadge
2 Q. (Continuing) I'm sorry, Mr. Atlas.
3 A. Mr. Atlas advised me to say to you that
4 I would be willing to discuss any of my opinions on
5 these matters to make sure it had gone onto the
6 record.
7 MR. ATLAS: I just wanted to make
8 sure, since I did not have the entire
9 transcript in front of me, that the
10 professor covered both the areas of his
11 testimony and the opinions he was going to
12 cover. I wanted to make sure that he had
13 covered that as fully as he believed was
14 appropriate and, if not, to respond to that
15 question which I may have cut him off on
16 before .
17 Q. Professor, I'm a lawyer and not an
18 electrical engineer or video specialist, and I take
19 it you're not a lawyer.
20 A. Correct.
21 Q. But do you understand what was just
22 said? And if you do, could you please help me?
23 A. I think I understand, and I think I have
24 expressed my opinions on the areas on which I'll be
25 testifying, but we could go over them if you felt
81
1 Ramadge
2 necessary.
3 Q. You're saying this deposition has
4 covered as much as you know and as much as you're
5 prepared to testify to as we sit here today; is
6 that right?
7 Although there may be other areas that
8 your continuing researches reveal -- let me finish
9 the question -- but as you sit here today you have
10 no additional planned experiments, that you know of
11 in your mind, again, other than what you've already
12 described under oath here today.
13 Is that a fair statement to sum it up?
14 A. Within the context of this case, yes.
15 Q. Within the context of this case, yes.
16 Are you getting paid?
17 A. By Princeton University? For my work at
18 Princeton University?
19 Q. Are you getting paid in connection with
20 your testimony in the case?
21 A. No.
22 Q. Or any of your time?
23 A. No.
24 Q. No?
25 Again going back to your professional
82
1 Ramadge
2 views about the quality of video images, do you
3 regard VHS tape as presenting a fairly acceptable
4 image qualitywise?
5 MR. ATLAS: Objection.
6 But go ahead and answer.
7 A. These things are subjective.
8 Q. Right.
9 A. And standards are designed to meet
10 certain criteria, certain available bandwidth
11 criteria, for example, and there are trade-offs
12 involved in doing that.
13 Q. So you don't -- you have no view as to
14 the acceptableness of the quality of VHS.
15 MR. ATLAS: Objection. I'm not quite
16 sure what acceptable is, but --
17 A. I think it's subjective. I watched and
18 I continue to watch VHS movie tapes.
19 Q. At home?
20 A. At home.
21 Q. And as a consumer?
22 A. I've watched them and I am watching them
23 as an ongoing consumer.
24 MR. ATLAS: Are you asking him whether
25 he likes VHS or DVD better?
83
1 Ramadge
2 MR. HART: Hm, hm, hm.
3 Q. What did you review in connection with
4 your testimony here today in terms of things that
5 were filed in connection with this case?
6 MR. ATLAS: Objection.
7 Go ahead.
8 A. I reviewed Exhibit 1.
9 Q. Your declaration, right?
10 A. Yes.
11 Q. Anything else?
12 A. I had looked at my curriculum vitae.
13 THE WITNESS: Is that included here?
14 MR. ATLAS: I think it is.
15 THE WITNESS: Oh, yes.
16 Q. Actually, that's a good question.
17 Is your curriculum vitae attached?
18 A. Only partly I believe.
19 Q. Is there a reason why only part of it's
20 attached?
21 A. I believe what's attached is actually my
22 Princeton University electrical engineering web
23 pages.
24 MR. ATLAS: Correct. I'm sorry.
25 Q. Your web page doesn't have DeCSS on it,
84
1 Ramadge
2 does it?
3 A. Not to my knowledge.
4 Q. I'm sorry, and apart from your
5 declaration and your curriculum vitae, are there
6 any other things that you reviewed, that were filed
7 in this case, in connection with your testimony
8 here today?
9 A. I read the deposition of Andrew Appel.
10 Q. Anything else?
11 A. I believe I signed a nondisclosure
12 agreement.
13 Is that the correct term?
14 Q. Might be.
15 Go ahead. Anything else?
16 A. Let's see.
17 Just regular documents associated with
18 my research.
19 Q. OK, nothing more case specific.
20 A. No, not that I can recall.
21 Q. Have you reviewed the declaration of
22 Dr. Michael Shamos?
23 A. I'm sorry. Yes.
24 Q. Have you reviewed any of the
25 declarations of Robert Schumann?
85
1 Ramadge
2 A. No.
3 Q. With respect to Dr. Shamos' declaration,
4 when did you first review that?
5 A. Again, in the framework of two to three
6 weeks ago.
7 Q. Did you form any conclusions or opinions
8 about that declaration after reviewing it?
9 A. In preparation for the experiments that
10 are ongoing, that I'm doing, I noticed that there
11 was some pieces of information that I would have
12 liked to have seen included in that declaration, in
13 order to replicate those experiments, which were
14 missing from the declaration.
15 Q. Are you trying to replicate the
16 experiments that were described in his deposition?
17 A. That would be one interesting experiment
18 to perform.
19 Q. Are you planning on doing that?
20 A. Given that there was some lack of
21 information available in that declaration, I will
22 have to approximate that experiment.
23 Q. To the best of your recollection as you
24 sit here today what is the information that you
25 believe is missing?
86
1 Ramadge
2 A. I believe there were several things
3 missing. One of them might have the frame size in
4 pixels and the other one might have been the frame
5 rate in frames per second. There may have been
6 others, but those two things come to mind.
7 Q. Is frame rate in any way noted by the
8 codecs?
9 A. Some codecs allow you to choose the
10 frame rate.
11 Q. Anything else that was, in your view,
12 missing from Mr. Shamos' deposition?
13 By the way, it wasn't the deposition, it
14 was the declaration I was referring to. I
15 apologize for that.
16 Anything else that was missing, in your
17 view?
18 MR. ATLAS: Do you have a copy of it?
19 Maybe you could take a look at it.
20 Q. (Continuing) Just sitting here today, as
21 you can recall it.
22 A. I recall thinking it would be useful to
23 have known some other pieces of information, but
24 right off the top of my head, without having it
25 here in front of me, I don't recall what those
87
1 Ramadge
2 were.
3 Q. And sitting here thinking about the
4 experiments you'd like to duplicate that Dr. Shamos
5 did, can you say what, in your mind, is missing
6 that you prefer to know as a bit of information --
7 MR. ATLAS: Objection.
8 Q. -- for you to duplicate those
9 experiments?
10 A. I think I gave you the two things that I
11 can recall.
12 Q. Is there anything else that you reviewed
13 in preparation for your testimony here today?
14 A. I believe that's all.
15 Q. Now, is there a difference, in your
16 estimation, between copying video content to RAM as
17 opposed to the hard drive, and if so, what's the
18 difference?
19 MR. ATLAS: Objection to form.
20 A. Copying it from where?
21 Q. Well, let's use DeCSS, apply it to a
22 DVD, right? And I believe you've already testified
23 that that causes a copy to go to the hard drive
24 file.
25 Is that your understanding?
88
1 Ramadge
2 A. That's correct.
3 Q. And the hard drive is something
4 different than RAM copy in computer terminology,
5 correct?
6 A. It has different characteristics.
7 Q. And what's the difference in the --
8 A. They're much faster to copy it to RAM,
9 assuming you have an adequate supply of RAM.
10 MR. ATLAS: Just could we go off the
11 record for a second?
12 (Discussion off the record.)
13 Q. Now, once the decrypted movie content is
14 copied to the hard drive, what uses can be made of
15 that unencrypted copy from the hard drive?
16 MR. ATLAS: Objection.
17 Can you sit here and say all the uses
18 that can be made?
19 A. It's a decrypted copy of the vob file --
20 well, assuming that you had a player which would
21 play that decrypted copy, you could play it. If
22 you have a transcoder, you could transcode it to a
23 different format or a different compression.
24 Q. The hard drive's dumb, though, isn't it?
25 It doesn't --
89
1 Ramadge
2 MR. ATLAS: Objection.
3 Q. -- discriminate from what uses can be
4 made on it; isn't that correct?
5 It's just storing content, right?
6 A. Well, I think the situation can be a
7 little bit more complex. Because hard drives are
8 quite smart. They actually have built-in
9 electronics. They can have password protection.
10 They can have all sorts of things built into them.
11 Q. But absent a password protection as to
12 who has access to the hard drive, assuming there's
13 no limitation on who accesses the hard drive,
14 virtually anything can be done with the content
15 once it's on the hard drive.
16 MR. ATLAS: Objection.
17 That's true.
18 A. The situation is quite complex, because
19 you talk to the hard drive through the operating
20 system, and so depending upon how the operating
21 system has been set up, that could control various
22 activities that can happen on the hard drive.
23 Q. What kind of operating system are you
24 using in your lab?
25 A. We're using several different types of
90
1 Ramadge
2 computers with various operating systems. We have
3 SGI machines running an SGI version of UNIX. We
4 have Sun web stations running a Sun version of
5 UNIX. We have Intel based machines running
6 Windows NT. I have a laptop running Windows 98.
7 And we have PCs running Linux.
8 Q. And the computer through which you
9 downloaded DeCSS and decrypted "Contact," what
10 computer is that running on?
11 A. That was -- I did it on two separate
12 computers. I have two computers in my office. One
13 runs Windows NT the other runs Windows 98.
14 Q. What kind of an Internet connection do
15 you have from your home?
16 A. From my home?
17 Q. Yes.
18 A. Currently a 56 kilobyte dial-up
19 connection.
20 Q. And what about at the university?
21 A. The university, I have a connection to
22 the university's back bone through my department
23 connection, and hence to the university has a
24 connection to the Internet.
25 Q. Right. So ten megabytes a second, a
91
1 Ramadge
2 hundred?
3 A. Which part?
4 Q. Your connection from your office, let's
5 say, at the university.
6 MR. ATLAS: I think he's saying there
7 are various parts --
8 Q. OK, let me just rephrase it. I
9 apologize.
10 The immediate connection that your
11 computer has to the network that it's hooked up to,
12 what's the bandwidth of that?
13 A. I believe that's a category five cable.
14 Q. Which is what in terms of --
15 A. That's rated at 100 megabytes per
16 second.
17 MR. HART: I don't have anything else
18 for you, doctor. Thank you for your time
19 and your candor. And I personally find this
20 fascinating.
21 THE WITNESS: Thank you.
22 MR. ATLAS: Can we just take one
23 second?
24 MR. HART: Do you want five seconds?
25 Do you want me to go out of the room?
92
1 Ramadge
2 MR. ATLAS: You can go off the record.
3 (Discussion off the record.)
4 MR. ATLAS: I don't have any
5 questions.
6 (Time noted: 1:16 p.m.)
7 ____________________
8 PETER RAMADGE
9
10 Subscribed and sworn to before me
11 this ___ day of __________, 2000.
12
13 _________________________________
14
15
16
17
18
19
20
21
22
23
24
25
93
1
2 C E R T I F I C A T E
3 STATE OF NEW YORK )
4 : ss.
5 COUNTY OF NEW YORK )
6
7 I, SHAUNA STOLTZ-LAURIE, a Notary Public
8 within and for the State of New York, do
9 hereby certify:
10 That PETER RAMADGE, the witness whose
11 deposition is hereinbefore set forth, was
12 duly sworn by me and that such deposition is
13 a true record of the testimony given by the
14 witness.
15 I further certify that I am not
16 related to any of the parties to this action
17 by blood or marriage, and that I am in no
18 way interested in the outcome of this
19 matter.
20 IN WITNESS WHEREOF, I have hereunto
21 set my hand this 14th day of July, 2000.
22
23 _____________________
24 SHAUNA STOLTZ-LAURIE
25
94
1
2 ------------------- I N D E X -------------------
3 WITNESS EXAMINATION BY PAGE
4 PETER RAMADGE MR. HART 5
5
6 -------------------- EXHIBITS -------------------
7 [RAMADGE] FOR ID.
8 [Ramadge] Exhibit 1, Mr. Ramadge declaration. 5
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25