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)
816
1 UNITED STATES DISTRICT COURT
SOUTHERN DISTRICT OF NEW YORK
2 ------------------------------x
3 UNIVERSAL CITY STUDIOS, INC.,
]et al,
4
Plaintiffs,
5
v. 00 Civ. 277 (LAK)
6
SHAWN C. REIMERDES, et al,
7
Defendants.
8
------------------------------x
9
July 21, 2000
10 9:00 a.m.
11 Before:
12 HON. LEWIS A. KAPLAN,
13 District Judge
14 APPEARANCES
15 PROSKAUER, ROSE, L.L.P.
Attorneys for Plaintiffs
16 BY: LEON P. GOLD
CHARLES S. SIMS
17 SCOTT P. COOPER
18 FRANKFURT, GARBUS, KLEIN & SELZ
Attorneys for Defendants
19 BY: MARTIN GARBUS
ERNEST HERNSTADT
20 DAVID ATLAS
21
22
23
24
25
817
1 (Trial resumed)
2 THE COURT: A couple of preliminary things on my list
3 before we get started. First of all, have you been able to
4 reach agreement or otherwise come to a view on the specific
5 findings of the Microsoft case that I asked whether you would
6 stipulate to?
7 MR. SIMS: Your Honor we haven't yet. We will over
8 the weekend.
9 THE COURT: Just so I am clear, if I don't hear an
10 objection on the specific findings that I have identified, I
11 will take judicial notice of them and the deadline for any
12 objections is 11 o'clock on Monday. As I say, they are in the
13 nature of definitions and I --
14 MR. GARBUS: I can't see why it would be a problem?
15 MR. SIMS: I am advised we have gotten to it, we have
16 no objection.
17 THE COURT: You might consider also because I will be
18 looking further at that opinion for other things that might be
19 useful by way of background here, to see whether there is a
20 description in it of Linux that you can agree on as to what it
21 is, what its relationship to Microsoft windows is. You may be
22 able to do that, you may be able to save time. OK.
23 The second item on my list is that I am going to
24 alert you to an issue I thought I had because I am going to
25 ask for a brief on this if by the end of the trial I think it
818
1 is relevant and you might as well start thinking about it. If
2 I am understanding at least certain of the arguments here, the
3 defendants are arguing that there are encryption utilities
4 other than DeCSS that would decrypt CSS protected DVDs, that
5 there is no proof, direct or otherwise -- as to direct I think
6 it is stipulated -- that any particular decrypted movie that
7 may have been offered over the Internet or on a file-sharing
8 utility or whatever other relevant modes of distribution may
9 be, that was decrypted with DeCSS as opposed to some other
10 decryption utility and what the defendants I think would ask
11 me to conclude from that is that the plaintiffs therefore
12 can't have had any injury actually or threatened by reason of
13 these defendants' actions.
14 Just by way of a parenthetical digression, I am well
15 aware that the issue of the extent to which there is
16 sufficient proof that there are in fact decrypted movies
17 available over the Internet has not been decided and we are
18 going to have further discussion about that. My question
19 presupposes, just on a hypothetical basis, if there is
20 adequate proof to find that there are such movies out there.
21 The thought that occurred to me last night and on
22 which I may want briefs is whether the burden of proving that
23 whatever decryptions are out there came from the defendant or
24 more broadly DeCSS is on the plaintiff as distinguished from
25 the burden being on the defendant to prove that they did not.
819
1 The case that prompted me to think about that, I know we all
2 read this in law school a long time ago, Summers v. Tyce,
3 decided by the California Supreme Court in 1948 which seems at
4 least superficially to be analogous to this problem.
5 The case involved a group of hunters out hunting with
6 shotguns, two hunters fired simultaneously. One shotgun
7 pellet hit the plaintiff. Because they were shotgun pellets
8 as opposed to bullets out of a rifled weapon, it was
9 impossible to tell which shotgun the pellet came from. The
10 California Supreme Court said that the burden of proof was on
11 the defendants to show that their shotgun did not discharge
12 the bullet that hit the plaintiff and in the absence of such
13 proof, everybody who fired a shotgun at that time was jointly
14 and severally liable.
15 I am not inviting argument on that this morning, but
16 I wanted you to know that I am thinking about that and to give
17 you an opportunity to address it at the appropriate time.
18 OK. That said -- and I should add there is a lot of
19 development in the law of products liability, I am sure you
20 all the know the case Summers v. Tyce as its starting point
21 and some of you may know that. Let's continue where we are.
22 Mr. Corley, come on up. You are still under owes.
23 MR. GARBUS: We had that open question before you
24 concerning certain aspects of Mr. Corley's testimony. I
25 presume we are not getting into it until such time as a
820
1 further discussion of the court.
2 THE COURT: Absolutely. Absolutely. If it is
3 something you are going to want to go into Mr. Gold, tell me
4 you need to talk to me privately with Mr. Garbus.
5 MR. GOLD: Thank you, your Honor.
6 CROSS-EXAMINATION
7 BY MR. GOLD:
8 Q. Good morning, Mr. Goldstein.
9 A. Good morning.
10 Q. We left off yesterday discussing a statement that you had
11 made and I am going to give it to you again. Did you not at
12 one point in time, after October of '99, state that DeCSS was
13 a free DVD decoder that allows people to copy DVDs?
14 A. That was a statement that was made on our web site which I
15 did not write but I take responsibility for. It was at some
16 point after it was published on our web site I believe in
17 November that we realized that it was a lot more complicated
18 than just that and we -- future writings on the web site
19 reflected this.
20 Q. What do you mean when you say it was a lot more
21 complicated than that?
22 A. It wasn't simply a bit of code that allowed someone to
23 copy DVDs. At that point in time, I wasn't even that familiar
24 with the technology. I wasn't aware of how the encryption
25 worked, I wasn't aware of the extent of existing DVD piracy
821
1 and wasn't aware that simply copying a DVD was a trivial thing
2 that had been done for years, so I misread the actual facts of
3 the case. When I realized it was more complicated than what
4 we had reported, the case actually became very much more
5 interesting to me.
6 Q. Did you discover after you first made the statement we
7 have just referred to that it was incorrect?
8 A. Did I discover -- first of all, I didn't make the
9 statement. I discovered after looking at various public
10 forums, discussions about the subject, that this was not a
11 case about copying. It had nothing to do with copying. It
12 was about something called access control which was, prior to
13 November, I had been unfamiliar with that.
14 THE COURT: The question, Mr. Corley, wasn't about
15 what this case was about. The question was about the
16 statement that appeared on your web site.
17 THE WITNESS: I am sorry. Not the case but what was
18 going on at the time, the people whose web sites were being
19 shut down, the story itself, that apparently was not about
20 copying DVDs as we had initially stated.
21 Q. Let me ask you, Mr. Goldstein, is it not true that DeCSS
22 is a free DVD decoder that allows people to copy DVDs?
23 A. I believe it allows the data to be copied to a hard drive
24 at some point during its operation as do many other utilities
25 but I believe that it can do that if you write it in a
822
1 particular way.
2 Q. Then the prior statement you made was correct, I guess?
3 A. Not entirely because it did not exist for the purpose of
4 copying DVDs. One would not run that program to copy DVDs.
5 Q. But you said that it was a free DVD decoder that allows
6 people to copy DVDs and if I understand your testimony this
7 morning, you are saying the same thing?
8 A. It allows you to do that but that's not the purpose of it.
9 Again, I am looking at this from a journalistic perspective.
10 Q. I am not asking you the purpose, I am asking you what it
11 does.
12 MR. GARBUS: I will object to Mr. Gold cutting off
13 the witness' answer.
14 THE COURT: The witness is being unresponsive.
15 A. I am explaining it as best I can. Basically the program
16 exists to allow someone to bypass the CSS encryption on the
17 DVD. Now, whether that's the purpose of that is to copy a
18 DVD, that's something that we initially didn't understand
19 properly and I realize, I realized in November that it was --
20 that was not the case at all, that it was about defeating
21 access control, being able to have a player that worked under
22 the Linux operating system among other things.
23 Q. You began posting DeCSS on your web site in November of
24 1999?
25 A. That's correct.
823
1 Q. Is it your testimony that you did that as a journalist to
2 write a story?
3 A. That's correct.
4 Q. Could you have written the identical story without the
5 posting, using the letters DeCSS as many times as you wanted
6 in the story?
7 A. Not writing a story that would have been respected as a
8 journalistic piece, no, because in a journalistic world, you
9 have to pretty much put up or shut up. You have to show your
10 evidence and in this particular case, we would be writing an
11 evidence without showing what we were talking about and
12 particularly in the magazine that I work for, people want to
13 see specifically what it is that we are referring to, what bit
14 of technology that doesn't work, what new advancement, what
15 evidence do we have and simply saying that somebody else said
16 something just won't cut it.
17 So in this particular case, we pointed to the
18 evidence itself which was already firmly established out there
19 in the Internet world. We just put it up on our site so we
20 could write our perspective on it and show the world what it
21 was all about.
22 Q. Getting back to the question, in November of 1999, was it
23 possible for you to write a story about DeCSS on your web
24 site, using the letters DeCSS next to each other as many times
25 as you wanted, without posting DeCSS?
824
1 A. I will take another shot at it. I -- basically, the story
2 would not hold any value to our readers if we simply printed
3 allegations without showing evidence.
4 Q. I am not asking you, sir, what value your story would
5 have.
6 THE COURT: Mr. Garbus.
7 MR. GARBUS: I object. The questions are
8 argumentative. And bad as to form.
9 THE COURT: Certainly the one in the process of
10 winding up is.
11 Do you have another question, Mr. Gold? Ask it,
12 please.
13 Q. In order to post DeCSS, didn't you have to go to your
14 computer and get it?
15 A. Go to my computer and get it?
16 Q. Doesn't anyone have to -- yes, go to your computer, the
17 same one you used to write the story?
18 A. We had to get DeCSS from a computer, not our computer
19 because we didn't have it. We took it from one of the other
20 sites that had it at the time and posted it based on that.
21 Q. Once you did that, when you wrote DeCSS in your story, in
22 effect now you had posted it, isn't that right?
23 A. That's correct.
24 Q. Couldn't you have written the same story using the same
25 exact words and using DeCSS without going out and getting
825
1 DeCSS and posting it?
2 A. No. It would not have been the same story. It is
3 analogous to printing a story about a picture and not printing
4 the picture. People want to see what you are talking about.
5 Q. Does your web site now have a slogan "Stop the MPAA"?
6 A. It is probably on there somewhere, yes.
7 Q. Doesn't it stand out in bold letters?
8 A. I'm not exactly sure where it would stand out, but
9 wherever it is, I'm sure it stands out graphically.
10 Q. And what is the meaning of that phrase, what do you mean
11 to convey by stopping the MPAA?
12 A. Well, it is basically a phrase that I didn't coin myself.
13 It has come from a number of people that see the actions of
14 this lawsuit and other events over the past few months as
15 ominous and something that should be stood up to and that's
16 what people are doing simply by vocalizing that.
17 Q. Have you tried to stop the MPAA by writing stories about
18 the MPAA and describing what they do as inappropriate?
19 A. I don't know if they carry that much weight where I could
20 stop the MPAA by writing a story, but I have written stories
21 to educate people as to the facts of the case.
22 Q. Have you ever tried to stop the MPAA by making
23 contributions to organizations that were espousing the repeal
24 of the antisurf protection laws?
25 A. We have tried to raise funds, I am not in a position to
826
1 make contributions unfortunately, but we are trying to raise
2 funds to help pay for our defense.
3 Q. Have you promoted through your web site the taking or
4 downloading of DeCSS?
5 A. We have promoted the linking to various sites that still
6 have it posted because that has not been ruled illegal in any
7 way.
8 Q. Is it your belief as a reporter that to write a story
9 about trafficking in illegal drugs in Manhattan, you would
10 have to spend time actually trafficking in illegal drugs?
11 A. No, I don't think it is similar at all. One is a computer
12 file, one is an actual substance.
13 Q. Is it true that the number of sites posting CSS has
14 increased since January 2000 when you started linking?
15 A. It is hard to say exactly how many there are at any one
16 particular point in time. There were a lot already up when we
17 started the story. We came in kind of late. I believe CSS
18 was cracked, I am hearing dates of October, and we didn't post
19 the story until nearly a month later.
20 At that point in time, there were already hundreds of
21 sites that had it and that was a good part of the story that
22 we posted. I don't know how many there are now, and I don't
23 know how many came up in January. I do know there was an
24 increase after the preliminary injunction against us and I'm
25 not sure where that stands today or how many sites actually
827
1 have it or how many sites that haven't been reported that have
2 it.
3 Q. Mr. Goldstein, are you a leader in the hacker community?
4 A. Some people may say that. I don't like to refer to myself
5 as a leader but I am certainly somebody who certainly follows
6 that particular aspect of society and I try to speak
7 intelligently on it.
8 Q. You publish the Hacker Quarterly?
9 A. Yes.
10 Q. Your web site 2600 is known to virtually every hacker
11 throughout the United States?
12 A. And quite a number of nonhackers as well, yes.
13 Q. You have appeared on radio and television?
14 THE COURT: We are not going to repeat what Mr.
15 Garbus did yesterday, are we? We went through the TV programs
16 and magazine articles. Let's go, Mr. Gold. He has a right
17 not to like this law and he has a right to say so.
18 MR. GOLD: Absolutely.
19 THE COURT: The question is whether he violated it
20 and that's a different question.
21 MR. GOLD: My point, yes.
22 Q. There was a hacker convention last weekend, is that true?
23 A. That's correct, the third HOPE convention, call H2K.
24 Q. And there was a Mr. Johansen appeared at that?
25 A. Joe Johansen appeared on panel.
828
1 Q. And he spoke?
2 A. He and his father both spoke.
3 Q. Was a mock trial held of this case on --
4 A. Yes, actually we had scheduled that before the trial had
5 been scheduled so the timing was kind of strange but yeah,
6 that was organized by somebody else.
7 Q. Is it fair to say that the hacker community in the United
8 States is following this trial carefully?
9 A. I would say the hacker community, open source community,
10 the Linux community, yes.
11 THE COURT: What do you mean by the open source
12 community?
13 THE WITNESS: The open source community is basically
14 people who write software, release the source code, share
15 information, it overlaps into the Linux world. I am actually
16 not a part of that community. I didn't know much about that
17 community before this case.
18 THE COURT: OK, thank you.
19 Q. Is it true that the 2600.com site provides a form for
20 people to fill out so they can add their site to the 2600.com
21 mirror list?
22 A. Yes, we have two forms. We have a form where people know
23 about a site that carries DeCSS or know of a site that carries
24 DeCSS, they can submit that, or if they know of a site that we
25 are listing that no longer has it, they can fill that out as
829
1 well. But I should also point out that that has not been
2 worked upon for a couple of months simply because we are too
3 busy with other projects.
4 Q. But the form has been there for many months?
5 A. The form has been there from the beginning, yes.
6 Q. And it still is?
7 A. I believe so.
8 Q. You don't know?
9 A. I haven't looked at our web site in a while but I don't
10 believe any changes were made, no.
11 Q. How long a while?
12 A. I don't look at every page.
13 Q. How long a while haven't you looked at your web site?
14 A. I have looked at my web site. That particular page I have
15 not looked at probably in a couple of weeks but I believe it
16 is still up there because we don't make changes.
17 Q. Do you understand that it is wrong to break through a
18 protective device that protects a digital copyrighted work in
19 order to take that work?
20 MR. GARBUS: I object to the question.
21 THE COURT: Sustained.
22 Let me ask you something about the mirror list and
23 some of the sites to which you have linked. Are there sites
24 to which you are linked on which by clicking on the link on
25 your site, you reach a page at the linked site that has a
830
1 variety of content on it which may or may not include DeCSS on
2 the first page that comes up?
3 THE WITNESS: Yes, that's correct. I would say
4 that's probably the case more times than not.
5 THE COURT: Do some of those sites then give the
6 person who reaches it through clicking on the hyperlink on
7 your site then give the user or visitor, whatever you want to
8 call it, the option to go further in the -- I'll call it the
9 transferee site if the visitor wishes to download DeCSS?
10 THE WITNESS: The person is given the choice as to
11 whether or not they want to proceed further? Yes.
12 THE COURT: Are there sites to which you are linked
13 or have been linked where the effect of clicking on the
14 hyperlink on your site is to begin a download to the user of
15 DeCSS without the user or the viewer taking any further
16 action?
17 THE WITNESS: You would always have to take some bit
18 of further action even for something as simple as verifying
19 where you want to put the file on your hard drive. There are
20 no instances that I know of where you can only click once on
21 one of our hyperlinks and immediately start a download.
22 THE COURT: From your understanding, is it possible
23 to construct a link of the type I have described, whether it
24 is on your site or not?
25 THE WITNESS: I imagine anything is possible but I'm
831
1 not familiar with that particular application.
2 THE COURT: Are there links listed on your mirror
3 list in which the user, when the user clicks on the link is
4 transferred to a point on a transferee site that has no
5 content that comes up on the screen other than DeCSS or a
6 dialog box that requires the user to do something like verify
7 that the user wishes to download DeCSS? That is to say, no
8 other content?
9 THE WITNESS: Yes, that's certainly possible.
10 THE COURT: OK, thank you. Go ahead, Mr. Gold.
11 Q. Mr. Goldstein, have you ever publicly advocated that it is
12 wrong to infringe on anyone's copyright?
13 A. Yes, I support the copyright laws as I understand them.
14 Q. Have you, sir, publicly advocated that helping someone to
15 commit a copyright infringement is wrong?
16 A. I believe that helping somebody to break the law is wrong,
17 yes.
18 Q. What do you mean by wrong?
19 A. Illegal, immoral, something I wouldn't do.
20 Q. Now, at the end of 1999, did you know that CSS was a
21 protected device protecting digital copyrighted movies?
22 A. I knew it was an encryption standard that had been applied
23 to DVDs. I did not see it as the same thing as preventing
24 illegal copying, no.
25 Q. Why not? What's the difference?
832
1 A. Again, I'm not a lawyer, so my understanding of the
2 nuances might be a bit vague, but --
3 Q. Sir, I would like you to not give me any legal opinions.
4 THE COURT: Mr. Gold, he was trying to answer so
5 let's hear the answer.
6 MR. GOLD: Thank you, your Honor.
7 A. My understanding is the protection, the copyright law is
8 meant to protect the owner of the copyright from having
9 illegal works distributed whether by copying or any other kind
10 of infringement where the work or the copyright holder is not
11 compensated for the work. And my analysis of CSS and what CSS
12 accomplished was not the same thing as that. In other words,
13 copying of DVDs was not affected by whether or not one
14 decrypted CSS or one did not decrypt CSS.
15 Q. Did you know at the end of 1999 that the movies made by
16 the major Hollywood studios were copyrighted?
17 A. Did I know they were copyrighted? Yes, of course.
18 Q. Is the quarterly published, Hacker -- forgive me --
19 A. Hacker Quarterly.
20 Q. Is that copyrighted?
21 A. Yes, we copyright both the magazine and material on the
22 web page.
23 Q. Why?
24 A. We don't wish for people to be able to simply copy
25 everything on our site and claim ownership as theirs.
833
1 Q. Did you make the copyright application or authorize that
2 it be made?
3 A. Yes.
4 Q. Now, by providing links through the 2600.com site, are you
5 making it possible for any member of the general public to
6 download DeCSS if they have a computer?
7 MR. GARBUS: I object to the form of the question.
8 THE COURT: What's the objection?
9 MR. GARBUS: Let me hear the question again.
10 (Record read)
11 MR. GARBUS: I will withdraw the objection.
12 A. If they have a computer and a connection and they choose
13 to go to our site, yes, they can download it that way.
14 Q. Is it your understanding that once someone has downloaded
15 DeCSS through the 2600.com site, neither you nor 2600 has any
16 control over what they do with that, what they did with DeCSS?
17 A. That's correct, we have no control what people do after
18 they leave our site.
19 Q. Did you make any efforts to confine your providing DeCSS
20 to people who were going to use it for any specific purpose?
21 A. I don't think such control is possible on the net. So the
22 answer would be no.
23 Q. Is it true that anyone who downloads DeCSS can use it for
24 any purpose that they want to put it to?
25 A. Any purpose they want --
834
1 Q. Any purpose at all?
2 A. I'm not -- I am having trouble following that. DeCSS can
3 only do a limited number of things. It can't do anything
4 somebody wants.
5 Q. Is it not true that all DeCSS can do is decrypt CSS?
6 A. DeCSS exists for the purpose of bypassing CSS, correct.
7 Q. Is it true that your posting of DeCSS and your linking to
8 other sites that post DeCSS is being done for the sole reason
9 that you are a journalist?
10 A. The reason the story appeared was because it was a
11 journalistic piece. The reason we continued to write about it
12 and talk about it is because we believe it continues to be
13 that kind of a story.
14 Q. Writing it and being a journalist was your only purpose of
15 posting or linking to other sites that --
16 A. That's the purpose for our site, that's the purpose for
17 our magazine. It is a journalistic endeavor, yes.
18 Q. Did you ever write that the mirroring of DeCSS was a
19 demonstration of electronic civil disobedience?
20 A. I can't say that I wrote those specific words but I
21 believe those words may have appeared on our web site.
22 Q. And you're responsible for them?
23 A. I take responsibility for what appears on our web site,
24 yes.
25 Q. What is your definition of civil disobedience?
835
1 MR. GARBUS: I object.
2 THE COURT: Sustained.
3 Q. When those words appeared on your web site, what did you
4 understand them to mean?
5 A. My definition of civil disobedience, electronic civil
6 disobedience, I suppose simply people taking a stand in a way
7 they perceive as morally just. I really think it should be
8 left up to the individual person to make that definition on
9 their own. That's how I would picture it on the web site.
10 Again, I don't know the exact context of how it appeared on
11 our web site. I would have to look at that.
12 Q. Is it not true, to your knowledge, that civil disobedience
13 means violating some law for the purpose of making a
14 statement?
15 A. It many cases, it does involve sitting in front of a door,
16 for instance, a minor violation to prove a point, yes.
17 Q. It is limited only to minor violations?
18 A. It can be bigger than that, I suppose.
19 Q. Now, do you write a news story on your web site or does a
20 news story appear on your web site relating to DeCSS every
21 single day?
22 A. I write a story occasionally. There are other people who
23 write stories occasionally on the web site. They do not
24 appear every day, no.
25 Q. How often in the course of a week in the last three months
836
1 have stories about DeCSS appeared on your web site?
2 A. I would say probably an average of about one a week. And
3 that's simply because we don't have very much in the way of
4 staff.
5 Q. But I gather that you link to sites that post DeCSS every
6 single day, every single minute, every hour of the year?
7 A. It is not an ongoing thing. We are not consciously doing
8 it -- basically every story we write on our web page stays up
9 until the end of time or until we are ordered to take it down.
10 Q. Is it possible for you to remove it?
11 A. Yes.
12 Q. You haven't done that?
13 A. No, I have not.
14 Q. And you don't take it down on each day when you are not
15 writing a story about DeCSS, is that true?
16 A. No, we believe in keeping our stories up even when a story
17 has a factual error in it. Our philosophy is it would be
18 wrong to rewrite history and pretend we didn't say something
19 that was wrong.
20 Q. Have you made any money by posting DeCSS or linking it to
21 other sites that post DeCSS?
22 A. I would strongly suspect not.
23 Q. Why would you only suspect? Do you know?
24 A. I'm certain I haven't made any money from posting DeCSS.
25 I have probably lost quite a bit of money because I have not
837
1 been able to devote my time to the things I am supposed to be
2 doing.
3 Q. There has been a great increase in people coming to your
4 web site since you first started to post DeCSS, is that not
5 true?
6 A. I have no way of knowing that since we haven't kept a
7 counter since last summer.
8 Q. Wasn't that because so many people were coming to your web
9 site?
10 A. That was last summer before all this started. I have no
11 way of knowing how many people are hitting our site. We have
12 no ads on our site so we have no economic need for more people
13 to come there. I do note a lot of people are talking about
14 it. I can suspect that more people are coming to our site but
15 I have no real way of proving that.
16 Q. I gather al of your income comes from the Hacker
17 Quarterly?
18 A. That's correct.
19 THE COURT: Mr. Gold, isn't that whole line
20 irrelevant, in line of Greeging and a whole line of cases all
21 the way back to Greeging?
22 Q. Have you talked with anyone at all with about
23 advertisements in the Hacker Quarterly in the last several
24 months, the possibility of advertising?
25 A. No, our policy is never to accept advertising.
838
1 Q. And you have had a discussion with no one about that?
2 A. No.
3 Q. I gather, Mr. Goldstein, that you have never been involved
4 personally in reverse engineering?
5 A. No, I'm not an engineer.
6 Q. And I gather you have never been involved in cryptographic
7 research?
8 A. No, I am not a cryptologist.
9 MR. GOLD: Your Honor, I believe it would be
10 appropriate to have a sidebar at this point before I get into
11 another subject.
12 THE COURT: Come to the sidebar.
13
14 (Continued on next page)
15 (Pages 839-842 filed under seal)
16
17
18
19
20
21
22
23
24
25
843
1 (In open court)
2 MR. GARBUS: I would ask that this part of the record
3 be deemed confidential.
4 THE COURT: The transcript of the sidebar will be
5 separately bound and filed under seal. It will be available
6 to counsel but not otherwise.
7 Implicit in that of course is there will be no
8 discussion by counsel or revelation to anybody about what the
9 sidebar was all about.
10 THE COURT: Anything else, Mr. Gold?
11 MR. GOLD: Your Honor, I would like to offer pages 58
12 and 59 of Mr. Goldstein's deposition transcript as an exhibit
13 in the record at this point.
14 THE COURT: You have your objection, Mr. Garbus. I
15 take it subject to the objection. If I don't refer to it in
16 my decision, you should assume that I sustained the objection.
17 BY MR. GOLD:
18 Q. Mr. Corley, I have noticed that a picture of a telephone
19 and a telephone booth appears to an appear on the back cover
20 of every issue of the Hacker Quarterly?
21 A. That's correct. We print foreign pay phone pictures.
22 Q. For what purpose?
23 A. People seem to be interested in what pay phones in other
24 countries look like so they send us their photos when they go
25 on vacation or if they live in that particular country and we
844
1 compile them together and print four pictures on the back of
2 every issue.
3 Q. And the 2600, that title of your quarterly, is that --
4 does that relate to a situation occurring with frequency in
5 the '80s and '90s relating to phone service?
6 A. No, actually as I said yesterday, that dates back to the
7 1960s. It is something known as in-band signaling where a
8 particular frequency sent over a long distance telephone line
9 cause certain conditions to occur. And committing a 2600
10 hertz tone would drop a subscriber into what was known as
11 operator mode where they could route phone calls, route
12 themselves to internal operator and explore the entire system.
13 That has not worked for quite a while and it is a symbolic
14 type of name.
15 Q. Going back to the description you just gave us, was it
16 possible to use what you described to make phone calls without
17 paying for them?
18 A. That's one use, yes, it could have been used in that way
19 as well.
20 Q. Yesterday we were reviewing certain articles or your
21 counsel was reviewing with you articles that appeared in the
22 Hacker Quarterly. I am not going to go through them line by
23 line at this point. However, do you recall whether that
24 warning about -- that goes something like if I remember it, we
25 don't advocate the breaking of the law so we are not
845
1 advocating that you perform the acts described above, did that
2 appear in every column in the last four or five years?
3 A. It is something that has recurred. We don't print it on
4 every page in every issue. So for instance, if somebody
5 writes a letter to the editor and I think you could point to
6 virtually every expression of this, somebody expresses an
7 intent to commit a crime, we will without fail advise them not
8 to do that or try to convince them that's a really bad idea.
9 Q. Such descriptive articles about acts which were in fact
10 illegal did not contain that warning?
11 A. The articles are written from the author's perspective.
12 If the author wrote the article from a different perspective
13 as ours, there would be no such correction in there.
14 Q. By the way, how do we know that those authors weren't you?
15 A. Well, I can tell you they weren't me but other than that,
16 I don't know how I can prove it.
17 Q. So --
18 A. Without violating --
19 Q. So Bull Finch and Crypton and whoever people sign, are you
20 sure those people aren't you?
21 A. I'm very sure, very sure. For one thing, I don't have the
22 technical expertise to write many of those articles. I write
23 the editorial, I write the responses to the letters and
24 occasionally news updates.
25 Q. Do you recall one such article where there was a
846
1 description of how one could climb up a tree on his street
2 where there would be a black box that related to telephone
3 service in the community --
4 A. I believe you made reference to that yesterday. It is not
5 really a black box.
6 Q. What color is it?
7 A. It is different colors, but beige, white, silver.
8 Q. Those are the boxes?
9 A. Basically that particular story entailed people who could
10 climb poles or go into basements or closets in buildings and
11 make phone calls off subscriber lines.
12 Q. Do you remember in one paragraph, there was a description
13 of how you could climb the tree or pole and get to the black
14 box and whatever color it was and rip it apart so there would
15 be no telephone service to the community?
16 A. I don't recall specifically that paragraph but that could
17 have been something that the author wrote.
18 Q. And you did not write that article?
19 A. No, I did not.
20 Q. Does that article, if you remember, contain a warning
21 about don't break the law and don't do the things that are set
22 forth above?
23 A. No, as I said, we warned people through letters and our
24 editorial policy. We don't insert our comments into other
25 people's articles.
847
1 Q. I am asking if you remember that that particular article
2 did not contain such a warning?
3 A. I don't remember the particular article but our policy
4 states we don't insert our editorial policy into articles.
5 THE COURT: Mr. Gold, I think you made your point.
6 MR. GOLD: Thank you, Mr. Goldstein. Thank you I
7 have no further questions.
8 Oh, yes, if I may, there is one area that I haven't
9 gotten into. I thank my colleague.
10 Q. Mr. Goldstein, is it true that unless you are enjoined,
11 you intend to keep providing links on 2600.com to other sites
12 where DeCSS is available, do you not?
13 A. It is my understanding that that's legal so until we are
14 told otherwise, yes.
15 Q. The answer is yes?
16 A. Yes.
17 MR. GOLD: Thank you.
18 THE COURT: Thank you, Mr. Gold.
19 MR. GARBUS: I have one question.
20 REDIRECT EXAMINATION
21 BY MR. GARBUS:
22 Q. The judge asked you about linking. If www.garb.com comes
23 to you and asks you to link, do you know what material I will
24 have on my site when you give me permission to link to you?
25 A. Well, first of all, I believe you mean permission for us
848
1 to link to you.
2 Q. Yes?
3 A. What we had been doing while we were keeping the list
4 updated was basically going to the site, seeing if the file
5 names were there, were present, and if so, and this is because
6 we were getting so many submissions, we would just say yes,
7 OK, we will post this one.
8 What we later found out, there were some dummy files
9 out there. Some sites had the file names but they weren't
10 really the files. They were gibberish or even empty files.
11 It is simply impossible for us to go out there and verify that
12 each file is exactly what it says it is. That's one of the
13 reasons we stopped doing it, because it was kind of pointless
14 after a while.
15 Q. Turning to the judge's question, do you know if you are
16 going to link to www.garb.com who allegedly has DeCSS there,
17 whether it is on the first page, second page, fifth page or
18 what other information is contained on that site?
19 A. I am sorry, could you rephrase that a little?
20 Q. Let's, let's assume on your site 2600.com, you had
21 ww.garb.com as a linking site. Do you know -- and that
22 linking site allegedly has something about DeCSS, do you know
23 whether -- it has comments on DeCSS, do you know whether it
24 has the actual mirror and do you know where that appears on
25 the first page, the fifth page of the site?
849
1 MR. GOLD: Objection, your Honor.
2 THE COURT: Overruled.
3 A. We would know when we initially verified the existence of
4 the site if it was on the fifth page or first page or wherever
5 but we wouldn't retain that knowledge and that could change,
6 too. A person could take down DeCSS the day after we link to
7 them and until somebody told us that, we wouldn't know to take
8 it off the list. As I mentioned over the past few months, we
9 haven't maintained it at all.
10 Q. Do you know how many clicks it takes once you get to that
11 site to get to the actual mirror of DeCSS?
12 A. It differs with every site. Some may have a diatribe of
13 have a various political ideology that they want you to read
14 before you get to the particular mirror of the file where you
15 click on it. Some might have nothing but the file name when
16 you get to their site that you click on. So it varies with
17 every site. We have no way of controlling that.
18 THE COURT: Ever heard of the term "deep linking"?
19 THE WITNESS: I have heard the term. I am a little
20 confused by it because I know of one type of linking. I might
21 have an idea of what that is. I think it involves going into
22 a site beyond the main page, I think Ticket Master was doing
23 something like that or somebody was doing something like that
24 to Ticket Master.
25 THE COURT: Now, your web site has a home page,
850
1 right?
2 THE WITNESS: Correct.
3 THE COURT: That's the page you get if you enter the
4 universal resource locator www.2600.com into your browser,
5 true?
6 THE WITNESS: That is right.
7 THE COURT: You have other pages that are, so to
8 speak, behind the home page, right?
9 THE WITNESS: That's correct.
10 THE COURT: Every one of those other pages has its
11 own universal resource locator or URL, correct?
12 THE WITNESS: That's correct, yes.
13 THE COURT: So someone who wished to link to your
14 site and not to your home page but rather to some specific
15 page beyond your home page could enter in his own web page a
16 hyperlink not to your home page, but rather to the page behind
17 your home page in which the person was interested in linking,
18 right?
19 THE WITNESS: That's true, yes.
20 THE COURT: So if, for example, let's just make an
21 assumption, the New York Times, which is www.NYTimes.com, has
22 a web site, the home page has the URL that I just indicated,
23 right?
24 THE WITNESS: That's correct.
25 THE COURT: And let's just suppose that as somebody
851
1 suggested somewhere in the bowels of the New York Times web
2 site is the DeCSS code, all right? Are you with me?
3 THE WITNESS: Yes.
4 THE COURT: That code would appear on a page or out
5 there in hyperspace with its own unique URL, true?
6 THE WITNESS: Correct.
7 THE COURT: Now, if the New York Times wanted you to
8 carry a link to its mirror, you might link to that mirror in
9 more than one way, isn't that true?
10 THE WITNESS: There are several ways you can do it,
11 yes.
12 THE COURT: You could link by inserting the hyperlink
13 to www.NYTimes.com and then leave the user to navigate through
14 the New York Times web site, true?
15 THE WITNESS: Correct.
16 THE COURT: Or you could link by inserting the
17 appropriate URL to the specific page on the New York Times web
18 site that had the DeCSS code, am I correct?
19 THE WITNESS: That's correct.
20 THE COURT: And it was your practice to verify the
21 existence of proposed mirrors before you linked to them, when
22 you put them on your mirror list, right?
23 THE WITNESS: That's correct.
24 THE COURT: So the means of knowing whether the link
25 that you posted on your mirror list took someone who clicked
852
1 on that hyperlink to a page containing a whole bunch of
2 content, some of it relating to DeCSS and others relating to
3 God only knows what, as compared to taking you directly to the
4 source code for DeCSS was in your hands, true?
5 THE WITNESS: If we had spent more time on it, we
6 could have refined it so it took you directly to the actual
7 files, yes.
8 THE COURT: Or not?
9 THE WITNESS: Yes. What we did was we took what they
10 gave us and looked at that and said does this go to DeCSS,
11 either on the first page, second page or whatever and if it
12 did, we just put it in the way they gave it to us.
13 THE COURT: OK. Anybody want any further
14 examination?
15 MR. GARBUS: One thing.
16 If the court were to issue a declaratory judgment
17 saying that the posting of DeCSS were illegal or
18 inappropriate, would you stop posting the linking?
19 THE WITNESS: Yes, of course.
20 THE COURT: Mr. Gold?
21 RECROSS-EXAMINATION
22 BY MR. GOLD:
23 Q. One question, do you know what a declaratory judgment is?
24 A. I am sorry.
25 Q. Do you know what a declaratory judgment is?
853
1 A. It is similar to an injunction, preliminary injunction.
2 MR. GOLD: Thank you.
3 THE COURT: OK, anything else? Mr. Corley, you are
4 excused. Thank you.
5 (Witness excused)
6 MR. GARBUS: Your Honor, can I approach the bench for
7 a moment.
8 THE COURT: OK, you and Mr. Cooper or Mr. Sims,
9 whoever wants to come.
10 (At the sidebar)
11 MR. GARBUS: I think the Summer v. Tyce issue that
12 you raised is an issue that we have been conscious of since
13 the beginning and just to use the analogy, if one of the
14 bullets is defective --
15 THE COURT: I don't want to hear argument about it
16 now.
17 MR. GARBUS: I don't want to also, but I just want to
18 go into the question and relate it and I don't want to get
19 contumacious, the whole question of discovery and the question
20 of robustness is directly related to that issue and it has
21 always been our argument, which we have not had a chance to
22 document through discovery, there is the Merden report, the
23 Macrovision report which indicated that DeCSS was not a bullet
24 in the same way that the other utilities were bullets and it
25 has been, that's been our position from the beginning.
854
1 THE COURT: I am glad to hear that, Mr. Garbus,
2 because this is the very first time you have said it.
3 MR. GARBUS: That robustness was an issue and the
4 efficacy of DeCSS --
5 THE COURT: You just told me the Summer v. Tyce issue
6 has been in your mind from the beginning and the first person
7 to raise it in this case is me on the 5th day of trial. You
8 are trying to make a record for another purpose. If you want
9 to make a motion at an appropriate point for whatever relief
10 you think is appropriate, you make that motion, but we are
11 going to hear evidence right now.
12 (In open court)
13 THE COURT: Mr. Hernstadt, your next witness?
14 MR. HERNSTADT: Very briefly, before you go to the
15 next witness, yesterday we sent a letter in asking the Court's
16 permission to use a computer to do a demonstration. We
17 thought it was for next week but we have managed to bring the
18 witnesses in for today, so we have another letter asking
19 permission to do it this afternoon.
20 THE COURT: Tell me what it is about. Tell me what
21 the demonstration is supposed to be about, Mr. Hernstadt.
22 MR. HERNSTADT: This afternoon we would like to do a
23 demonstration of the LiVid Linux DVD player. The software is
24 still in an alpha -- in other words, even before it is
25 released for general testing, but it is functioning software
855
1 so we would like to show you how it works.
2 THE COURT: All right, I have signed the letter. I
3 didn't see the letter you sent yesterday.
4 MR. HERNSTADT: The other possible demonstration is
5 similar to the one Dr. Shamos did.
6 THE COURT: Mr. Cooper?
7 MR. COOPER: Your Honor, we have had no notice of
8 either of these demonstrations to my knowledge. It is
9 difficult for me to understand the relevancy of the first
10 described demonstration to any of the issues in this lawsuit
11 and we would like an opportunity to get more detail from
12 opposing counsel regarding what the demonstrations are.
13 THE COURT: How long is it going to take, Mr.
14 Hernstadt?
15 MR. HERNSTADT: The demonstration?
16 THE COURT: Yes.
17 MR. HERNSTADT: A few minutes.
18 THE COURT: Relax, Mr. Cooper.
19 Next witness?
20 MR. HERNSTADT: The defendants call Larry Peterson.
21
22 LARRY PETERSON,
23 called as a witness by the defendant,
24 having been duly sworn, testified as follows:
25 DIRECT EXAMINATION
856
1 BY MR. HERNSTADT:
2 Q. Good morning, Professor Peterson. Can you tell us about
3 your educational background please?
4 A. I have a Ph.D. in computer science from Purdue University.
5 Q. What year did you graduate?
6 A. 1985.
7 Q. What was the subject of your dissertation?
8 A. It had to do with several issues relating to computer
9 networks and focused primarily on e-mail and use of e-mail.
10 Q. After you received your degree, where were you employed?
11 A. I then took an assistant professorship position at
12 University of Arizona. I was there for 13 years, last two of
13 which I was department head.
14 Q. Of which department?
15 A. The computer science department.
16 Q. For the last couple of years, where have you been?
17 A. Princeton University.
18 Q. What courses do you teach at Princeton?
19 A. At Princeton, I have been teaching the computer networks
20 class and introductory programming class.
21 Q. What is involved in the computer networks class?
22 A. Primarily I walk students through the motivation behind
23 the Internet, the problems that cropped up over a number of
24 years as the Internet evolved, and how researchers had solved
25 them to evolve the Internet to where it is today.
857
1 Q. How long have you been studying the Internet?
2 A. Since I was a graduate student in 1980.
3 Q. How long has the Internet been around?
4 A. For a few more years than that. In 1980 is when it took
5 on the form that we are familiar with today roughly. Before
6 that, it was a packet switch network called the Arpanet.
7 Because of the advent of ethernet and wanting to bring more
8 networks into the fold, the Internet technology was deployed
9 in the very early '80s.
10 Q. And what are your research subjects?
11 A. My is research crosses between networks and operating
12 systems. Both of them are broadly the study of computer
13 systems. We build systems, we measure them, and evaluate how
14 they work, so it primarily has been focusing on how we can
15 build operating systems so that networks can be more effective
16 for users. We take a very, what we call, end-to-end
17 perspective so we are concerned not just with one particular
18 thing and how fast you can transmit over it or even the fact
19 that the links are catenated together to form the network or
20 that there is a computer attached to either end and
21 application programming or processes right on those computers,
22 so we have been focused on how to get data from an application
23 on this side of the network to an application on the other
24 side of the network.
25 Q. Have you ever researched a high speed connections from a
858
1 network to a computer?
2 A. Back during the early '90s, we were a part of what was
3 called the gigabit test, it was a national initiative to push
4 the edges of gigabit network technology and at that time, what
5 we in particular were doing was trying to move bits between a
6 processor on one computer to a processor on another computer
7 at the same speeds that the link connecting those two
8 computers operated which at the time was 622 megabits a
9 second, technology called OC12, and we were in fact able to
10 move the data from one machine to another at that speed.
11 There are a number of obstacles we had to overcome to do that.
12 Q. Is OC12, could you explain what OC12 is?
13 A. It's stands for Optical Carrier 12. There is a base unit
14 of transmission -- let me go back. There is basically the
15 technology having to do with fiber optics and there is a base
16 rate at which people can transmit which is roughly 55 megabits
17 per second and you keep multiplying that by 2 or 3 or 6 or 12,
18 so 6 times that base rate gives you 622. something megabits
19 per second. It is the way the bandwidth is measured on
20 optical links.
21 Q. Is OC12 used in the Internet today?
22 A. Yes, it is.
23 Q. Where do you find it?
24 A. You would find it in the backbone of the Internet. So let
25 me just give you a broad picture of the Internet and I can
859
1 tell you where different technology applies.
2 You can characterize the very core of the Internet
3 which is a set of long hold networks running across the
4 country from Los Angeles to New York and so on and cities in
5 between as the backbone of the Internet. A number of carriers
6 have provided backbones. In the original days of the
7 Internet, they were typically provided by government agencies
8 so there was the NSF net in the early '90s and eventually
9 commercial companies replaced that. So today you go to Sprint
10 or AT&T and they have a backbone. The links that make up the
11 backbone running between Chicago and Houston or whatnot might
12 be at 622 megabits per second.
13 Out near the edges of that backbone then, you connect
14 in various ISPs, Internet Service Providers, so your local
15 cable company might in one direction provide you a service at
16 home, but they have to turn around and connect into the
17 Internet at the high end, they would possibly use a 622
18 megabit link from their site into SprintNet, for example.
19 MR. HERNSTADT: Your Honor, may I approach?
20 THE COURT: Yes.
21 Q. I have handed you Exhibit BDC. By way of explanation,
22 could you tell us what this is?
23 A. This is just the layout of Sprint's Internet backbone
24 network. I found it on the web a few days ago. It shows
25 links connecting various cities. I know Sprint has a
860
1 significant presence in Kansas City, for example, so it will
2 show you the link running from Kansas City to Fort Worth and
3 collectively this would make up the Sprint backbone.
4 Q. Can you explain what the different links are?
5 A. This is color coded and it is hard to tell for sure which
6 are which, but these are the various links technologies being
7 used in the Sprint backbone, DS3, an older technology, 45
8 megabits per second. OC3 is 155; OC12, 622; 0C48, that's 6.4
9 gigabits per second.
10 MR. COOPER: From a foundational standpoint, can we
11 get some information from the witness about where this
12 document comes from?
13 THE COURT: He said he found it on the Internet a
14 couple of days ago.
15 MR. COOPER: We have not seen it before and I am
16 looking for some indication about how he went about locating
17 it and whether this is some area of his expertise or whether
18 he did some research and pulled it down.
19 THE COURT: Is that about the size of it?
20 THE WITNESS: Yeah, I went to Sprint.com as a good
21 guess of where to find Sprint's backbone, some links off
22 there, I couldn't give you the exact URL, and I found this
23 page.
24 THE COURT: What do you want me to do with it if
25 anything?
861
1 MR. COOPER: I am trying to explore the foundation of
2 documents of this type so we have an even playing field as far
3 as cross-examination is concerned, your Honor.
4 MR. HERNSTADT: Your Honor, we would move this into
5 evidence insofar as it assists your Honor in understanding
6 what this --
7 THE COURT: Now you are not worried that things that
8 appear on the Internet that are hearsay. But there is no
9 objection so I will receive it.
10 MR. HERNSTADT: Thank you.
11 (Defendant's Exhibit BDC received in evidence)
12 Q. Professor Peterson, have you received any honors or awards
13 in connection with your work?
14 A. Several of my papers have been Outstanding Papers at
15 different conferences, two in particular were Student Paper
16 awards where I was co-author with one of my students at
17 SigComm which is premiere conference for the networking
18 community.
19 Q. Are you an editor now of one of the --
20 A. I am editor and chief of ECM which is the computer
21 sciences professional organizations tracking on computer
22 systems. That's the premiere journal on computer systems.
23 Q. Have you authored any publications?
24 A. Several. I think I counted this morning over 50.
25 Technical papers.
862
1 THE COURT: Do you have a CV?
2 MR. HERNSTADT: I was about to ask if I could
3 approach.
4 THE COURT: Yes.
5 Q. Professor Peterson, is this your curriculum vitae?
6 A. Yes, it is.
7 Q. Does this accurately reflect your publications?
8 A. Yeah, I believe it does.
9 MR. HERNSTADT: Your Honor, we offer this into
10 evidence.
11 THE COURT: Received.
12 (Defendant's Exhibit BBD received in evidence)
13 Q. Professor Peterson, what is Internet 2?
14 A. Internet 2 is a consortium of 150-odd universities. In
15 the old days, the Internet was a play thing of the researchers
16 at universities, but of course it has become a very commercial
17 entity today and that has impacted the way research is
18 conducted. So primarily research universities have gotten
19 together to build an another piece, another corner of the
20 Internet that they can use for research purposes and done in
21 collaboration with some of the carriers, equipment companies
22 and the like, Cisco and Nortel, I am not precisely sure who
23 all the players are, so that they can have high speed
24 connectivity so their astrophysicists can send huge files to
25 each other in the national labs.
863
1 Q. Will it replace the Internet?
2 A. No, it is supplemental purely for the use of education.
3 In fact, to join the consortium, you have to agree you will
4 only use the connectivity and bandwidth for research and
5 teaching purposes.
6 Q. Are you being paid for your appearance today?
7 A. No, I am not.
8 Q. Or for any of the work you have done in connection with
9 this matter?
10 A. No.
11 Q. Can you describe how the Internet works by, for example,
12 if you were to send a file from your office to a colleague's
13 office, what are the various steps it would take?
14 A. As I have already briefly described that there is a
15 backbone and off the edge -- there are multiple backbones.
16 Off the edge of those backbones will be individual service
17 providers. They may be cable companies, local phone
18 companies. There are a variety of people who have gotten in
19 the business of providing Internet service. They will then
20 connect at some number of links into the various backbones
21 They might connect for one or might connect to more.
22 Collectively, you can think of that as the majority of the
23 Internet in terms of its capacity.
24 What happens at the edges then is that individual
25 users or consumers will buy some link into one of those
864
1 service providers. So maybe it is just they have a 56-kilobit
2 modem and they dial into the service provider periodically or
3 maybe they have released a line which is sometimes referred to
4 as the last mile, last mile link into the Internet.
5 So now we have the backbone surrounded by service
6 providers surrounded by the last mile links. That constitutes
7 then all the bandwidth that makes up the Internet. One point
8 I should make, if you go into the Internet, it is not just
9 links. There are nodes, computers, in essence, that connect
10 those links together sometimes called packet switchers,
11 sometimes called routers. So at every one of the connection
12 points, there is a router of some sort. Of course at the
13 edges, individual PCs and laptops and whatnot that the end
14 user has is connected to that last mile link.
15 Q. OK, could you tell us what file transfer speed is?
16 A. File transfer speed would be the number of bytes in the
17 file divided by how long it took to transfer the file. That
18 would give you a bit rate or byte rate.
19 Q. Are connections to the Internet or pieces of the Internet
20 rated by how fast you are capable of --
21 A. Any one of these links, you can talk about the bandwidth
22 it could potentially deliver. Whether or not you get the same
23 transfer speed as you have individual links depends on a
24 number of factors. It certainly depends on the fastest link
25 between the sender and receiver. So I couldn't go -- the
865
1 transfer time couldn't be any faster than the slowest link
2 between the source and the destination. There are other
3 factors that come into play though.
4 Q. In practical terms, does one ever get the maximum
5 theoretical speed of any piece of the Internet?
6 MR. COOPER: Your Honor, it is ambiguous and lacks
7 foundation.
8 THE COURT: Overruled.
9 A. We get it in the line. We can get conditions set up just
10 right that we can transfer at the theoretical maximum rate
11 that the link would provide. On any given day at any given
12 time, you might see something approaching that speed, might
13 see half of it, a tenth of it, it is hard to predict. It
14 depends on a number of factors, not the least of which is how
15 many other people are attempting to use the network at the
16 same time and so -- I could expand upon that, but there is
17 clearly going to be a point at which, as I have described it,
18 we have got millions and millions of end users connecting to
19 their ISPs, connecting to backbone. There is some point at
20 which there is a link that goes from Kansas City to Fort Worth
21 that is trying to carry all of that traffic and that link
22 could potentially be a bottleneck, a pinch point in the
23 network.
24 Q. Is that, what you have just described, is that what is
25 known as congestion?
866
1 A. Let me be more precise, congestion is a state where you
2 have multiple incoming flows of packets attempting to go out
3 on a shared link and the input rate exceeds the output
4 capability. So what happens is the packets get skewed as much
5 as they can be, but there is a limited amount of memory in any
6 one of those routers. As soon as that queue is full, packets
7 are dropped and they simply are not delivered. The state at
8 which you drop packets is the state of congestion.
9 Q. What is the impact of the congestion on the rates of
10 speed?
11 A. Because congestion means packets are being dropped and not
12 being delivered, if you go to the software on the in-points,
13 they are charged with the responsibility of making sure every
14 packet gets through exactly as it was presented, so as the
15 packets are coming out is exactly the same as the packets that
16 go in. If you ever detect a packet didn't get through, then
17 the sending software is responsible for retransmitting that
18 packet. But the bottom line, you will not get the same rate
19 that the theoretical capacity of the links might have
20 suggested you would.
21 THE COURT: When you say they are dropped, does that
22 mean they are never transmitted or they are delayed?
23 THE WITNESS: They are dropped. They will be delayed
24 as long as they are queued in any one of the routers, but they
25 are literally, if I have no room to hold them, I reject the
867
1 packet as it is coming in.
2 THE COURT: The question whether a packet is dropped
3 or simply delayed depends on the capacity of the router?
4 THE WITNESS: That is right.
5 THE COURT: Go ahead.
6 Q. What are some of the other pinch points that you
7 mentioned?
8 A. Well, the software that I have been talking about is TCP,
9 transmission control protocol. It is the key protocol of the
10 Internet technology. TCP is a very delicate protocol in that
11 it has to be tuned just right to get the transfer rates that
12 the underlying bandwidth might suggest. So, for example, TCP
13 will allow some number of packets to be in transit before
14 getting an acknowledgement which is a packet that says I got
15 packet 42, before getting an acknowledgement back.
16 You would not want to be in a situation where I sent
17 you one packet and waited for you to respond I got it and then
18 send a second packet and then respond because then I have only
19 one packet in flight and cross-country legacies being what
20 they are, you have very low bandwidth, you get very far from
21 the capability of underlying links.
22 So what TCP does, it will send multiple packets and
23 multiple packets in flight, ideally as many as the Internet
24 capacity can hold. As I said, it is a simple matter of
25 tuning. If your TCP has not been tuned to keep enough packets
868
1 in flight, you will get far less bandwidth than you might have
2 expected.
3 Q. What are some of the solutions if any for congestion?
4 A. Well, there is no solution for congestion aside from
5 putting in more capacity. All we can do is program nodes so
6 they detect when congestion is happening so the sources stop
7 sending so fast because if they continue to send fast, the
8 network will eventually collapse, which is get no useful work
9 done, I am spending all my time sending packets that will
10 eventually be dropped so nothing gets through. So the
11 solution is what TCP does today which is slow down whenever it
12 detects that congestion is happening on the network.
13 Q. Have you ever heard of technology called Napster?
14 A. Yes, I have.
15 Q. Is Napster or use of Napster clogging the Internet?
16 MR. COOPER: Foundation.
17 THE COURT: Sustained.
18 Q. Are you aware of any impact on the Internet of Napster?
19 A. I am not aware of any studies that say Napster is
20 accounting for "X" percentage of the packets being exchanged.
21 It has not come up in any circles I have been in where people
22 point to Napster as detectable in the traffic.
23 Q. Are you aware of any other file-sharing technology similar
24 to Napster?
25 A. Gnutella is the only one I can think of.
869
1 Q. Are you aware of any impact on the Internet of Gnutella?
2 A. No.
3 Q. Is there, at Princeton University, is there what is
4 reported in the press as a Napster problem? Do you understand
5 what I mean by that?
6 A. Princeton has not taken any action to limit the use of
7 Napster is all that I know.
8 Q. What is the network topology?
9 A. Well, roughly speaking it is like this, and I'm not the
10 Princeton system administrator so I am sure there are details
11 that I don't know. Each dorm at Princeton currently has a 10-
12 megabit shared network for all the students within that dorm
13 that would then be connected by 100-megabit ethernet into the
14 central facilities of the university. So it is one of a --
15 like a tree, like the central facility, the core, 100 megabits
16 going down the dorms and various departments as well and in
17 the dorms, you would have 10 megabits shared. Within
18 individual apartments, you have richer connectivity because
19 there are other things they are trying to do.
20 Q. Are you aware of use of Napster in uploading, downloading
21 MP3 files at Princeton?
22 A. Not that I am aware of, no.
23 Q. Could the Internet sustain file transfers of a size of 650
24 megabytes in any kind of significant volume?
25 MR. COOPER: Object to the form, and foundation.
870
1 THE COURT: Yes, sustained.
2 Q. Are you aware of the size of a Napster file?
3 A. Napster files are in the neighborhood of 3 or 4 megabytes.
4 Q. Are you aware of a general volume of Napster files that
5 are being transferred over the Internet?
6 A. I'm not aware of the volume of transfers, no.
7 Q. Could the Internet as it is now configured handle a
8 significant number of transfers of 650 megabit files?
9 MR. COOPER: Object to the form and foundation.
10 THE COURT: Sustained as to both.
11 Q. Based on your experience and your expertise with
12 networking the Internet, is it your opinion that the Internet
13 as it is presently configured could handle thousands of 650
14 megabyte files transferred?
15 MR. COOPER: Same objection.
16 THE COURT: Sustained.
17 Q. I would like you to assume the following scenario, that
18 tens of thousands of 650-megabit files are being transferred
19 each day and could you tell us what your opinion --
20 A. A minute ago you asked me if I knew what the volume of
21 Napster was and I don't know in any detail what the volume of
22 Napster is. But I can say, I can make a comparison with 650
23 megabyte video files and say for whatever fault that volume
24 is, I can talk about how that would strain the Internet.
25 Q. Please do so.
871
1 A. It doesn't matter what that volume is. If you take
2 Napster to be a problem and 10,000, 100,000 or a million and
3 you suddenly just translated 3 megabytes into 650 megabytes,
4 then I could talk about that.
5 MR. COOPER: I would still like a foundation if I
6 could, your Honor.
7 THE COURT: I think you need it.
8 MR. HERNSTADT: I am asking him to assume -- he
9 stated he knows the size of a Napster file. I am asking him
10 to assume there are tens of thousands --
11 THE COURT: Look, this is a little bit along these
12 lines, you have the Tappan Zee bridge going across the Hudson
13 River up there, and we don't know how many cars are going
14 across it, and so far we don't know how many lanes it is and
15 you are saying if instead of whatever numbers of cars are
16 going across it, we had a lot more very big trucks, would
17 there be a problem, and I think you can see that there are
18 some problems with that question.
19 MR. GARBUS: Your Honor, can we take a five-minute
20 break?
21 THE COURT: Yes.
22 (Recess)
23 Q. Professor Peterson, have you ever heard of DivX?
24 A. I heard of DivX, the intake for compression tool, yes.
25 Q. Do you know what -- do you understand a part of this
872
1 lawsuit is about the threat that plaintiff has alleged will be
2 posed by movies being on DVDs being decrypted by DeCSS
3 compressed to a 150-megabyte file and then sent on the
4 Internet?
5 A. Yes, I understand that's the basic idea.
6 Q. Does that as a new technology and new threat -- in other
7 words, a year ago, DeCSS has been around since approximately
8 October of 1999.
9 THE COURT: Is there a question?
10 MR. HERNSTADT: Yes.
11 Q. Let me ask you to assume that as of July 2000, there is
12 additional traffic on the Internet consisting of these
13 650-megabyte DivX files of DVD movies?
14 A. If people started to transmit 650-megabyte DivX --
15 MR. COOPER: Your Honor, is there a question?
16 THE COURT: There is not yet.
17 Patience, Mr. Peterson.
18 Q. My question to you, Professor Peterson, based on your
19 experience, do you have an opinion as to what the effect on
20 the Internet would be posed by a new flow of 650-megabyte
21 files being added to it?
22 MR. COOPER: Objection to the form. It is an
23 incomplete hypothetical.
24 THE COURT: It sure is. I am going to hear it sooner
25 or later, so I may as well hear it now and then I will ask the
873
1 next question and then we will be past this I think.
2 Go ahead, answer.
3 A. I believe the question being asked is if the transfer of
4 650-megabyte files in the Internet -- I'm not sure if that's
5 what he asked -- I am sorry. I am having a little trouble
6 knowing exactly what it is I should answer.
7 THE COURT: I think I know where Mr. Hernstadt is
8 going and I will help him get there.
9 The net has a finite capacity today, right?
10 THE WITNESS: Correct.
11 THE COURT: It has whatever level of congestion or
12 lack of congestion that it has, right?
13 THE WITNESS: Right.
14 THE COURT: If there is a big new load placed on it,
15 depending on how big the load is and what the current capacity
16 is, it may or may not put a strain on the capacity of the
17 system, true?
18 THE WITNESS: True.
19 THE COURT: When networks such as the Internet in the
20 past have encountered limitations due to the volume of traffic
21 pressing their capacity, the tendency has been to expand the
22 capacity, right?
23 THE WITNESS: Right.
24 THE COURT: Kind of like the interstate highway
25 system?
874
1 THE WITNESS: That's an accurate analogy.
2 THE COURT: All right, let's go, Mr. Hernstadt.
3 MR. HERNSTADT: Thank you. Actually, I thought you
4 were doing very well, your Honor.
5 Q. Professor Peterson, in such an instance, how much of an
6 expansion would there have to be?
7 A. Well --
8 MR. COOPER: Your Honor.
9 THE COURT: Obviously the answer is it depends. How
10 many messages, what the capacity is, whether there is
11 overcapacity or undercapacity, unless you have got some
12 specifics here, you are nowhere, and even if you have the
13 specifics, you have to address the expansion question.
14 Q. If we are talking about the addition of 650-megabyte files
15 being traded or being sent via the Internet by let's say --
16 and let us assume for the purposes of this at an additional
17 rate of 10,000, 10,000 650-megabyte files per day, will that
18 pose a strain on the Internet?
19 A. I can't say whether 10,000 transfers exactly would pose a
20 strain. What I can say, if you look at the Internet today as
21 to what is a common-sized file that's transferred and make the
22 assumption that the Internet is engineered for what is
23 happening today, that you see files in the neighborhood of
24 like the -- like the music example, MP3, 3, 4, 5, 6, 7
25 megabytes, that's a typical transfer file size at the high
875
1 end. There are certainly smaller transfers, but that is
2 considered to be a sizable transfer in today's Internet, 650
3 megabytes is two orders of magnitude or a hundred times larger
4 than that.
5 (Continued on next page)
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
876
1 BY MR. HERNSTADT:
2 Q. And based upon your experience, do you have an opinion as
3 to how long it would take for the backbone of the Internet to
4 be increased by two orders of magnitude?
5 MR. COOPER: No foundation for that question, your
6 Honor.
7 THE COURT: Sustained.
8 Q. Professor Peterson, do you have any experience with the
9 growth of the backbone of the Internet?
10 A. Yes, I do.
11 Q. Do you know how fast the Internet, the backbone of the
12 Internet, has developed, how much speed on the Internet has
13 increased over the last ten year?
14 A. Yes, I could walk through that if you'd like.
15 Q. Please.
16 MR. COOPER: Your Honor, I would like to have some
17 foundation for the information the professor is about to give.
18 THE COURT: I'm satisfied. Overruled.
19 A. Sir, there are different ways of measuring exactly what we
20 are talking about here, but let me give you a couple of
21 example ways of looking at this.
22 If you start in 1990, approximately ten years ago,
23 the Internet backbone at that time run by the National Science
24 Foundation consisted of 1.5 megabits per second links. In
25 1993 that was upgraded to 45 megabit per second links.
877
1 THE COURT: What was the second one?
2 THE WITNESS: 45 megabits. If you go to 1995, we are
3 now seeing some new technology introduced, we start to see the
4 commercial carriers become bigger and bigger players. A
5 cutting age network that was a follow onto the Nation Science
6 Foundation's network was running at 155 megabits per second.
7 That was the cutting edge technology in 1995 for the long haul
8 lengths that make up the Internet.
9 If you keep following that history, and I don't have
10 an exact date here, but it's in the '98 neighborhood, I don't
11 know exactly when this particular turnover took place, we
12 started to go see the OC12 that we were talking about a little
13 earlier in the backbone, 622 megabits per second.
14 Today you can find OC48 links, they are at 2.4
15 gigabits. So if you look at the peak technology available
16 over a ten year period, you will see that it's increased two
17 orders of magnitude over that ten year period, if I've got my
18 numbers right.
19 If you focus just on the last five years, which I
20 think is probably a little bit more appropriate because that's
21 the technology that's carrying us today, OC3, OC12, OC48,
22 we've gone from bandwidths measured in the hundreds of
23 megabits per second to gigabits per second. That's one order
24 of magnitude improvement in the backbone's capacity in a five
25 year period. It could be a little bit more precise, but for
878
1 now I'll stay with orders of magnitude.
2 We are looking at if you take the assumption that you
3 need to improve the capacity of Internet by two orders of
4 magnitude, so that 650 megabyte file transfers, because as
5 common place as today, 3 megabyte or 4 or 5 megabyte
6 transfers, that's going to take two orders of magnitude
7 improvement in capacity. As I just said, it was a five year
8 period to see the last order of magnitude improvement.
9 Q. Could you define what is an order of magnitude?
10 A. Factor of ten. Two orders of magnitude is a factor of
11 100.
12 Q. What is the connection between the increased capacity of
13 the backbone and the increased connection speed for the
14 average consumer, connection to the home?
15 MR. COOPER: Can we have a foundation, your Honor?
16 THE COURT: I don't even understand the question.
17 Try again.
18 Q. Professor Peterson, is there a connection between how fast
19 a connection is available to the home and how much capacity is
20 available on the backbone?
21 A. Yes, because more consumers in the home start sending in
22 and receiving data, that data is traveling over the shared
23 capacity of the backbone, the backbone capacity has to keep
24 pace as you add more users or you add larger files that those
25 users are transferring.
879
1 Q. Professor Peterson, in your opinion, what would be the
2 effect on the effective transfer schemes of files if the
3 average file size being transferred increased but the Internet
4 backbone did not?
5 MR. COOPER: I object to the form of the question,
6 because it's an incomplete hypothetical, and I think it still
7 lacks foundation. I think it's the same question.
8 THE COURT: Look, I am going to take it, but let me
9 say this: Obviously I have a lot of respect for Professor
10 Peterson's expertise, but this just has so many limitations
11 that it's close to being of no value, because there are so
12 many factors here that you are not controlling for and so many
13 things that we don't know about, that it's just not of much
14 help. But go ahead.
15 Q. Professor Peterson, what are the factors that determine
16 the transfer rate that a home user would obtain when sending a
17 file from his or her home to somebody else?
18 A. The factors that would come into play would be the
19 capacity of the links between the sender and receiver, and we
20 have already been talking about the backbone, but it might
21 most likely be limited by those two edge links which would be
22 the uplink speed of the sender and the download speed of the
23 receiver, because these links are typically not symmetric into
24 the home. And while I may be the only one using my end link,
25 the links that are in the middle of the network are being
880
1 shared with thousands and millions of other users, and so
2 congestion is going to happen on those links, and of course it
3 depends on the level of congestion exactly what transfer fee
4 I'm going to get.
5 Q. Is the Internet capacity today sufficient so that
6 congestion is only an occasional problem in terms of slowing
7 transfer rate down?
8 A. Congestion happens all the time on the Internet today.
9 Q. If everything else remained the same but the size of the
10 average file being sent increased, is it predictable what the
11 impact would be on the transfer speeds?
12 MR. COOPER: Same objections, your Honor.
13 THE COURT: Overruled.
14 A. There would be increased congestion. I would expect there
15 to be increased congestion.
16 Q. And what is the connection between increased congestion
17 and transfer speeds?
18 A. Congestion is probably the predominant factor after the
19 actual link speed that affect the transfer rate.
20 Q. And to get a sense of what two orders of magnitude means,
21 could you compare that to an increase, for example, in
22 microprocessor speed of two orders of magnitude?
23 A. Just as an example a lot of people today would consider
24 having a 500 megahertz PC as being pretty good. Two orders of
25 magnitude back in time, that was 5 megahertz, which has been
881
1 quite a while ago in a lot of people's memory.
2 MR. HERNSTADT: Thank you very much, Professor
3 Peterson.
4 THE COURT: Thank you. Mr. Cooper.
5 CROSS-EXAMINATION
6 BY MR. COOPER:
7 Q. Good morning, professor.
8 A. Good morning.
9 Q. I believe you testified regarding some information
10 regarding the Internet backbone and made reference to OC12.
11 Those are routers, is that correct?
12 A. OC12 is a link speed.
13 Q. Okay. And the speed is 622.08 megabytes, is that correct?
14 A. I forget if the .08 is correct, but it's 622. Something
15 megabits per second.
16 Q. And the speed at which the switchers and transfers are
17 I'll to operate now have increased significantly from the
18 OC12, have they not?
19 A. Today there are OC48 links in the backbone.
20 Q. Are there not OC96?
21 A. There are probably some OC96. I'm not aware specifically
22 but that's possible.
23 Q. Are there not OC92 that have an effective rate of 9.93
24 megabytes current any use?
25 A. I'm not aware of their use in the Internet backbone.
882
1 Q. Are you aware of their use in connection with work being
2 done at the high speed connectivity consortium?
3 A. I'm not specifically aware of that consortium.
4 Q. Have you not heard of the consortium, a consortium that
5 involves the Carnegie Mellon University, Cisco Systems Inc.,
6 the Corporation for National Research Initiatives among
7 others?
8 A. There are lots of consortiums like that. I'm not familiar
9 with that particular one.
10 Q. Your writing partner in your book is Mr. David, is that
11 correct?
12 A. Um-hum.
13 Q. He is a fellow with Cisco Systems, is he not?
14 A. That's correct.
15 Q. The only experiment you conducted in connection with your
16 testimony today, as I understand it, was to check the
17 effective download speed of your home DSL, is that correct?
18 A. That's correct.
19 Q. And the effective download speed that you experienced was
20 2 meg, correct?
21 A. 2 megabits per second, yes.
22 MR. COOPER: I have no further questions, your Honor.
23 THE COURT: Mr. Hernstadt.
24 REDIRECT EXAMINATION
25 BY MR. HERNSTADT:
883
1 Q. One question. Professor Peterson, what is your Internet
2 connection from your home?
3 A. It's a DSL.
4 Q. Where does it go?
5 A. It goes directly into the department, so I have a leased
6 line directly into the department, so the department is in
7 essence my ISP.
8 Q. What is the connection of your department?
9 A. The department into the Internet?
10 Q. Yes.
11 A. We are connected by 100 megabit into the campus. The
12 campus is then connected to the Internet at I believe 50
13 megabits per second.
14 MR. HERNSTADT: Thank you very much.
15 THE COURT: Professor Peterson. Thank you very much.
16 MR. ATLAS: We are going to be calling Professor
17 Peter Ramadage. I wonder if we could have a short break so he
18 can set up a laptop computer.
19 THE COURT: Sure. We will take 15 minutes.
20 (Recess)
21 THE COURT: Mr. Atlas.
22 MR. ATLAS: Good morning, your Honor. Defense calls
23 professor Peter Ramadage.
24 PETER RAMADAGE,
25 called as a witness by the Defendants,
884
1 having been duly sworn, testified as follows:
2 DEPUTY COURT CLERK: State your name, spelling your
3 last name.
4 THE WITNESS: Peter Ramadage, R-A-M-A-D-G-E.
5 THE COURT: Proceed, Mr. Atlas, please.
6 DIRECT EXAMINATION
7 BY MR. ATLAS:
8 Q. Professor Ramadage, where are you presently employed?
9 A. Princeton University.
10 Q. What do you do at Princeton?
11 A. A professor in the department of electrical engineering at
12 Princeton University, so I am engaged in teaching both
13 graduate and undergraduate students, and also research.
14 Q. Do you teach in any specialized area?
15 A. Yes, I do teaching and research in the area of digital
16 video libraries, digital signal processing for video, search
17 techniques for digital video, video compression and
18 transcoding of digital video.
19 Q. Could you briefly go through each one of those areas in
20 which you teach and just describe them for the Court, please.
21 A. My primary focus at the moment is in digital video
22 libraries. With the ever increasing amount of digital video
23 that's available, it becomes important to have an indexing,
24 search and browsing mechanism for retrieving video content
25 once it has been archived. We are developing tools which will
885
1 go into producing systems to achieve that goal. This is a
2 relatively new area, so we are only at the very beginning of
3 the research. We are developing very elementary building
4 blocks for different types of searches, different types of
5 browsing mechanisms to aid in the formation of digital
6 libraries.
7 Q. Could you describe the area of signal processing for us,
8 please.
9 A. Okay. Signal processing generally is the area in which
10 you input a given signal, let's suppose it's a video signal,
11 for example, and then you pass those bits that you are waiting
12 through an algorithm, whose purpose is to come up with either
13 a different version of what has been input or to answer some
14 question that the user has input based on the video content
15 that you are searching or inputting into the algorithm.
16 Q. I think you also mentioned compression technology.
17 A. Yes.
18 Q. What do you do in compression technology?
19 A. Let's restrict our attention to video files. Because
20 video is a very, very intensive signal to convert to digital
21 format, it requires very many bits to convert adequately into
22 the right format. It results in a very large file. Now to
23 store those files in an efficient way, or to transmit them to
24 another person in an efficient way one normally uses a
25 compression technology. That compression technology comes in
886
1 two forms. It's a form called loss less compression, and in
2 loss less compression the file is transformed into a format
3 which is smaller and requires less space but which can be
4 decompressed and you get back the original content completely
5 without any change. So it's called loss less, because no
6 information was lost.
7 Another type of compression technology is lossy
8 compression, and lossy compression you are willing to trade
9 off accuracy of the decompressed files, so the decompressed
10 file would be an approximation to the original, but you will
11 benefit by being able to get much, much greater compression.
12 That's typically used for consumer video, video on the
13 Internet, for example, video phones, wireless multimedia, they
14 all use lossy compression.
15 Q. Are you familiar with any other types of compression
16 technology?
17 A. Those are the main two. There is also a general purpose
18 compression technology used on computer files. Typically it's
19 under the name of zip files or LZW double compression
20 techniques. That's a loss less compression technology.
21 Q. You also mentioned transcoding. If you could briefly
22 describe that for us.
23 A. Okay. There are various applications when after a video
24 has been produced and it has been stored in a compressed
25 format that one needs to actually compress it further or to
887
1 transform it inn some other way to a different format.
2 To give you an example, if I have a video stored and
3 I wanted to transmit it to someone over a wireless
4 communication channel, there simply may not be enough
5 bandwidth on the wireless communication channel to transmit
6 the video in a reasonable amount of time in its current
7 format, so you would take that current format and you would
8 transcode it or basically you would take an existing
9 compressed file, process it, and to achieve greater
10 compression or change other attributes of the video such as
11 the frame rate or the frame size to make it better suited to
12 transmit over the wireless communication channel. That would
13 be one application of transcoding.
14 Q. How long have you taught at Princeton?
15 A. 16 years.
16 Q. What degrees do you hold and from what schools?
17 A. I have two undergraduate degrees, one in physics, one in
18 electrical engineering from the University of New Castle in
19 Australia. I have a masters degree in electrical engineer
20 from the University of New Castle. I have a Ph.D. from the
21 University of Toronto in Canada.
22 Q. Have you received any professional honors?
23 A. Yes. I have been awarded a complication medal from my
24 undergraduate institution. I have been awarded a best paper
25 award by the IEEE. I have been awarded several teaching
888
1 awards both from inside Princeton University and agencies
2 outside Princeton University; NSF research initiation grants;
3 IBM young investigator awards, I think.
4 Q. What is the IEEE?
5 A. The Institute of Electrical and Electronic Engineers.
6 Q. Have you received any teaching awards separate and award
7 from what you described?
8 A. Those are the main ones.
9 Q. Have you been published?
10 A. Yes, I have published over 80 referee journals and
11 conference articles.
12 MR. ATLAS: I am going to show the witness what we
13 have marked as Defendants' Exhibit BDE. It's a copy. I will
14 let the witness describe what it is.
15 THE COURT: Any objection?
16 MR. MERVIS: No objection, your Honor.
17 THE COURT: Your name is?
18 MR. MERVIS: Michael Mervis.
19 THE COURT: BDE is received.
20 (Defendants' Exhibit BDE received in evidence)
21 Q. Can you tell us what BDE is?
22 A. This is my curriculum vitae. I would say it's the latest
23 one that I just produced about two or three weeks ago.
24 Q. Does this reflect your published articles and conference
25 papers?
889
1 A. Yes. Yes, it's got the most recent articles we will be
2 presenting this September at the IEEE conference on image
3 conferencing. Those have been added.
4 Q. Have you ever testified before as an expert?
5 A. Not in court, no. I have been employed as an expert
6 witness on other court cases, but I haven't actually
7 testified.
8 Q. Have you been deposed before as an expert?
9 A. Yes, I have.
10 Q. In which case?
11 A. There is a case I. Omega v. Cyquest concerning the patent
12 infringement on removable media disk drives.
13 Q. Were you being compensated for your time in that case?
14 A. Yes, I was. I was being paid an hourly rate of -- it
15 started at $200 an hour and it went up to $250 an hour by the
16 end of the case.
17 Q. Are you being compensated for your time in this case?
18 A. No, I am not.
19 Q. Do I take that to understand that you are offering your
20 service as an expert for no compensation?
21 A. That's correct. I believed it is important to present my
22 point of view and perhaps the point of view of other
23 researchers in this area, that it's important for academics
24 and researchers in industry to have fair access use of digital
25 content in my own area of research. That happens to be
890
1 digital video. And I think it's very important for
2 researchers in digital video today to have access to the
3 digital video that's in the marketplace for fair use, use in
4 research. And down the road when this technology is mature I
5 think it would be important for other researchers in
6 disciplines not necessarily related to technology also to have
7 access to this digital information.
8 Q. Can you describe in a little bit more detail the specific
9 areas of research you are currently engaged in now at
10 Princeton?
11 A. Okay. Sir, we have I think several projects running
12 currently. One project has to do with searching video by
13 example, so we have stored in a data base a large amount of
14 video. This is a hypothetical example. What we would like to
15 do is someone comes in and says I would like to see what
16 you've got in your data base, clips that look like this
17 example, and they bring with them an example of what they
18 would like to see, what they would like to retrieve from the
19 data base, and we have been working on algorithms for quickly
20 searching through the data base to try and match the example
21 of what is in the data base and pull those thing out of the
22 data base.
23 Now because the video in the data base is stored in
24 compressed form to save space, it is important that these
25 algorithms actually operate on the compressed video. So we
891
1 have been developing algorithms that do that. That's one
2 project.
3 Another project has to do with multicamera video. We
4 have video from two cameras taking a video of a scene
5 simultaneously, and we are creating a synthetic video as seen
6 from a virtual camera on the baseline between the two real
7 cameras. That might have application in replays of sporting
8 events, it might have application in surveillance. The Navy
9 is very interested in this because they are putting cameras on
10 the decks of aircraft carriers, so we will also be talking to
11 the Navy about these possible applications.
12 Another project is using the information which is
13 embedded in the compressed video to quickly search through
14 that based on camera motion, try to estimate how the camera
15 was moving by using the information embedded in the compressed
16 video. And once you extracted what the camera was doing, that
17 can often help you say what is happening in the video.
18 To give you a specific example, in a basketball game
19 the camera often follows the ball very closely, so by
20 determining the camera motion you can quite easily pick out
21 things like fast breaks, jump shots, lay-ups and things like
22 this.
23 Q. Any other areas of research you are currently engaged in?
24 A. We also are developing some novel compression algorithms.
25 Based on some other work we think we might have some ideas
892
1 that will lead to novel compression algorithms. These are not
2 general purpose algorithms. These are rather specialized
3 compression algorithms that might be applicable in computer
4 graphics, compressing computer graphics or compressing video
5 game video.
6 Q. What is the general purpose of your research in this area?
7 A. We are working towards providing tools for the browsing,
8 manipulation and searching and indexing of digital video.
9 That's our general objective. This is a very new discipline,
10 so we are not working on everything right now. We are working
11 taking the very first steps and working on the very elementary
12 building blocks that would go into building such a system.
13 Q. When you say this is a new discipline, you are referring
14 specifically to your research or the general research in this
15 area?
16 A. The general research in this area is relatively new,
17 really has come to the forefront in the past five years.
18 Q. Do you have any expectation of how this research will
19 develop in the future?
20 A. The expectation is that after maybe in ten years time,
21 after we have all the elementary pieces worked out, people
22 will start to put these together into commercial systems or
23 even public domain systems, and then people from the
24 humanities, people from the social sciences, as well as people
25 from technology and the sciences will be able to use these
893
1 tools as part of their research. It will become a research
2 tool for people outside of the immediate technological area
3 where they were developed.
4 Q. Now in your research, and specifically in the development
5 of the algorithms you testified about, do you use digital
6 content?
7 A. Almost exclusively everything is based on digital content.
8 Q. How do you use the digital content?
9 A. One of the advantages of having digital video is that you
10 can search through it in a very quick way. Digital video
11 enables many things which are not possible with analog video.
12 You can search through the video, you can jump into it in a
13 random access type of way rather than sequentially have to
14 start from the beginning. You can store it very conveniently,
15 and you can create data bases and libraries of it very
16 conveniently. The idea is you can create a library of video
17 content, both video, sound as well as text.
18 Q. What are the sources of digital video content that you use
19 in your research?
20 A. We have a variety of sources, and I also should tell you
21 that at different stages of the research we use different
22 types of video, so we have our own camera, it's a digital
23 camera and it stores its video on a high quality digital tape.
24 From there we can transcode it into different formats,
25 whatever format we find most convenient. We find MPEG1 the
894
1 most convenient to work with, so we often transcode into MPEG
2 1 format.
3 Q. Explain what MPEG1 is.
4 A. MPEG 1 -- there are various standard compression
5 technologies. One of the very first is called MPEG1. MPEG
6 stands for Motion Picture Expert Group. It was a consortium
7 of people interested, industries interested in digital video,
8 as well as representatives of the standards organizations, and
9 they produced over a period of three to four years an ISO
10 standard called MPEG1.
11 We also download pieces of video that other
12 researchers have, and they make small clips of video available
13 on their web pages, either displaying the result of their
14 algorithms, or often in conjunction with that they will put up
15 the original video, unprocessed video, to allow other people
16 to try to replicated their results on videos. Those are very
17 short videos, less than ten seconds.
18 Occasionally my students are able to download some
19 digital video from the web that a movie studio has put up as a
20 publicity piece for a movie or something like this. So we
21 also have something like that.
22 In addition, on two occasions we have been able to
23 negotiate through an industrial partner permission to use an
24 extensive piece of video, copyrighted video from the content
25 producer. That is subject to much more restricted use.
895
1 Q. Those are the four primary sources of digital content you
2 have available to you now?
3 A. There is another source, and that is we can take analog
4 tapes and convert them into digital form. Several years ago
5 that was how we relied for getting digital video, but it
6 wasn't totally satisfactory. It's subject to the noise of the
7 analog recording process, then the noise of the analog
8 playback process, and then the peculiarity of the particular
9 digitizer that you use and compression hardware that you use
10 to actually do the digitization and compression. So we prefer
11 not to have to rely on that too much.
12 As I said, algorithms rely on processing the
13 compressed domain video, and we don't want to get tied to a
14 particular compressor, a particular piece of hardware that
15 does the compression, because then maybe our algorithms will
16 only work with that particular piece of hardware, so it's
17 important for us not to get too dependent on that mechanism.
18 Q. In terms of the variety of digital content that you have
19 just gone through, the five categories, do you find the
20 variety available in those categories best suits the type of
21 work you're doing?
22 A. There are problems with each of those video sources that I
23 have mentioned. Let me elaborate on that a little bit.
24 First we have our own camera, but it's quite a steep
25 learning curve to learn how to use the digital camera, so our
896
1 graduate students spend some time learning it if they need to
2 produce video, but the resultant video is not particularly
3 high quality. They are not professional cameramen, and they
4 are faced with various hurdles that they have to come up with,
5 and they are quite innovative in trying to overcome those
6 hurdles, but the resulting video is not commercial quality
7 video. For example, we can't hold the camera very steady.
8 It's a hand-held camera. We can't take multicamera video. We
9 only have one camera. And we can't control the lighting very
10 well.
11 The video that we download from the web from other
12 researchers is typically very, very short and sometimes has a
13 lower frame rate than we would like to work with, and
14 sometimes a very small frame size, small number of pixels in
15 it.
16 Video that we digitize from analog tape I think I
17 said already there is a couple of problems with that. I won't
18 go over those again.
19 By far the best quality digital video that we can
20 obtain is from DVDs.
21 Q. Have you heard of DeCSS?
22 A. Yes, I have.
23 Q. Have you used DeCSS?
24 A. Yes, I have. I have used it. I went to the --
25 Q. Wait. What do you understand DeCSS to be?
897
1 A. It's a program which will read the contents, the table of
2 contents of a DVD disk, and then you can ask it to descramble
3 the scrambled VOB files, the video object files on that disk,
4 and store them to your hard drive.
5 Q. You testified a moment ago that you used DeCSS. How did
6 you obtain DeCSS?
7 A. I went to a search engine, Google.com and I just typed in
8 DeCSS and did a search. It came up with about over 7,000
9 hits, so I did a bit of a search through those hits until I
10 found a site that had the software, and I downloaded the
11 source and executable program.
12 Q. Do you recall the site you downloaded it from?
13 A. No, because it involved a search through all of the
14 various hits I don't actually remember the site I eventually
15 found it on. Sometimes when you go to these sites you don't
16 actually end up on the front page of the site, you end up on
17 some lower page that has the feel you want, but it's not
18 immediately clear what site this is ultimately connected to.
19 Q. Do you know whether it's 2600.com?
20 A. I don't believe it was 2600.com, no.
21 Q. What form was DeCSS in when you downloaded it?
22 A. It was a zip file which is a loss less compression, that's
23 a general type of loss less compression used to compress
24 general computer files. When I decompressed that, I believe
25 it gave me the source, the executable and maybe another
898
1 document, I can't remember, a "read me" file or something like
2 that.
3 Q. Do you have an opinion on whether DeCSS would be useful to
4 you in your area of research?
5 MR. MERVIS: Objection, your Honor. There is no
6 foundation.
7 THE COURT: Overruled.
8 A. My initial experiments have indicated that's by far the
9 best source of high quality digital video available to us
10 today.
11 Q. What is the best?
12 A. The DVD. In terms of the breadth of video that's
13 available to us, the wide variety of video available to us
14 from various sources, and the high quality of the content,
15 DVDs are the best source for us.
16 Q. What I'm asking you is do you have an opinion on whether
17 the DeCSS utility would be useful to you in your area of
18 research?
19 A. Oh, absolutely, because the video content on DVDs is
20 scrambled, and so to get access to it we have to unscramble it
21 and that's exactly what DeCSS does, it unscrambles that video
22 content.
23 Q. If you were to have access to a wide variety of high
24 quality digital content like on DVDs, why would that be
25 helpful to your research?
899
1 A. Okay. Initially when we first start developing an
2 algorithm, we usually use very short pieces of video, because
3 videos are a very time consuming object to work with and it
4 takes a lot of space. But after we've got the prototype
5 working, it's very important to test the algorithm on a wide
6 variety of different video. It's important for two reasons.
7 First, you want to make sure your algorithm isn't somehow
8 dependent on the type of video you took or the type of encoder
9 you used. That's step number one.
10 Step two is you would like to search for video for
11 which it doesn't work, and to do that you need to get out
12 there and search through a whole range of different types of
13 video. You are specifically looking for video where your
14 algorithm fails to help you improve your algorithm or maybe
15 start up a different research direction.
16 Q. In order to use the video, the high quality video that's
17 available on DVDs, do you have an understanding of whether you
18 need to decrypt them first?
19 A. Yes. There are several types of files in the DVDs. There
20 are files with the extension IFO, which I believe is an
21 abbreviation for information, and those contain like a table
22 of content type information about what is on the DVD. There
23 are files with the extension BUP, which I believe stands for
24 back up. Those I think are back-ups for the IFO files. Then
25 there are files with the extension VOB, which I believe is an
900
1 abbreviation for video object. And those actually contain the
2 video and those are scrambled, and those are the things that
3 need to be unscrambled before you can actually use the video.
4 Now I should also add that that video is already in a
5 compressed form. It's in a compressed form called MPEG2,
6 which was the second extension of the ISO standard from MPEG1
7 to MPEG2, and that dealt with a higher quality video. The
8 initial base standard for MPEG2 was intended to produce
9 compressed video of comparable quality to broadcast video.
10 But it also includes higher level video standards as well.
11 Q. In terms of the digital video content that's commercially
12 available on DVDs, is that preferable to the digital content
13 that you described before, the digital content that's
14 currently available here?
15 A. Yes, it's very high quality, clean, no noise. It is
16 already digitized, which is excellent, which means we avoid
17 having to digitize it ourselves and then pick out any
18 peculiarities our own digitizer exhibits. So, it's a very
19 sound and preferable source of digital content. Also because
20 of the wide variety of DVDs available, it satisfies our need
21 for a large source of different varieties of video.
22 And in addition, I want to add one other thing. When
23 we use our own camera we try to avoid introducing any bias.
24 We don't want to sort of take video which is too favorable to
25 what we are trying to do, because the scientific method
901
1 demands that if you want to thoroughly test your algorithm you
2 have to use the video which has been taken independently by
3 somebody else. That's a basic fundamental premise of the
4 scientific method.
5 Q. Is there a relationship between the amount or variety of
6 digital video that you would use in connection with your
7 research and the confidence you would have in the results of
8 your research?
9 A. Absolutely. One of the criticisms -- I have been to
10 conferences. I have also given talks in industry, and one of
11 the criticisms that industry often has mentioned is that you
12 really need to test your algorithms on a wider variety of
13 video. I often hear that from people in the industry.
14 Q. Are there any other applications of your research that
15 would benefit from having access to the high quality digital
16 content of DVDs?
17 A. I think I have covered the main points.
18 Q. In terms of compression technology, do you use compression
19 technology in your research?
20 A. Yes, as I have said, our algorithms are designed to work
21 with compressed files, and because of that we need to have a
22 working knowledge of the compression technologies that are
23 employed. We also do our own transcoding of video from
24 different compression standards to a second compression
25 standard in order to test our algorithms on a variety of
902
1 compression standards.
2 Q. I believe earlier you testified that there was lossy
3 compression and loss less compression?
4 A. Yes.
5 Q. Do you know what type of compression is used on
6 consumer-oriented video like the DVDs you buy in a store?
7 A. Yes, both the MPEG1 standard and the MPEG2 standard are
8 lossy compression technologies. That means when the video is
9 compressed, information is thrown away. If it's a good
10 encoder, a good compressing algorithm, it will first try to
11 throw away information which is least perceptually
12 significant, but the more compression you ask for, the more
13 bits it has to discard or the more information it has to
14 discard, and eventually it is discarding important
15 information, and that shows up as artifacts in the resultant
16 video when you uncompress it.
17 Q. As a general matter, why would you use lossy compression
18 if you can use loss less compression?
19 A. Okay. Loss less compression that can achieve compression
20 factors of 2 is quite typical. If you are very lucky you
21 might get higher than 2. So, a 2 gigabyte file would be
22 reduced to 1 gigabyte with a compression factor of 2. By
23 applying loss compression, you can -- I'm quoting here from
24 the MPEG standard -- you can achieve compression factors of
25 around 20 to 30 without any visible perceptual difference
903
1 between the uncompressed video and the compressed video. So,
2 that's a factor of ten better at least than the loss less
3 compression technology. And in practice you often find
4 compression factors as high as 40 used.
5 Q. Now, could you just describe for us a little bit in
6 greater detail as you use more and more compression on content
7 what happens to that content.
8 A. Okay. Almost all of these compression technologies are
9 block based. What I mean by that is that each frame of the
10 video is divided into small blocks, nonoverlapping blocks in
11 the simplest case. These blocks are about 16 by 16, 16 pixels
12 horizontally by 16 pixels vertically.
13 The compression is based on those elementary blocks,
14 so when you start to throw away too many pieces of
15 information, those blocks don't get represented correctly in
16 the uncompressed video, so you start to actually visibly see
17 these blocks in the uncompressed video, and you start to see
18 miscoloring of the blocks.
19 One of the things that underlies most of the
20 compression technologies is that high frequency information is
21 discarded first. Now, high frequency information encodes
22 things like edges, where there are sharp transitions, sudden
23 changes in the image, so when you have discarded enough of
24 this high frequency information or too much of this high
25 frequency information, you start to see artifacts at the edge
904
1 boundaries, a phenomenon called ringing, where there seems to
2 be a time varying fluctuation around the edge, and you can
3 visibly see this in highly compressed video.
4 Q. Is that what you referred to before as an artifact?
5 A. That's an artifact. The fact that you can see the blocks
6 is an artifact. The fact that the blocks are miscolored is an
7 artifact, and this ringing is an artifact. There are other
8 artifacts as well, but those are the main ones.
9 Q. Do you have an understanding of an average size of a DVD
10 film?
11 A. Well, I have looked at the DVD for the movie Contact, and
12 the total amount of information on that DVD disk was about
13 seven and a half gigabytes. Most DVDs contain the original
14 movie as well as some additional add-ons such as director's
15 comments, or things that were cut from the movie might be
16 added on there, or special effects might be added on.
17 In the particular case of the DVD Contact, the actual
18 video file and the audio file took up about 6 gigabytes, and
19 the remaining gigabyte and a half were the extras and the
20 add-ons.
21 Q. Do you have an understanding of how much an average CDR
22 can hold in terms of megabytes or gigabytes?
23 A. The standard is 650 megabytes.
24 Q. So, if I wanted to copy a DVD that I went out and
25 purchased and copied it onto a CD, I would have to compress
905
1 the content to go from 6 or 7 gigabytes down to 650 megabytes?
2 A. That's correct.
3 Q. How would that be done?
4 A. Well, you would use a transcoder. Your video that you
5 have from the DVD after you've descrambled it, you must
6 descramble it first or else it won't work at all. Let's say
7 you descrambled the VOB files, now they are in MPEG2 format.
8 That has a bit rate -- and I will explain what bit rate is in
9 a second -- it has a bit rate of about 6 to 10 megabits per
10 second. Now the bit rate is how many bits are coming out of
11 the player per second in order to display the video on the
12 screen on average. That's an average rate. Sometimes it's
13 higher, sometimes it's lower. But on average it's about 6.7
14 megabytes per second. You have to now transcode that down to
15 a much, much lower rate.
16 You have available on the CD a space of 650
17 megabytes. Let's take the movie Contact. That's a two and a
18 half hour movie, so I have to get two and a half hours at 6.7
19 megabits per second down into a 650 megabyte disk. If you do
20 the conversion of units to make all the units appropriate, and
21 then work out the math, the bit rate after you have done the
22 transcoding for both the audio and the video needs to be
23 around 590 kilobits per second. So you need to go from 6.7
24 megabits per second down to 490 kilobits per second.
25 That's an enormous compression.
906
1 Q. Would that type of compression result in artifacts being
2 present on the ultimate product you end up on the CDR?
3 A. Yes, most definitely. The artifacts would be most visible
4 when there are scene changes, camera motion or when there is
5 movement of objects in the scene.
6 Just as a point of comparison, MPEG1, the rate for
7 MPEG1 which is generally to be believed below broadcast
8 quality video is 1.5 megabits per second.
9 Q. Based on your experience, if someone were going to
10 compress a film file to send over the Internet, is it more
11 likely that such a person would use lossy or loss less
12 compression?
13 A. If they used loss less compression you would take a 6
14 gigabyte file down to about 3 gigabytes. That is way too big
15 to transmit over the Internet. So that forces you, you must
16 use lossy compression if you are going to get it down to any
17 reasonable size. And even at 650 megabytes that's a very
18 large file.
19 Q. In terms of the use of lossy compression, are there
20 trade-offs to the user in connection with this type of
21 compression?
22 A. The user, if you want to do a transcoding from say MPEG2,
23 which is the DVD format, down to a much more highly compressed
24 format, there are various parameters you get to choose. The
25 most important parameter is the bit rate. Because you only
907
1 have a limited amount of space, you have 650 megabytes, and
2 you have a two and a half hour movie, that determines the bit
3 rate. That's fixed.
4 Now you can play with other factors. You can change
5 the size of the image. You can make the image smaller. You
6 can change the frame rate so the frame rate of the DVD is
7 29.97 frames per second. You could attempt to do better by
8 decreasing the frame rate.
9 Q. What effect would decreasing the frame rate have in terms
10 of viewing the ultimate product?
11 A. You would see fewer frames per second. If the video was
12 interlaced, which MPEG2 can be an interlaced video, I think
13 the interlacing would be much more prominent at the lower
14 frame rate.
15 Q. Could you just explain to us what the term "interlacing"
16 refers to.
17 A. Yes. Broadcast video doesn't present each image in the
18 video all at once on the screen. It actually does it in two
19 passes. In the first pass it draws every second line on the
20 screen. Then 1/60th of a second later it comes back and draws
21 the lines it missed. That's called interlacing. These two
22 scans are interlaced and they are 1/60th of a second apart.
23 Now if there is motion in the scene, then during that
24 1/60th per second things will have moved, so one of the
25 problems with interlacing is that you begin to see slight
908
1 jagged edges around objects that are moving because of the
2 interlacing.
3 Computer screens don't use interlacing. They use
4 what is called progressive scan, and progressive scan you
5 start at the top and you draw each line sequentially down the
6 screen.
7 Q. In your opinion, what is the current state of compression
8 technology?
9 A. Okay. Sir, compression technology has been a very hot and
10 active area of research for about the past 12 years. The
11 MPEG1 standard started around the end of the 1980s, started to
12 meet and form MPEG1 standard. The MPEG2 standard was even
13 started before the MPEG1 standard was finalized in the early
14 1990s. The JPEG standard was also in the first half of the
15 1990s. We are now just past, at the end of 1999 the second
16 generation of the JPEG standard, JPEG 2000. At this point
17 compression technology is maturing and we are seeing fewer and
18 fewer gains in the compression for each new generation of
19 technology that comes along.
20 In the past there have been some interesting
21 suggestions about potential ways of increasing the compression
22 rates dramatically. None of those have panned out yet. In
23 fact some of them have even been put aside now as areas of
24 research that are not likely to be productive.
25 Q. Looking out for the next five years or so, do you see any
909
1 breakthroughs in compression technology that would allow a
2 person to compress a file to a greater extent than is
3 currently available now?
4 MR. MERVIS: Your Honor, I object. There is no
5 foundation for this testimony.
6 THE COURT: I will take the answer. Overruled.
7 A. In talking to my colleagues, we often have lunch together,
8 some of them are very very actively involved in the
9 compression community and actually have graduate students
10 working together putting the pieces into the next ISO
11 standards.
12 THE COURT: Excuse me. I had understood from your
13 earlier testimony that you were offering yourself as an expert
14 on compression, is that right?
15 THE WITNESS: I am. I also want to let you know that
16 this issue is a hot topic of discussion in the department
17 about where is compression going.
18 THE COURT: Go ahead.
19 THE WITNESS: And it seems right now that the only
20 technology which looks promising is the wavelet technology,
21 and that has been the basis of the JPEG 2000 standard which
22 has just recently been issued.
23 So, it looks like from this point on there is only
24 going to be incremental improvement. There is nothing that is
25 known to be on the horizon which promises drastic gains.
910
1 Everything right now looks like it's going to be incremental
2 for the next five years. I can't see any further ahead than
3 the next five years, so I don't want to speculate beyond that.
4 Q. Fair enough. Have you heard the term fractal compression?
5 A. Yes.
6 Q. Do you have an understanding of what that is?
7 A. Around 1988 a researcher by the name of Barnsley proposed
8 this compression technology called fractal compression. I
9 think he has two patents on that technology issued around 1990
10 and 1991. He formed a company I think which is called
11 Iterated Systems, but since then nothing has really panned out
12 from that technology. The latest I heard in the journals and
13 other reading that I have done is that his latest version
14 produced by his company is not quite as good as JPEG.
15 THE COURT: Now, isn't there software on the market
16 in the digital photographic area known as genuine fractals?
17 THE WITNESS: I haven't heard of that software.
18 THE COURT: Isn't it in commercial use and supported
19 by service bureaus for high compression of digital
20 photographic files?
21 THE WITNESS: I have no information on that.
22 THE COURT: All right. Go ahead, Mr. Atlas.
23 Q. Have you heard the term DivX before?
24 A. Yes, I have.
25 Q. What is DivX, as you understand it?
911
1 MR. MERVIS: Your Honor, can I be heard on this? I
2 have an objection. This witness was deposed a week ago on
3 July 14, and as of that time the testimony was that he heard
4 of DivX, that he downloaded some things from the Internet that
5 he thought might be DivX, but he had done no further work on
6 the subject. So, whatever we are going to hear from the
7 witness in response to that question is something we have had
8 no discovery on, no notice of, and we will not have any
9 ability to effectively cross-examine.
10 THE COURT: Mr. Atlas, what about it?
11 MR. ATLAS: Actually at his deposition he testified
12 about what he had done with DiVX and what he intended to do
13 with DivX between the time of his deposition and the time he
14 would be testifying, and I suggest that if plaintiffs wanted
15 any further deposition before trial, they had an opportunity
16 to seek it. But I think the witness clearly put them on
17 notice of what he intended to do and what he looked to get out
18 of it.
19 THE COURT: Let me see the deposition and tell me
20 what you're referring to.
21 MR. MERVIS: Yes, your Honor. Can I approach?
22 THE COURT: I'm sure you will tell me the numbers.
23 MR. MERVIS: The subject of the DivX was covered on
24 the following pages: 15 through 17, 21 through 22, 25 to --
25 THE COURT: I can't remember them all that fast.
912
1 MR. MERVIS: I'm sorry. Perhaps you let me know when
2 you're ready.
3 THE COURT: Yes. What's after page 17?
4 MR. MERVIS: Pages 21 to 22, your Honor.
5 THE COURT: Next?
6 MR. MERVIS: 25 through 28, Judge.
7 THE COURT: Next.
8 MR. MERVIS: Page 70, your Honor.
9 THE COURT: Was there any representation made to you
10 about whether or not he would testify in this area?
11 MR. MERVIS: The only representation that I know of,
12 your Honor, was that his studies were continuing. And I would
13 simply point out in response to my adversary's comment, this
14 is expert testimony. Of course obviously we have been
15 proceeding somewhat informally, but the notion that we would
16 be given no notice of any further developments in this
17 expert's experiments I think is inconsistent with the general
18 standards of expert discovery. All of a sudden we are here
19 today hearing about DiVX, and we have never heard a peep out
20 of defense counsel.
21 MR. ATLAS: Your Honor, at page 16 he was asked
22 whether he had made any evaluation of DivX. He said he was in
23 the process of doing it.
24 THE COURT: I read it.
25 MR. ATLAS: You read that. I don't think this comes
913
1 as any surprise, and we have been talking to each other every
2 day -- not Mr. Mervis and I -- but they certainly were aware,
3 and I think if they wanted further examination of this witness
4 before trial, with eight letters a day, I think they could
5 have included that in a letter.
6 THE COURT: Is that all you get?
7 MR. ATLAS: It's not a surprise. I think it's clear,
8 given the time constraint, he put them on notice that he would
9 be doing this before trial.
10 THE COURT: I will hear the evidence.
11 MR. MERVIS: Your Honor, may I make just one further
12 application? I am prepared to cross-examine Professor
13 Ramadage on everything so far. I don't know what he is going
14 to say right now, but I do think that at the very least we
15 should have the opportunity -- and I understand the Court
16 wishes to hear the evidence now, but I do think we should have
17 the opportunity to take the witness's deposition on these
18 points and perhaps bring him back or perhaps just submit that
19 deposition testimony, because I don't think it's appropriate,
20 and I don't think it's fair to the Court, the witness or the
21 defense counsel for me to be doing cross-examination on that
22 topic. I think it's a better use of time to have a short
23 deposition and we can submit the appropriate exerts if there
24 are not --
25 THE COURT: Why not do that?
914
1 MR. ATLAS: I have no objection to that.
2 THE COURT: It's easy. You will finish your direct,
3 Mr. Mervis will cross-examine him to wherever he gets to where
4 he wants to insert DivX. You will get a deposition done in
5 the next three days sometime, indeed before the trial is over.
6 I certainly don't necessarily want to interfere with the
7 professor's weekend if I don't have to. Lawyers are
8 different.
9 MR. MERVIS: My son might disagree, your Honor, but I
10 understand.
11 THE COURT: I understand, but it's an occupational
12 hazard for sons of lawyers. And then we will proceed that
13 way.
14 MR. MERVIS: Thank you.
15 THE COURT: All right.
16 MR. ATLAS: Could you repeat back the last question
17 and answer, please.
18 THE COURT: I might observe this, and I do so because
19 I want to be corrected if I misunderstand. Don't I have in
20 evidence here Sleepless in Seattle and the Matrix, both in the
21 form of the original DVD and in the form of a DeCSS decoded
22 DivX'd compressed CD-ROM copy? Don't I have that?
23 MR. MERVIS: You do, your Honor, that's correct. I'm
24 not sure we have moved the -- yes, we have. Your Honor, you
25 do have both of them.
915
1 THE COURT: Okay. So I mean I suppose the purpose of
2 all that is to ask me to look at the two movies and see to
3 what extent I think that the copy is or isn't a commercial
4 threat in the marketplace.
5 MR. ATLAS: That's part of it, your Honor. I think
6 your Honor permitting, the witness has done his own test along
7 the lines of what Mr. Shamos did. It will not take very long.
8 THE COURT: That's fine. I said I would hear it, but
9 I'm just, you know, I don't know what form the testimony is
10 coming in in, but I think movie fans probably don't think in
11 terms of artifacts per movie. I think they look at the movie
12 and they say is this an acceptable quality to watch.
13 MR. ATLAS: I agree. That's exactly where we are
14 going in terms of whether it is acceptable quality. Can I
15 hear the Last question and answer.
16 (Record read)
17 A. DivX is what is called a codec, and a codec is an
18 abbreviation of two words, coder, decoder. So when you
19 compress a video, you also have to at some later time
20 uncompress it or decompress it. Each coder comes together
21 with its appropriate decoder. That pair is called a codec.
22 And DivX is an example of a codec, a coder/decoder pair.
23 Q. Have you used a DivX before?
24 A. Yes, I downloaded it from the web, installed it on my
25 laptop computer and on the work station in my office. Then I
916
1 employed a program called flask MPEG.
2 Q. Is this in connection with your work for this lawsuit?
3 A. Yes, it was with this lawsuit. I used a program called
4 flask MPEG to select the DivX codec. Now, let me add here
5 that there were actually two DivX codecs, one called a low
6 motion codec and one called a fast motion codec.
7 I experimented with both of these codecs, and I took
8 the descrambled VOB files from the movie Contact and I passed
9 them through this codec to compress them further down to a
10 much lower bit rate. I said before we need to go down to an
11 average of 590 kilobits per second. That's for both the video
12 and the audio. If we subtract off from that 590 128 kilobits
13 per second for the audio, then that tells us what our bit rate
14 has to be for the video, and so I have employed a bit rate of
15 about 450 kilobits per second to further compress the video
16 for the movie Contact into a compressed file in what is called
17 an AVI format. AVI is a Microsoft format.
18 (Continued on next page)
19
20
21
22
23
24
25
917
1 Q. What size are you looking to compress?
2 A. That was to get it down to 650 megabytes.
3 Q. Why was that?
4 A. That's the size you can write onto a CD writable disk.
5 Q. How long, the point we are at now in terms of this
6 experiment you conducted, can you tell me with respect to each
7 step how long it took?
8 A. Roughly. To download DeCSS took maybe three-quarters of
9 an hour. To use it to descramble all of the VOB files on the
10 DVD took about an hour. Then to use the DivX codec through
11 the flask MPEG program to transcode it down to about 650
12 megabytes, that was done in various stages because each VOB
13 file is about a gigabyte and there are about seven of them in
14 the movie. It took about two and a half hours per gigabyte.
15 At least one of them was not playable after I had transferred
16 it, there was some error, so I had to redo that one. After it
17 was all said and done, I would say it took about 20 hours to
18 do the transcoding.
19 Q. What did you do next in your experiment?
20 A. I needed to ascertain whether this -- all of these codecs,
21 including the DivX codec are what are call variable bit rate
22 codecs, so even though you tell it I want 450 kilobits per
23 second, it actually decides itself how many bits to use based
24 on the -- what's happening in the video at any given time.
25 If there is lots of fast motion, it will use more
918
1 bits and if there isn't very much action, if it is a very slow
2 scene, it would try to use fewer bits. So it tries to budget
3 the bits and do the best it can depending on how much action
4 there is in the scene. So even though you tell it ahead of
5 time to use about 450 kilobits a second, you may use more, you
6 may use less.
7 I did an experiment to see exactly how flexible the
8 DivX codecs were. It turns out that one of the codecs, the
9 one called the fast motion codec is very stingy with its bits,
10 it tries to use as few bits as possible to save space, but the
11 result in quality suffers. The low motion codec is very
12 generous with the bits. When more bits are needed, it uses
13 them and it hopes it can save later when the video isn't
14 moving as quickly. Generally, the video content was better
15 using the low motion codec.
16 Having decided that the low motion codec was the best
17 one to use, I then transcoded in particular -- I am sorry.
18 Let me step back a second. Because the low motion codec is a
19 variable bit rate because, if you say you want 450 kilobits a
20 second doesn't guarantee you will be under 450 kilobits a
21 second. In fact, you tended to go over by about 2 to 5
22 percent.
23 Q. So what did you do next in your experiment?
24 A. So I selected three representative two-minute slices of
25 the video for further examination. These pieces were selected
919
1 so that they contained some slow motion as well as some fast
2 motion, try to get an average sample of what it is going to do
3 on this slice. Two minutes of video is quite long. That's
4 3600 frames. And then I transcoded those two-minute segments
5 down to a target rate of 450 kilobits per second and I
6 examined the resultant quality of the video.
7 For my first example, I actually did it at several
8 rates to see what the effect of the rates would be and to try
9 to keep it of course below the required size and for the next
10 two examples, I just did it at the one rate, set rate of 450
11 kilobits per second. In all three examples, there are a large
12 number of visible artifacts, a sufficient number in fact that
13 I think the average person would say this stinks.
14 MR. ATLAS: With the court's permission, I believe
15 the witness can demonstrate the examples he was just talking
16 about on his laptop computer. If your Honor is willing, I
17 think the professor will take us through it.
18 THE COURT: Sure, let's get the exhibits marked.
19 MR. ATLAS: I have got Exhibit RDF is a copy that --
20 BDF.
21 THE COURT: Say it once more.
22 MR. ATLAS: Defendant's Exhibit BDF, the Professor
23 has this material on his hard drive. I have one for the court
24 and one for the defendants.
25 THE COURT: Are we playing it from the hard drive or
920
1 off this disk?
2 MR. ATLAS: We could do it off either.
3 THE COURT: Let's play it off the disk. I don't know
4 if what is in the hard drive and I don't know if I have a
5 record of what's in the hard drive.
6 THE WITNESS: In order to do a comparison, I would
7 first like to show the original CD.
8 THE COURT: Is that marked?
9 MR. ATLAS: Do you have it? I think I may have to
10 ask you to give up your copy of Contact.
11 THE COURT: Let's get it marked, please.
12 MR. ATLAS: Do you have the box it came on?
13 THE COURT: Mark it on the disk.
14 MR. ATLAS: Will that affect --
15 THE COURT: On this side.
16 THE WITNESS: It shouldn't affect on this side.
17 THE COURT: The original CD of the movie Contact is
18 Defendant's Exhibit BDG.
19 BY MR. ATLAS:
20 Q. Did you purchase this, by the way?
21 A. I did.
22 MR. ATLAS: If plaintiff's counsel wants, I will buy
23 you a copy but you seem to have a ton back there.
24 A. Let me open to the original DVD.
25 Q. First in terms of viewing it, where would the best place
921
1 for us to stand be?
2 A. Let me get it set up. So because it is a laptop screen,
3 the best thing to do it is view it perpendicular to the laptop
4 screen. We can adjust the screen. Please feel free to do
5 that.
6 THE COURT: It is fine.
7 A. The best place to stand is right around this little circle
8 here. That's a list of all the files on the DVD. The VOB the
9 encrypted video files. To try it, we double-click on the
10 info-file which is the table of contents and that will start
11 the DVD playing then what I would like to do is show you at
12 least two of the examples from the DVD that I transcoded and
13 that will be the best way of comparing the resultant quality
14 after transcribing to the quality before.
15 We will skip to the video start and then I think I
16 will jump to one of the scenes. Let's see if we can get there
17 OK. The first scene that I transcoded is this scene. I
18 transcoded the first two minutes starting from roughly around
19 here where you see Jodie Foster get up and start to walk
20 outside. So there is some slow motion at the beginning.
21 Notice how clear that antenna is right there. We will jump to
22 some fast motion. That's enough from that scene.
23 Let me jump to another part of the DVD. This is my
24 second example. I will fast forward a little bit until I get
25 to the scene that I have transcoded. This is my second
922
1 example. I have transcoded two minutes starting around here.
2 Q. Why did you choose this particular scene?
3 A. Again, I am looking for examples where it contains some
4 fast motion as well as some slow motion.
5 THE COURT: We will take a recess. Five minutes.
6 (In the robing room)
7 MR. GARBUS: I just saw the professor in the
8 bathroom. He is throwing up. He is very sick, what I would
9 suggest is we take a lunch break.
10 THE COURT: I guess we have no practical alternative.
11 All right, 2 o'clock. Fine.
12 MR. GARBUS: I presume he will be all right by 2
13 o'clock. I know nothing about it. If he is not, we will put
14 somebody else on.
15 THE COURT: Right, thank you.
16 (Luncheon recess)
17 (Continued on next page)
18
19
20
21
22
23
24
25
923
1 AFTERNOON SESSION
2 2:00 p.m.
3 THE COURT: Professor, better?
4 MR. ATLAS: He is. Before we begin, some
5 housekeeping matters in terms of our work for the weekend,
6 depositions and documents. I wanted to have a good idea of
7 how the Court wanted to handle these things if you wanted to
8 take five minutes now or at the end of the day.
9 THE COURT: Let's do it at the end of the day. Maybe
10 we can get this man on the road before rush hour.
11 Are we ready for the matinee?
12 THE WITNESS: Yes.
13 I am just starting the original DVD. We will see the
14 header first and then I will go to the second one, the second
15 scene.
16 THE COURT: Right.
17 THE WITNESS: The second scene will begin around
18 here, last for two minutes. I have selected this scene again
19 because it has both parts where there is not much motion,
20 camera pans as you see here, and other parts where there are
21 lots of moving pictures, so it is sort of a representative
22 two-minute sample. It is not much motion, pan of Jodi Foster,
23 here is a camera pan. Notice the clarity of people against
24 the background. You will see some ladies on top of a van in a
25 second singing. I want you to pay close attention to how
924
1 clear they took against the blue sky.
2 The ladies in the green T shirts against the blue
3 sky. This Elvis has something written on the back of his
4 guitar. See if you can read it. So you get the idea. That's
5 the crispness, another slow scene. I go all the way to the
6 point where this gentleman points to the Jodi Foster.
7 OK, so now let's go to the last example. That's
8 going to begin when part is released right about here and we
9 will see a two-minute clip from the point where I started this
10 video. This again has mixed scenes, some very fast action, a
11 lot of things changing all at once, interspersed with some
12 close-ups of Jodi Foster where there is a little bit of change
13 but not a fast change. This will be the most challenging of
14 the three samples after you are done transcoding.
15 Notice the individual lines here in this
16 computer-generated display in which she is traveling through
17 this tunnel, they are all clearly visible. And now a slow
18 scene. And then it will take off again and you will see a
19 similar transversal through another tunnel. I think you get
20 the idea.
21 Let's close the DVD now and I will put this CD, 650
22 megabyte CD but I have some short samples burnt into the CD.
23 THE COURT: That is BDF?
24 MR. ATLAS: It is marked, yes.
25 Now, let's close the DVD player. And open up that CD
925
1 which should be down here, right there, and here are my three
2 examples. Let's start with example one. I will play this, it
3 is in AVI format and I will play it through Microsoft media
4 player.
5 BY MR. ATLAS:
6 Q. Before you play it, what is AVI?
7 A. It is a format for video. It was a Microsoft format. The
8 media player is a Microsoft product. It comes standard with
9 the windows operating system. So this has been reduced in
10 size from the original frame size of the DVD, so I am going to
11 play this at full screen and this is the first scene that I
12 showed you on the DVD.
13 Notice the fuzziness of the general appearance. Look
14 at all the detail lost around the antenna area. If you look
15 closely around edges, you will see the ringing effect, the
16 halos around edges and along the tree line. This is the slow
17 scene, the slow part of the scene so this isn't the
18 challenging part of the scene yet.
19 These individual pieces aren't as crisp as they were
20 in the original DVD. Now here is the fast part. Notice all
21 of the artifacts, there is a phenomena called contouring where
22 you have big spreads of constant color. It is trying to code
23 this but it doesn't have enough bits. It is saying I don't
24 have enough bits for this, let me use fewer bits for this
25 whole area. You will see here in the Jodi Foster area a
926
1 pattern of blocks, that's the block artifact that I discussed
2 before. If we wait until she gets out of the car, we will see
3 it more prominently on the back of her shirt. Right there.
4 Notice the loss of detail of all of the faces in the
5 scene. Again, the coder, transcoder is trying to save bits
6 and place the bits were things are graphically changing, so
7 all the detail on the background has been lost. You can tell
8 it is rocks, but all of the fuzziness in here, and you can see
9 the detail has been lost. That's the first example. It is a
10 mixture, as I pointed out, of slow and fast.
11 We will close that one and start the second example.
12 The second example is the crowd scene. Again, you notice the
13 general fuzziness, not as crisp as before. Notice the ringing
14 artifacts around all of the moving edges. You will see the
15 effect of the block, blockiness there. The slow part, it is a
16 little bit better on the slow part. Ringing, see the ringing
17 around that lady just a second ago.
18 Whenever there is a scene change where there is a
19 rapid change, you notice all these artifacts more prominently.
20 The slow part, it does better on the slow parts. Ringing
21 around there. Let me stop it for a second and you can see all
22 the block artifacts there right across the scene. That
23 contributes to the general fuzziness of the appearance. There
24 are three ladies, green ladies coming up. Generally much
25 fuzzier, ringed, with the ringed artifacts. Here is Elvis,
927
1 see the back of his guitar, it is gone.
2 THE COURT: I wouldn't say it was quite gone. It was
3 not as distinct but it was not gone.
4 THE WITNESS: OK. That's the end of the second
5 example.
6 And the third example is the last scene and the part
7 that leaves the launch station and starts to go through the
8 tunnel. Here again, you see it struggling, it doesn't have
9 enough bits to fill in the detail. Notice all the block
10 artifacts right there. It is a little better on the close-up
11 scene where it is closer, lot of block artifacts. Everything
12 is changing here very rapidly. It just can't keep up. The
13 lines of the tunnel are not distinct, disappeared. We just
14 see the general color.
15 The slow part again, contouring artifacts. Rapid
16 change, you see the block artifacts all over the scene there.
17 Again, all the lines are lost, individual lines are lost,
18 block artifacts clearly visible. And I can stop it and you
19 can see all of the individual block artifacts. In fact, this
20 is a constant color, the contouring effect, it doesn't have
21 enough bits to fill in the colors that were there and says I
22 am going to fill in that patch with just one color. You have
23 seen all of the artifacts that are present.
24 THE COURT: OK, thank you.
25 MR. ATLAS: I just have a few other questions. If I
928
1 can move BDF and BDG into evidence.
2 THE COURT: Any objection?
3 MR. MERVIS: Only with respect to what we might see
4 in the deposition, your Honor.
5 THE COURT: All right, received on that basis.
6 (Defendant's Exhibits BDF and BDG received in
7 evidence)
8 Q. A few more questions, Professor.
9 Have you heard of the MPEG committee?
10 A. Yes, there are several MPEG committees. There was an MPEG
11 1 committee, an MPEG 2 committee. There was for a brief
12 period of time an MPEG 3 committee and MPEG 4 committee.
13 Those are all concerned with designing standards for video
14 compression.
15 Q. In terms of the MPEG 2 committee, is that the MPEG
16 committee that determined the standards for DVDs?
17 A. The MPEG 2 committee determined a standard and that
18 standard, part of what's called the baseline MPEG 2 standard
19 is what was used to encode and compress the video on that DVD
20 Contact that I used. I believe that is quite standard.
21 Q. Do you know who makes up the MPEG 2 committee?
22 MR. MERVIS: Objection, your Honor, relevance.
23 THE COURT: What's the relevance, Mr. Atlas?
24 MR. ATLAS: I think it will be clear in a minute.
25 THE COURT: Why don't you tell me the relevance.
929
1 MR. ATLAS: The MPEG committee, they have issued
2 certain findings in books and articles, they do studies of
3 different compression techniques and they have issued what
4 they consider findings on what level of compression is
5 acceptable for the consumer viewing.
6 THE COURT: How is that any better than your opinion
7 or mine?
8 MR. ATLAS: This is from a committee that is made up
9 of I believe motion picture companies and some of the computer
10 companies. These are the people who are in charge of deciding
11 what a good standard is for consumer viewing and I was going
12 to have the professor testify about the standards that they
13 have said are appropriate for consumers.
14 THE COURT: What's the basis, Professor, of your
15 knowledge, if any, about who is on the MPEG committee?
16 THE WITNESS: I don't know any of the individual's
17 names on the committee. I do know they are made up of
18 representatives from many industries and the international
19 standards office.
20 THE COURT: Do you know how many different
21 organizations are represented on the committee?
22 THE WITNESS: A large number but I don't know the
23 exact number.
24 THE COURT: Do you know whether they adopt standards
25 only by unanimity or by some vote short of unanimity?
930
1 THE WITNESS: I believe it is a vote short of
2 unanimity.
3 THE COURT: Go ahead, Mr. Atlas. We will see if
4 there is anything that the witness can testify to.
5 BY MR. ATLAS:
6 Q. Are you aware of any testing performed by the MPEG
7 committee, I gather the MPEG 2 committee, regarding different
8 compression technology?
9 A. All of the MPEG committees extensively test -- they call
10 for submissions or proposals for things to be included in the
11 standard and there is extensive testing of the submissions.
12 Then a final draft of standards is proposed. These are
13 extensively tested. It is my understanding from my reading of
14 articles and books they even do subjective testing with
15 bringing individuals in to rate the quality of the resultant
16 video.
17 MR. MERVIS: I would object and move to strike the
18 testimony on the grounds it constitutes hearsay.
19 THE COURT: I think so. It is stricken.
20 MR. ATLAS: I think these are things considered by
21 the expert in terms of giving his impression on compression
22 technology. This is part of the general literature.
23 THE COURT: If we are going to take what's part of
24 the general literature, we don't have a hearsay rule, do we,
25 as part of published materials?
931
1 MR. ATLAS: I will ask the witness in terms of
2 whether he relies on this material for -- I would like to ask
3 the witness whether he relied at all on these type of reports
4 in connection with any of the opinions he has given on
5 compression technology.
6 THE COURT: You better get, tie it up to a specific
7 opinion he has given in this case.
8 MR. ATLAS: May I take a minute?
9 (Pause)
10 MR. ATLAS: I have no further questions of this
11 witness, your Honor.
12 THE COURT: All right. Thank you. Cross-examination
13 Mr. Mervis?
14 MR. MERVIS: Thank you, your Honor.
15 CROSS-EXAMINATION
16 BY MR. MERVIS:
17 Q. Good afternoon, Professor Ramadge. Is it your belief that
18 you need to get access to encrypted DVD materials in order to
19 further your research into digital video?
20 A. I believe it would provide us with a very widespread,
21 reliable, high quality source of digital video which would be
22 a great benefit to our research.
23 Q. I see. The type of research that you described on your
24 direct, you have been conducting that sort of research for
25 about five years, isn't that right?
932
1 A. Correct.
2 Q. And the first time that you ever encrypted a DVD was in
3 connection with this lawsuit, isn't that correct?
4 A. That's correct.
5 Q. Now, on direct, you talked about working with industrial
6 partners, is that right?
7 A. Yes.
8 Q. Who are these industrial partners?
9 A. One of them is IBM. They have given us or negotiated us
10 the rights to use two pieces of video. The other one is the
11 Navy.
12 Q. And is it your contemplation that one of the end products
13 of the research you are developing now would be to develop
14 commercial tools?
15 A. No, I am not in the process of developing commercial tools
16 myself. We are developing elementary pieces or elementary
17 blocks which later might be put together by people who develop
18 commercial tools.
19 Q. To your knowledge, would that be something, is that
20 something that your industrial partners are contemplating?
21 A. That's something they are very interested in, yes.
22 THE COURT: Let me ask you, you said the Navy, this
23 technology that you have addressed in terms of indexing,
24 searching and browsing, without by any means disclosing any
25 confidential information, is of a great deal of interest to
933
1 the military, is it not?
2 THE WITNESS: It is.
3 THE COURT: It might, for example, be used to enable
4 targets to be identified electronically in aircraft that move
5 too quickly to permit a human being to identify the targets on
6 their own, right?
7 THE WITNESS: Certainly aspects of technology could
8 be used for such purposes but that's not the area of my
9 research.
10 THE COURT: No. I didn't suggest that. Let's leave
11 it at that.
12 I suppose the Department of Defense has an ability to
13 give you high quality video to work with, doesn't it?
14 THE WITNESS: They have various constraints on the
15 cameras. The cameras must be extremely robust because of the
16 environment and conditions in which they operate, so they have
17 a range of different qualities of video. We are working
18 actually with the Navy with a lower quality video.
19 THE COURT: All right, go ahead.
20 Q. You testified on direct one of your sources, current
21 sources of digital video content was content that your
22 industrial partners got from copyright owners by a license, is
23 that correct?
24 A. I don't know if it was by license, it was by some
25 agreement which I wasn't a party to.
934
1 Q. Your understanding is your industrial partners went out
2 and talked to the copyright owners and got permission to let
3 you use the content?
4 A. I don't know because I wasn't part of the negotiation. I
5 don't know who contacted who first to negotiate the permission
6 to use the copyrighted material.
7 Q. So is it fair to say that you have no firsthand knowledge
8 of any of your industrial partners ever approaching any of the
9 plaintiffs for permission to use any of their DVD products and
10 having that permission denied?
11 A. I wasn't present at such meetings.
12 Q. Is it correct, Professor Ramadge that lossy compression is
13 considered to be acceptable compression for consumer use?
14 A. There is a whole range of lossy compression. Certain
15 ranges of lossy compression have been tested and have been
16 agreed upon as acceptable for consumer use.
17 Q. Now, I am correct, am I not, that if you compress digital
18 content to 650 megabytes, you can put that on a single CD-Rom,
19 is that correct?
20 A. The file size has to be 650 megabytes, that is correct.
21 Q. And to do this, you need a few pieces of equipment, right,
22 a computer with proper software, you need a CD-writable drive,
23 and need the CD, correct?
24 A. Correct.
25 Q. And the computer that you need would run you roughly 2,000
935
1 bucks?
2 A. I think you need not CD but DVD.
3 Q. I am sorry. Thank you for correcting me. And the
4 computer would cost about $2,000, right?
5 A. That's in the right ballpark, yes.
6 Q. And the DVD drive would cost somewhere between 200 and
7 300, maybe a little bit less, right?
8 A. That's correct.
9 Q. And cost of the black CD, that's a writable CD, that's
10 what, a dollar?
11 A. Correct.
12 Q. You testified the film you were working with, Contact,
13 that was two and a half hours?
14 A. Correct.
15 Q. It was too long to put on a single CD?
16 A. No, it is possible to put it on to a single CD. I
17 designed those samples you just viewed, I designed the
18 compression factor to fit the movie Contact to a single CD.
19 Q. You could have used a lesser degree of compression and put
20 the film on two CDs, correct?
21 A. Correct.
22 Q. And the sum total of those two CD's would be $2?
23 A. Excuse me?
24 Q. The total cost of two CDs would be $2, right?
25 A. Yes.
936
1 Q. Now, Professor Ramadge, with all of your expertise and
2 experience with digital video, am I correct in understanding
3 that you don't own a DVD, sir?
4 A. I am sorry, a DVD player? There is a DVD player on my
5 laptop right there.
6 Q. What do you use for home viewing?
7 A. I use VHS, a tape player.
8 Q. You have never used a DVD player at home, have you?
9 A. I have used that DVD player at home.
10 Q. Since when?
11 A. Since I bought the laptop.
12 Q. When did you buy it?
13 A. Two years ago. But I don't own a DVD player at home.
14 Q. The 20 hours that you testified to that it took you to,
15 from beginning to end to complete the DivX-ing process, was
16 that 20 consecutive?
17 A. How would we count this, the files on the DVD are
18 separated into approximately 1 gigabyte chunks. There are a
19 few smaller chunks because the movie doesn't divide into an
20 even number of 1 gigabyte chunks. Each one-gigabyte file was
21 processed separately. Before I processed the next file, I had
22 to wait for the computer to finish and load in the next file,
23 type in the necessary commands to do that, double-check the
24 parameters to make sure nothing had changed.
25 After all, if it is going to take a while, you don't
937
1 want to make a mistake and come back, change the file name
2 where it was stored, check that there was enough empty space
3 on the disk, nothing will go wrong and run it again. It is
4 consecutive in that sense, do one file, stop, do another file.
5 However, in my initial investigations, I didn't do
6 all the files consecutively. I did some timing analysis based
7 upon largest of the files of one-gigabyte size and estimated
8 how long it would take and set aside certain portions of my
9 time to accomplish the rest.
10 Q. Am I correct in understanding that you didn't sit there in
11 front of the computer for 20 consecutive hours to complete
12 the task?
13 A. As I said, it depends how you count this. I didn't sit
14 there for 20 consecutive hours because I didn't do it all in
15 one sitting.
16 Q. Thank you very much. Is it also correct that certainly a
17 good portion of the time that it took, you had no reason to
18 look at your computer screen or have any other interface with
19 a computer, right?
20 A. For the purposes of my experiment, I did look occasionally
21 at the computer screen. I wanted to check the bit rate, the
22 current bit rate because it is a variable bit rate coda so
23 even though I say use such-and-such a bit rate, it could be
24 using less or using more. It tries to use the allocated
25 budget in an intelligent way. I periodically checked it to
938
1 see what the bit rate was.
2 Q. I understand but it is fair to say a good deal of the
3 DivX-ing process is automated, is that right?
4 A. It depends what you mean. The programs seem rather
5 delicate. I crashed the program several times. Several times
6 I tried to open the one-gigabyte files and Microsoft media
7 player crashed the system, so I think when you say automated,
8 it creates the impression you press the button and it is done.
9 It wasn't quite ought that automated. You needed to pay
10 certain attention to what you were doing. It was a hands-on
11 process.
12 Q. You didn't need to pay 20 hours attention?
13 MR. ATLAS: Objection.
14 A. I did glance away from the computer screen probably 50
15 percent of the time at least while it was doing the
16 transcoding.
17 MR. MERVIS: Thank you, Professor. No further
18 questions.
19 THE COURT: Thank you. Any redirect, Mr. Atlas?
20 MR. ATLAS: No further questions.
21 THE COURT: Thank you. Thank you, Professor.
22 (Witness excused)
23 THE COURT: Next witness?
24 MR. HERNSTADT: Defendants call Matthew Pavlovich.
25 He also has a computer and I don't know if you prefer he set
939
1 it up now. It won't take but a few minutes or if you prefer
2 he do it during his testimony when we get to that point --
3 THE COURT: I am indifferent.
4 THE WITNESS: If it is not a problem, I can set it up
5 right now.
6 THE COURT: Let's do it.
7 MATTHEW PAVLOVICH,
8 called as a witness by the defendant,
9 having been duly sworn, testified as follows:
10 DIRECT EXAMINATION
11 BY MR. HERNSTADT:
12 Q. Mr. Pavlovich, where do you live?
13 A. I currently live in Dallas, Texas.
14 Q. What is your present occupation?
15 A. I am an -- I work for a new technology start-up Media
16 Driver LLC.
17 Q. Are you one of the founders of that company?
18 A. Yes, I am a cofounder and currently the president.
19 Q. When did you first use a computer?
20 A. Could you repeat the question?
21 Q. When did you first use a computer?
22 A. We got our first computer when I was about 9.
23 Q. How long ago was that?
24 A. Thirteen years.
25 Q. Do you know the Linux operating system?
940
1 A. Yes, Linux is an open source alternative to other
2 operating systems on the market. It is developed by
3 individuals in the technology community.
4 Q. What does open source mean as relates to the Linux --
5 A. Open source is a term used by people who developed the
6 model, that applications that make up the operating system,
7 that the source code or recipe to program is available
8 publicly for anybody to use. The source code is generally
9 licensed and copyrighted, but the structure of the license
10 makes it so anybody can use it, make modifications, and things
11 of that nature.
12 Q. When did you first use Linux?
13 A. I first used Linux when I was a sophomore in high school.
14 Q. Is it your primary operating system?
15 A. It has been my primary operating system for the last five
16 years.
17 Q. How many computers do you have?
18 A. I have four.
19 Q. Do you use Linux on all of them?
20 A. Yes, they all -- all four run Linux.
21 Q. Can you tell us about your education, please?
22 A. I went to high school in Texas, senior class president for
23 Marcus High School in Flowermount, Texas. When I went to
24 college, I studied computer engineering at Purdue University.
25 Q. Did you graduate from Purdue?
941
1 A. No, I left Purdue University in mid April. I took a leave
2 of absence to start Media Driver.
3 Q. Could you tell us about your employment history?
4 A. I started back when I was in high school, worked for Best
5 Buy Incorporated. I managed the series of PC repair
6 technicians. All during high school and through college, I
7 have worked during the school years and in the summers and
8 some of the jobs kind of overlapped and intertwined.
9 After working at Best Buy, I also worked for a
10 company called Ping Tech Corporation. We built custom systems
11 and small networks for small businesses and things of that
12 nature.
13 While at Purdue University, second semester my
14 freshman year, I worked for Purdue Data Network. I worked on
15 the network upgrade, that was a big project there at Purdue.
16 After that, I worked for the school of liberal arts. I was a
17 systems and network administrator there, user base of 5 to 600
18 users.
19 After that, I worked for Allegiance Telecom which is
20 a publicly-held telecommunications company that specializes in
21 providing business telecom solutions.
22 Q. What did you do for Allegiance?
23 A. I was a support intern. I worked for hybrid networks
24 primarily UNIX and NT-based systems.
25 Q. Are you, do you know the Purdue network topology?
942
1 A. Yes, during my job there as network technician, we
2 installed, I installed hubs and switches, routers, connected
3 cable, during the major upgrade, the fiber optics, so, yes, I
4 know that work.
5 Q. What was upgraded, the internet connection upgraded?
6 A. Not at first. When I first attended Purdue, they had a
7 pair of T1s that was load balanced and what that does is take
8 two T1 lines and makes them act as one. T1 is a data
9 connection line of 1.55 megabits per second. And when I left,
10 they had upgraded the main connection to the internet to a T3.
11 However, the main focus of the upgrade was the internal
12 infrastructure at Purdue, the backbone, the connections to
13 labs, connections to dorms, things of that nature.
14 Q. What is a T3 line, what's is the access speed on that?
15 A. T3 is 44.5 megabits per second.
16 Q. Is that an increase of what magnitude?
17 A. 30 times.
18 Q. From what it was before?
19 A. I am sorry, it is 30 times larger than T1. We had a pair
20 of T1s, so that would only be 15 times.
21 Q. When the upgrade of the internet access to the university
22 was achieved, did that mean that the same increase in
23 magnitude was found in the dorm rooms?
24 A. No, when we were doing the network roll-out, updating the
25 systems, the primary goal was to improve efficiency on the
943
1 network. The early roll-outs were just put up almost in haste
2 just to get functionality and we were working through and
3 reworking things.
4 When you upgrade the main pipe to the internet, there
5 is still a lot of factors that take into account what the end
6 users, for example, the students in the dorm, are going to
7 experience when you upgrade a connection likes that. There
8 are several points within the network pinch points where
9 things bottleneck.
10 For example, our typical connection to a dorm would
11 only be 10 megabit shared connection and that shared
12 connection would be used by anywhere from 2 to 400 students.
13 When you upgrade the main connection to the internet, we
14 weren't necessarily upgrading those pieces.
15 Q. Did you live in the Purdue dorms?
16 A. Yes, I lived in the dorms my freshman and sophomore years
17 at Purdue.
18 THE COURT: When was that so I have a time frame?
19 THE WITNESS: That was fall of '96 through spring of
20 '98.
21 THE COURT: Thanks.
22 Q. What was the network available to your dorm room at
23 Purdue?
24 A. The --
25 MS. MILLER: Objection, I fail to see the relevance
944
1 to this line of questioning.
2 THE COURT: Overruled.
3 A. When I initially went to Purdue freshman year, the process
4 of the migration of the main upgraded backbone was in its
5 infancy and halfway through my freshman year, we actually saw
6 the fruits of it in the dorm. When we started, we had a 19.2
7 kilobyte connection and once the roll-out reached the dorm, we
8 had a 10-megabit shared connection. I had a 10-megabit
9 connection, every room had a 10-megabit line to it which would
10 funnel down so everyone in that one room would share a
11 10-megabit line to the campus backbone.
12 Q. What access speeds do you experience on the 10-megabit
13 line?
14 A. When I first got the line, when I first was connected, I
15 was one of the early few utilizing the new speed and we could
16 readily see 100-kilobits per second line when they first
17 installed it.
18 Q. You said 100 kilobits per second?
19 A. 100 kilobits per second.
20 Q. About 10 percent of the rated speed of the line?
21 A. Right, those numbers come -- I wasn't the only one using
22 the network and so other students using that.
23 The other thing when we first got that, the
24 university still had those load balance T1s and that line is
25 only 3-megabits per second line and the design of the network,
945
1 the dorms would never, one dorm room or one computer in a dorm
2 room would never have full access to the 3 megabits.
3 Q. Did the speed that you were able to experience change at
4 any time during your time in the dorm?
5 A. Yes, they actually got worse. During the roll-out,
6 everything was done in phases. We did one and two dorms at a
7 time and as we brought more dorms on line to the network, our
8 connection speeds and what -- the data transfer rates we
9 experienced gradually decreased.
10 Q. Before I asked you about 100 kilobits was 10 percent, I
11 meant to say 1 percent of the 10-megabit lines?
12 A. Yes, that's correct.
13 Q. Mr. Pavlovich, are you being paid to be here to testify?
14 A. No. However, I feel I should mention the Electronic
15 Frontier Foundation paid for my airplane ticket. I got a call
16 yesterday from Ed Hernstadt and they said they needed me up
17 here, so I caught the last plane of the evening, 7 p.m.
18 Q. We thank you for that.
19 Have you ever been involved in a project to develop a
20 piece of Linux software?
21 A. Yes. Just to kind of give you a quick background, when we
22 say project and what Linux is before we obfuscate and things
23 get confused, the development process --
24 THE COURT: When do we get to doing that, obfuscate
25 and things get confused?
946
1 THE WITNESS: I will do my best here.
2 A. We had a project, the way we define a project is the users
3 of the system, users of the Linux operating system and
4 primarily developers get to a point in their developing their
5 code that they realize we now have the foundation to work on
6 something else, whether it be 3D graphics, video playback, a
7 word processor program now, things like that.
8 Interested developers will make announcements on many
9 mailing lists, bulletin boards saying I am John Q. Developer,
10 I wish to develop this application, I will be working with
11 this web site, this mailing list, the code archive will be
12 stored here, and anyone else that wants to join in, come over
13 here and we will start working on here on this. In a short
14 period of time, anywhere from a half dozen to a hundreds of
15 developers will work on any given project.
16 Q. What is a code archive?
17 A. A code archive, we tend to give that name to a system we
18 use, when you have a large number of developers working on a
19 project, it is very easy for people to overwrite someone
20 else's work. So a code archive using a special program called
21 CVS allows you to make changes to code and store the changes
22 between each revision. So multiple developers can change the
23 same piece of code and if something gets squashed, you can go
24 back and revert to the original piece.
25 Q. What piece of code gets selected under the conversion?
947
1 A. I am sorry?
2 Q. How is there a selection of and different versions of the
3 same code?
4 MS. MILLER: Objection, foundation.
5 THE COURT: Sustained.
6 Q. Is one piece of code selected out of the different
7 versions of the same codes that are written?
8 A. When a developer, say we are developing a word processor
9 and fixing the spell check function and one developer sends up
10 a fix for a bug and another developer gets into a PC he meant
11 to get in two days ago and created another bug. What will
12 generally happen is they will look at it and just make another
13 patch to go on top of it. It is really a revolving process.
14 The last piece committed to a developer that has access to it
15 is generally the one people are working from.
16 Q. Have you ever worked in an environment where you have a
17 traditional process of making software?
18 MS. MILLER: Objection.
19 THE COURT: Sustained as to form. I don't know what
20 that means.
21 MR. HERNSTADT: I am sorry, your Honor.
22 Q. Do you understand what I mean when I say the traditional
23 way of writing software?
24 A. I think you mean the traditional commercial development
25 model. The open source model has many advantages in that it
948
1 levels the playing field, so if you are the most senior
2 developer or most junior developer, you both have equal say in
3 what goes on in the code base and the best idea wins or best
4 piece of code wins and that is very advantageous because a lot
5 of times, you will get a 17-year-old that is really sharp at
6 what he knows and it is a better way of doing something. In
7 traditional, it is more top-down manner, where the most senior
8 manager or developer says this is how we are doing it even
9 though it might not necessarily be the best way to do it.
10 Q. Which Linux projects have you worked on?
11 A. I have worked with several. The first one was the Debian
12 Linux distribution, a Linux distribution is a term we give to
13 another project that takes the collective of all the smaller
14 projects and puts it together in a way that end users can
15 easily install and maintain a system running Linux and it
16 makes it a lot easier than working from raw code base, so
17 there is much more continuity and stability to the system.
18 Q. Can you give an example of that?
19 A. An example of?
20 Q. A distribution, what might be bundles to go?
21 A. For example, Debian includes all the necessary tools and
22 applications one would need to set up a web server, a mailing
23 server, set up a desktop system for doing code development,
24 word processing, networks, virtually anything you can do with
25 a computer, there is an application available that you can do.
949
1 Q. What other projects have you worked on?
2 A. I worked -- I am sorry?
3 Q. What other projects have you worked on?
4 A. One of the developers I met while working on the Debian
5 project approached me and said he had received some
6 documentation on how to start getting 3D graphic support under
7 Linux. I was very interested in this because I was at that
8 time looking into what it would take to get a DVD working
9 under Linux, so I worked with him and we started the Utah GLX
10 project which was one of the founding projects for hardware
11 accelerated 3D graphics in Linux.
12 In fact, it was the founding project for that. My
13 role there, I lobbied several hardware manufacturers for
14 documentation. Our goal there was we needed specific
15 documentation to the various video chip sets in order for us
16 to support those cards in Linux.
17 Q. What does it mean when you say to support a card in Linux?
18 A. There is literally dozens of hardware vendors selling
19 common PC hardware and users often when they buy a system,
20 they are not sure which piece they get and in order for Linux
21 to run on the hardware they have, we have to have support for
22 their chips and so our goal is to get Linux to run on their
23 system, we have to have support for all the available chips.
24 Q. Does support mean that the operating system and the chip
25 can interact?
950
1 A. That's what I mean by support, the Linux can communicate
2 with the card, how the card was intended to be used for.
3 Q. Can you explain what a graphic accelerator is?
4 A. Graphic accelerator is a term used for the latest
5 generation of video cards where they specifically have designs
6 in them for doing 3D graphics, gaming, a lot of times they are
7 adding in support for hardware accelerated video play back,
8 motion compensation, video -- I can throw catch phrases all
9 day but --
10 Q. What was the next project you worked on?
11 A. After getting to the point where we had gotten to where we
12 needed to begin the DVD project, I spun a sister project off
13 from Utah GLX that became known as the Linux Video project or
14 for short, LiViD.
15 Q. Why did you start LiViD?
16 A. Quite frankly, I wanted to play DVDs on my Linux box. I
17 received documentation for a hardware decoder that worked with
18 my video code at the time and I wanted to be able to utilize
19 that decoder chip and the DVD drive and movies I bought under
20 Linux.
21 Q. What was your role with the LiViD project?
22 A. I was the founder, project leader of LiViD.
23 Q. What does it mean to be a project leader?
24 A. It can mean a lot of things, it depends on the project. A
25 lot of times the project leader in my case went out and
951
1 actively seeked developers working on similar projects,
2 video-related project. My idea was if we get all the
3 developers working on video and DVD-related projects, we could
4 all work from a common code base and have more a cohesive and
5 efficient code group as opposed to several projects working on
6 different things and not necessarily coming up with a
7 standardized set of tools.
8 Q. When did you start working on the LiViD project?
9 A. I founded LiViD June, July, in the time period of 1999.
10 Q. Are you still working on it?
11 A. Yes.
12 Q. How much time do you devote to LiViD?
13 A. When I originally started LiViD, I put upwards of 40 hours
14 a week, talking with developers, talking to hardware vendors,
15 looking at specifications, looking at code, writing code.
16 With my job and what I have been doing now, I have not been
17 able to commit as much time and now I would say I commit
18 anywhere from 10 to 20 hours a week.
19 Q. Are you being paid for any of your work on the project?
20 A. No, I am not.
21 Q. Have you ever heard of DeCSS?
22 A. Yes, DeCSS is commonly referred to as a Windows program
23 that decrypts DVD data.
24 Q. When did you first become aware of it?
25 A. Summer/fall period of '99 when it was first posted to the
952
1 LiViD mailing list.
2 Q. Did you try it out when it was first posted?
3 A. When it was first posted, it was posted in binary form.
4 We were more -- I was more interested in the source code
5 version which came shortly thereafter. When it first came
6 out, I did not use it.
7 Q. Was DeCSS part of or connected to the LiViD project?
8 MS. MILLER: Objection, your Honor, no foundation.
9 THE COURT: Overruled.
10 A. Yes, the DeCSS has actually a long history of being
11 related to the LiViD project. The CSS project or CSS process
12 has a few phases, the authentication between a decoder or the
13 piece whether it be hardware or software that takes the DVD
14 data and converts it here in audio and video presentation and
15 the actual decryption where it decrypts the encrypted content.
16 The first part of that process was the authentication
17 and that was written and released for and under the LiViD
18 project. DeCSS utilized the CSS routines from the LiViD
19 project as a piece of DeCSS. DeCSS, the source code was later
20 translated, the core functions were used in the decrypting
21 part of the DeCSS for the Linux video player.
22 Q. In other words, you are saying part of the DeCSS came from
23 Linux and the second part of DeCSS went back into Linux?
24 A. That is correct.
25 Q. Is the source code for DeCSS available now?
953
1 A. Yes.
2 Q. Is the source code for LiViD project available?
3 A. Yes, all the code for the LiViD project is available. We
4 released all the under the GNU general public license which is
5 the open source license that most Linux projects are using.
6 Q. Why was DeCSS a Windows program?
7 MS. MILLER: Objection, your Honor.
8 THE COURT: Sustained.
9 Q. Do you know why DeCSS was a Windows project?
10 MS. MILLER: Objection, we have already heard
11 testimony from Mr. Johansen.
12 THE COURT: Sustained.
13 MR. HERNSTADT: OK.
14 THE COURT: He didn't write the code.
15 MR. HERNSTADT: I will try it an a different way,
16 your Honor.
17 Q. Was DeCSS used by the project in developing the LiViD
18 player?
19 A. Yes. Many of the other developers that had -- I didn't
20 have a windows system at that time and many of the other
21 developers used DeCSS to test the tool, to decrypt DVD data to
22 use the unencrypted data for the other parts of the DVD
23 player. A DVD player would have ten steps to decode a DVD
24 fully and CSS represents two of those and we were unable to
25 develop many of the other pieces without having those two
954
1 steps. So many of the developers would use the unencrypted
2 content to start analyzing the other parts of it and many
3 developers analyzed the algorithm which was then brought back
4 into the LiViD player.
5 MS. MILLER: Objection, move to strike the last
6 answer. This witness has not laid a foundation for how he
7 knows what the other developers in the Linux project have.
8 THE COURT: Granted.
9 Q. You were the project leader of the LiViD project?
10 A. That's correct.
11 Q. As project leader, you observed all the different changes
12 of major source code as they were made, is that correct?
13 A. I observed most of them. It is difficult to see every
14 piece of code that goes in, but when a developer posts a
15 change to the LiViD project or LiViD archive, we set up a tool
16 to where the developer would commit his changes and it would
17 list it and then e-mail back to the mailing list, without the
18 developer having to do anything, the list of the files that
19 were changed and give an area for a note to what the developer
20 said he had just changed in that latest patch or update.
21 So it was very easy to track the developments by
22 following the mailed postings. They listed the files that
23 were changed and what was changed in those files.
24 Q. You said you followed the postings to the forum as well?
25 A. Sure, and the code archive itself.
955
1 Q. Was it your responsibility to know what was going on in
2 the LiViD project as its leader?
3 A. It was my responsibility in a sense that, you know, I felt
4 like I founded this project, I started it, this is something I
5 wanted to do. I had no obligation, no one was going to punish
6 me if I didn't follow everything but I had an interest and I
7 wanted to see a project that I started get completed so I
8 actively followed the developments of the project.
9 Q. Did other people working on the project report to you what
10 they were doing?
11 A. Yes. We would release status reports on occasion and I
12 would send out notes to other developers saying give me an
13 update on, you know, what piece you are working on, where you
14 see this going, if you need help in a certain area so we could
15 give the general community an idea of what was going on in a
16 project for all the people that weren't following the project
17 directly on the mailing list.
18 MR. HERNSTADT: I would suggest that that would
19 provide the foundation.
20 THE COURT: It shows that he knows a lot of hearsay.
21 I am very interested in what's going on in Camp David and I
22 read the paper every morning. I am not capable of testifying
23 to what's going on at Camp David in a court of law.
24 MR. HERNSTADT: I understand. He is the leader of a
25 project, they all report to him about what they do.
956
1 I will try it a different way.
2 Q. Mr. Pavlovich, given what you know about the LiViD
3 project, can you tell me if it would be necessary if you
4 wanted to, if you were to write a DeCSS program to do it in
5 Windows?
6 A. I am sorry. I didn't understand the question.
7 Q. Given what you know about the LiViD project and Linux in
8 or about the fall of 1999, if you are going to write a DeCSS
9 program, would you have to do it in Windows?
10 MS. MILLER: Objection. I don't understand the
11 question. It is completely ambiguous.
12 THE COURT: I think the witness and I do, Ms. Miller.
13 Overruled.
14 We are now going to hear about the file formats for
15 the third or fourth time.
16 A. The file system found on DVDs is the UDF support for Linux
17 was in infancy at the time, so one would need to have access
18 to read the data before being able to decrypt the data on the
19 disk, so yes someone would have to use windows or an operating
20 system that supported UDF to develop DeCSS.
21 Q. Did you consider testing the LiViD project using
22 unencrypted disks?
23 A. Yes, some of the developers found, you know, through the
24 course of their testing of the disks, they found a few
25 unencrypted DVDs. The main issue of that is there is no
957
1 labeling on a DVD when you buy it whether it is encrypted or
2 unencrypted. Through my research, I found only a dozen or two
3 dozen DVDs that are known to be unencrypted.
4 The other issue, some of the DVDs were found early on
5 to be unencrypted and either in later pressings or later
6 releases were then encrypted.
7 Q. Is there any other reason you didn't use unencrypted disks
8 to test the Linux project?
9 A. Most DVDs are encrypted and our DVD player wanted to be
10 full featured. We wanted to be able to play all the DVDs on
11 the market. The other thing here that we really were
12 developing for was our own personal use but also all the other
13 users for Linux and limiting the DVD player to only un
14 encrypted DVDs would have severely hampered the usability of
15 the Linux DVD player.
16 Q. Is the LiViD player project completed? Is there a player
17 that works now?
18 A. In our initial design, we have kind of had two phases of
19 the project. When we first started getting all the necessary
20 documentation and information needed to develop the LiViD
21 player, we did a proof of concept design where we put a bunch
22 of hodge-podge code together and we were able to test whether
23 everything was going to be technologically feasible.
24 Over last six or seven months, we designed a new
25 architecture which is going to allow us to have improved
958
1 playback performance, hardware acceleration, support for
2 multiple hardware decoder part boards and video cords and
3 currently the DVD player for Linux can play DVDs, decode the
4 audio. We have not had a formal release, a version 101 so to
5 speak, however the program is in a working state.
6 MR. HERNSTADT: I would like to have the witness give
7 a demonstration of the LiViD program here. It is about a
8 three or four-minute process to show you how it works and what
9 it is.
10 MS. MILLER: Plaintiffs object to that demonstration
11 on the grounds of relevance. Whether or not there is an
12 unlicensed LiViD DVD player right now has no significance, any
13 relevance to the issues in this lawsuit.
14 THE COURT: I take it your position is you don't
15 dispute that there is a working unlicensed DVD player, not
16 that you are conceding it, but you don't care to dispute it
17 for the case of the trial.
18 MS. MILLER: That's correct.
19 THE COURT: I don't think we need it then, Mr.
20 Hernstadt.
21 MR. HERNSTADT: If it is not contested, then --
22 THE COURT: It is not contested.
23 (Continued on next page)
24
25
959
1 BY MR. HERNSTADT:
2 Q. Does the LiViD player have a save function that permits
3 you to copy the movie on the DVD to some place on the hard
4 drive?
5 A. No. The design of the LiViD DVD player was never geared
6 around copying. Our main goal was to get compatibility and
7 performance of a DVD player so we could support it on as many
8 systems with as wide a range of hardware as any user would use
9 as possible, and copying was never a main focus of the
10 project.
11 Q. Have you ever used DeCSS to decrypt a DVD?
12 A. Yes.
13 Q. When did you do that?
14 A. During the course of this trial I used it on Windows to
15 test, run some tests, test the application, get a feel for it.
16 Q. What did you do with the files that you made when you used
17 the DeCSS program?
18 A. The Windows system I used has a very small space allocated
19 for Windows, only 1 gigabyte, so I'm not able to decrypt the
20 entire disk. I decrypted one small piece of it, just tested
21 it, to see that the application worked, deleted it when I was
22 done.
23 Q. Have you reviewed the source code of DeCSS?
24 A. Yes.
25 Q. Is that source code in the Linux DVD player now?
960
1 A. The derivatives of the DeCSS are. The core functions in
2 DeCSS, the mathematical algorithms and the process for
3 decrypting were converted from DeCSS back into the LiViD DVD
4 player.
5 Q. Do you know what DivX is?
6 A. Yes.
7 Q. What is it?
8 A. DivX is in the term I have seen it used in these court
9 proceedings for an MPEG4-based video format.
10 Q. Have you ever used DivX?
11 A. No.
12 Q. Have you ever used any program other than DeCSS and the
13 one instance you mentioned to decrypt a DVD movie?
14 A. I've got a DVD player on my Windows side, my Windows
15 installation. That's the only other one.
16 Q. But have you ever used a program like DVD Rip?
17 A. No.
18 Q. Or Speed Ripper?
19 A. No.
20 Q. Or any other "ripper"?
21 A. No.
22 Q. Do you own DVDs?
23 A. Yes.
24 Q. How many?
25 A. I've got 20, 25 DVDs.
961
1 Q. Did you buy them?
2 A. Yes.
3 Q. Have you ever played a DVD that you didn't buy?
4 A. Only the ones I rented.
5 MS. MILLER: Objection. Relevance.
6 THE COURT: Overruled.
7 Q. Have you ever made a decrypted DVD movie available on the
8 Internet?
9 A. No.
10 Q. Have you ever made a copy of a decrypted DVD on any
11 format, VCD, CD, DVD, and made it available to anyone else?
12 A. No.
13 Q. Or sold it to anyone else?
14 A. No.
15 MR. HERNSTADT: Thank you very much, Mr. Pavlovich.
16 THE COURT: All right. We will take our afternoon
17 break, 15 minutes.
18 (Recess)
19 THE COURT: Okay, Ms. Miller, you may proceed.
20 CROSS-EXAMINATION
21 BY MS. MILLER:
22 Q. Good afternoon, Mr. Pavlovich.
23 A. Good afternoon, Ms. Miller.
24 Q. I'm correct that you didn't have anything to do with the
25 creation of DeCSS, right?
962
1 A. Could you please repeat the question.
2 Q. I'm correct that you didn't have anything to do with the
3 creation of DeCSS, that's right, isn't it?
4 A. I did not develop any of the code. However, I started the
5 project for Linux in which the algorithms in DeCSS were part
6 of it.
7 Q. Now, you talked about performing a test of DeCSS for the
8 first time after getting involved in this lawsuit, is that
9 correct?
10 A. That is correct.
11 Q. The machine that you used to perform your test of DeCSS,
12 you told me in your deposition that machine actually has a 10
13 gigabyte hard drive, is that correct?
14 A. That's correct.
15 Q. That only 1 gigabyte was reserved for Windows to be able
16 to use DeCSS?
17 A. That's correct.
18 Q. You testified that the css-auth program was originally
19 developed as part of the Linux development project for a DVD
20 player, is that correct?
21 A. Yes.
22 Q. At what point in time was that program developed, if you
23 know?
24 A. That was in the summer of '99.
25 Q. Had you actually reviewed the source code for the css-auth
963
1 program?
2 A. Yes.
3 Q. What functions does the css-auth program perform?
4 A. Css-auth does what they call handshaking between the DVD
5 drive and the decoder on the system, and it passes the keys
6 that are on a DVD disk and the key that is present in the
7 player, and it does authentication and matching of those keys
8 to see if you have a viable match.
9 Q. And the css-auth program that you're referring to is
10 written for the Linux operating system, is that correct?
11 A. Yes, yes.
12 Q. Now, am I correct that no one on behalf of the LiViD
13 project sought a license from the DVD-CCA for CSS?
14 A. Can you please repeat the question.
15 Q. No one from the LiViD project ever sought a CSS license
16 from the DVD-CCA, is that correct?
17 A. There are contributors to the project that were familiar
18 with the licensing. The project as a collective never sought
19 out a license. There were individuals that were interested in
20 the development that had understanding of the license, other
21 people working on other projects.
22 Q. So the answer to my question is no one associated with the
23 LiViD project ever sought a license from the DVD-CCA?
24 A. The answer is "sort of," because when you say "a member of
25 the LiViD project," it's not like we have a list of people who
964
1 are in the group. It's an open development process, and there
2 were other people what might say they are part of LiViD, one
3 might say they are not, that did go seek out the license.
4 Q. But you testified in your deposition, did you not, that
5 the LiViD project basically resolved that they wouldn't seek a
6 DVD-CCA license because of the nondisclosure requirement,
7 isn't that correct?
8 THE COURT: Mr. Hernstadt?
9 MR. HERNSTADT: Could we have a page and line number?
10 THE COURT: Ms. Miller, I did issue an order on this.
11 Please just indicate the page and line and just read the
12 testimony. Then if you want to ask the witness if the
13 testimony was true, you can do that.
14 (Pause)
15 MS. MILLER: Sorry, your Honor. I just need one more
16 moment to find the testimony. I'm almost there.
17 THE COURT: Fine.
18 THE WITNESS: Could I have a copy to review when you
19 discuss, so I can get the context?
20 MS. MILLER: Certainly.
21 THE COURT: Do you want to perhaps pass to another
22 subject and have one of your colleagues look through that?
23 Mr. Sims will find it.
24 Q. Now, you have heard of a company called Sigma Designs, is
25 that correct, Mr. Pavlovich?
965
1 A. Yes, I have. I worked with one of their -- I believe he
2 is a marketing manager at Sigma Designs.
3 Q. And can you explain for the Court what Sigma Designs does?
4 A. Sigma Designs develops hardware decoder chips for doing
5 particularly DVD playback. They released a number of cards
6 into the market to support that.
7 Q. Does Sigma Designs also have a DVD player on the market?
8 A. They have an application that works with their hardware
9 boards.
10 Q. And does that application run under Microsoft Windows?
11 A. Yes.
12 Q. Now, you are aware, are you not, that Sigma Designs is a
13 CSS licensee, is that correct?
14 A. I have not reviewed their documentation. I don't know
15 firsthand if they are a licensee. It would lead me to believe
16 that they are, due to the fact that they released equipment
17 onto the market.
18 Q. Now, are you also aware that Sigma Designs recently
19 announced plans to develop a Linux DVD player? Is that
20 correct?
21 A. Yes, they are releasing plans to develop a DVD player to
22 support their next generation adapter. In fact I'm on the
23 list of beta developers for that card.
24 Q. Now, did anyone from Sigma Designs ever post anything to
25 the LiViD mailing list?
966
1 A. Marshall Goldberg. He was the gentleman who was the
2 marketing manager that I worked with.
3 Q. Now, it's correct, isn't it, that Mr. Goldberg posted a
4 message to the LiViD mailing list expressing concern about the
5 legality of LiViD's development activities in creating an
6 unlicensed DVD player, correct?
7 MR. HERNSTADT: Objection. Hearsay.
8 THE COURT: I take it it's not offered for the truth.
9 MS. MILLER: That's correct, your Honor.
10 THE WITNESS: Could you please repeat the question.
11 Q. It's correct, isn't it, that Mr. Goldberg expressed some
12 concern on the LiViD mailing list about the LiViD team's
13 development of an unlicensed DVD player for Linux, isn't that
14 true?
15 A. I believe he did. I would have to review the archives to
16 be sure.
17 Q. Do you recall testifying in your deposition about
18 Mr. Goldberg's mailing list posting?
19 A. Yes, I believe you showed me one of the postings, and upon
20 reading it I recognized the message.
21 Q. And isn't it also true that Sigma Designs declined to
22 release to LiViD certain of their specifications for their
23 interface cards because of their nondisclosure requirements
24 under the CCA's license?
25 A. I don't know the disclosure requirements under the CSS
967
1 license. However, Marshall expressed to me that they could
2 not release any specification on their most popular board,
3 which was the Sigma Designs known as the H Plus Board,
4 however, we did learn that that board did not do any CSS
5 routines that we knew of, and that board was reverse
6 engineered, or the driver for that was reverse engineered, and
7 support for that board is now going to be available in the
8 LiViD player.
9 Q. You also know the name Paul Bocco, is that correct?
10 A. That's correct.
11 Q. Paul Bocco is one of the leaders of another team referred
12 to as LS DVD, is that not correct?
13 A. That is correct. LS DVD has also worked with LiViD. We
14 have been working to try to create a cohesive set of tools so
15 that our code sets could work together as best possible.
16 Q. In fact the LS DVD project members and LiViD project
17 members shared information about their ongoing development
18 efforts for Linux DVD players, correct?
19 A. Yes, that's correct.
20 Q. Now Mr. Bocco also expressed some concern on the LiViD
21 mailing list about creating an unlicensed DVD player, isn't
22 that correct?
23 A. That is correct.
24 Q. And as far as you know, is the LiViD -- excuse me -- is
25 the LS DVD team now seeking licensing from the DVD-CCA?
968
1 A. I believe they are. Paul sent an update to the list,
2 mentioning that they had found a company to help fund the
3 necessary -- or to raise enough money to buy the licensing and
4 what not.
5 However, I think it's important to note that we have
6 seen no code from either of these companies yet, and that
7 their announcements have been met with skepticism. The common
8 term used is vaporware, meaning somebody announces they are
9 going to do something in the future, we are going to release
10 widget A or application B, and a lot of times they never
11 materialize. As of yet both of those people have made
12 announcements they are going to do it, however, we have not
13 seen any fruits of their labor.
14 Q. Now, you testified earlier this afternoon that you have
15 never used any rippers like DVD Rip or Power Rip and the Power
16 Ripper, is that correct?
17 A. That is correct.
18 Q. You submitted a declaration in this lawsuit, is that
19 correct?
20 A. I submitted two.
21 Q. Okay. And do you recall the date of the first declaration
22 that you submitted in this lawsuit?
23 A. I believe it was the end of April, beginning of May.
24 MS. MILLER: Your Honor, may I approach the witness?
25 THE COURT: Yes.
969
1 Q. Mr. Pavlovich --
2 A. Could you just give me one second to just take a quick
3 gander here.
4 THE COURT: You have given him Defendants' BBO, as I
5 understand it, is that right?
6 MS. MILLER: I actually thought it had been
7 designated as Defendants' S.
8 MR. HERNSTADT: Defendants' S is what I have.
9 THE COURT: S?
10 THE WITNESS: I have BBO at the bottom.
11 THE COURT: Mine and the witness's are BBO, and BBO
12 it is.
13 MR. HERNSTADT: May I inquire, is that the
14 declaration dated April 30?
15 THE COURT: Yes.
16 MR. HERNSTADT: Thank you.
17 Q. Mr. Pavlovich, have you reviewed the declaration?
18 A. Yes.
19 Q. Do you recognize this as the first declaration you filed
20 in this case?
21 A. Yes.
22 Q. I would like to direct your attention to page 6, paragraph
23 13 of that declaration.
24 A. Okay.
25 Q. Paragraph 13 you state, "At the same time, other tools for
970
1 copying movies from DVDs such as DVD Rip (a program that
2 intercepts the decrypted DVD data stream and creates a freely
3 copiable data file) and Power Ripper are widely available and
4 were for months prior to the release of DeCSS. Additionally,
5 they are the tools used when digital-to-digital (for instance
6 DVD to VCD) copies are made from DVDs, because they are much
7 easier to use than DeCSS."
8 Have I just accurately read paragraph 13 of your
9 declaration?
10 THE COURT: You don't have to ask that question.
11 It's right in front of us. Next question, please.
12 A. Yes.
13 Q. Did you write this paragraph of your declaration?
14 A. Yes.
15 Q. But you had actually never used either one of these
16 rippers --
17 A. No.
18 Q. -- prior to writing this paragraph?
19 THE COURT: In other words, it's correct that you
20 have never used any of these other rippers?
21 THE WITNESS: That is correct.
22 Q. You also indicated in your deposition, did you not, that
23 your only knowledge about Power Ripper is that it has
24 something to do with DVDs and that it's a Windows program.
25 A. Yes.
971
1 MR. HERNSTADT: Your Honor, could we have a page and
2 line?
3 MS. MILLER: Certainly. That would be page 99 of the
4 deposition, line 11 through line 25.
5 THE WITNESS: Could I get a copy of the deposition?
6 I have not received it.
7 THE COURT: It's not necessary right now,
8 Mr. Pavlovich. Let's please move on with this.
9 THE WITNESS: I apologize.
10 THE COURT: I wasn't being impatient with you,
11 Mr. Pavlovich. I issued an order in this case two weeks ago
12 explaining exactly how to do this, and I have reiterated it
13 several times during the trial.
14 Q. Now, Mr. Pavlovich, you also have told me in your
15 deposition, and I am reading now from page 100, line 14
16 through 15, and I asked you:
17 "Q. What about DVD Rip?" And your answer:
18 "A. I have never heard of DVD Rip."
19 Was that an accurate statement in your deposition?
20 A. I believe if that's how you are reading it from the paper,
21 yes.
22 THE COURT: That wasn't the question. Was the
23 testimony you gave in the deposition truthful, the portion
24 that was just read to you?
25 THE WITNESS: No, I mean I have heard of DVD Rip.
972
1 Q. And at what point in time did you hear of DVD Rip? Was it
2 prior to your declaration?
3 A. I don't have a date when I remember hearing about DVD Rip.
4 You know, through the course of developing the DVD player and
5 researching it.
6 THE COURT: Had you heard of it before your
7 deposition was taken?
8 THE WITNESS: Yes.
9 THE COURT: So when you testified in your deposition
10 you never heard of it, you testified falsely. Is that about
11 the size of it?
12 THE WITNESS: Yes.
13 THE COURT: Is there some particular reason why you
14 did that?
15 THE WITNESS: Not that I can think of now.
16 THE COURT: Let's go on.
17 MS. MILLER: No further questions, your Honor.
18 THE COURT: Thank you. Mr. Hernstadt?
19 REDIRECT EXAMINATION
20 BY MR. HERNSTADT:
21 Q. Mr. Pavlovich, do you remember being asked about DVD Rip
22 at your deposition?
23 A. Not particularly.
24 Q. Were you trying to hide something in your testimony when
25 you said you had never heard of it?
973
1 A. No. There is a lot of utilities out there available along
2 this genre, and I think it's difficult to keep track of the
3 exact names of everything.
4 When you know of an application and you wish to
5 research it and what it does, it's very easy to go and make a
6 search of it and find that application and find the specific
7 name of it. There are literally dozens of applications along
8 these lines, and I made a mistake.
9 Q. Mr. Pavlovich, where is the information about the LiViD
10 project stored?
11 A. On the LiViD website. It has the links to all the
12 information for the project.
13 Q. Is that information available to anyone?
14 A. Yes, it's publicly available.
15 Q. You were asked why you didn't go to the DVD-CCA for a
16 license. Why didn't you?
17 A. There was a long discussion on the LiViD development list
18 on that topic, and many people are expressing several
19 different opinions.
20 One of the major deciding factors was the license
21 that we were using for developing a Linux DVD player was the
22 GNU public license, and it would provide an incompatibility
23 with what we knew of the CSS license. And there is also the
24 issue of raising the funds and coordinating who would handle
25 the actual facilitation of the licensing and be bound to the
974
1 contract, etc.
2 Q. When you say "incompatibility," can you explain what you
3 mean by that?
4 A. The GNU public license is a license placed on the software
5 and it's a legal incompatibility. GNU public license was
6 designed to protect code that was released under that license
7 so people couldn't take it, make a modification and sell it
8 without providing the source which they changed. That's in a
9 nutshell what the core function of the license is. And what
10 we knew of the CSS license, we would not be able to release
11 the source code for the CSS in the DVD player.
12 Q. In writing software for Linux applications is reverse
13 engineering a tool that you use?
14 A. Reverse engineering is I think better defined as a method
15 or a process. It's kind of a general term. Yeah, reverse
16 engineering is used all the time under Linux development
17 projects. Hardware manufacturers often don't release all the
18 necessary information on products, whether it be an interface
19 for talking to graphics hardware, file sharing technology.
20 For example, one project that I followed very closely
21 over its development is the SAMBA project which incorporates
22 file sharing for UNIX systems in between Microsoft Windows
23 systems. Microsoft provided information on the file
24 technology, however they did not release all the information
25 needed. They made a few changes to the protocol so it would
975
1 be impossible to develop a compatible program to talk to
2 Microsoft systems on a non-Microsoft system. So the SAMBA
3 team had to reverse engineer the SMB protocol across the wire
4 to effectively communicate with Microsoft systems.
5 Q. Thank you. You were asked if you had done a declaration.
6 You said you had done two declarations.
7 MR. HERNSTADT: May I approach, your Honor?
8 THE COURT: Yes.
9 Q. Mr. Pavlovich, is that the second declaration you did in
10 this case?
11 A. Yes.
12 MR. HERNSTADT: I move the two declarations into
13 evidence.
14 MS. MILLER: Your Honor, objection to moving the
15 first declaration into evidence. As I read the declaration,
16 there are several statements in there that would purport to
17 give opinion testimony about Linux and other things having to
18 do with reverse engineering and system development. I don't
19 think that Mr. Pavlovich has been properly qualified as an
20 expert under Rule 702, so we would object on that basis.
21 Also, there seems to be some discrepancy particularly
22 as to paragraph 13 of the first declaration, whether in fact
23 that's true or if he has personal knowledge of the facts that
24 he related in that paragraph, given that he testified today
25 that he has never used any rippers.
976
1 THE COURT: And the second declaration?
2 MS. MILLER: The second declaration, in so far as
3 Mr. Pavlovich is explaining what DeCSS does, I am going to
4 object on the ground that we have already heard testimony from
5 Jon Johansen about what DeCSS does on that basis. That would
6 be cumulative. I also understand that he is responding in the
7 second declaration to some of the information in
8 Mr. Schumann's declaration, which is at issue with some of the
9 aspects of Mr. Pavlovich's first declaration. So on that
10 basis we have no problem, but certainly as to paragraph 5, the
11 second declaration, purporting to say what Mr. Johansen said
12 about the purpose of DeCSS, we would move to strike that or at
13 least not have it offered into evidence.
14 THE COURT: Look, Mr. Hernstadt, I have a problem
15 with this. You had the witness here on direct, you could have
16 asked him anything you wanted to ask him, and to come in on
17 redirect with two declarations, narrative material, part of
18 which is incompetent, part of which is hearsay, part of which
19 is beyond the scope of the cross, you are opening the witness
20 all over again.
21 MR. HERNSTADT: I will withdraw the question, your
22 Honor.
23 THE COURT: All right.
24 MR. HERNSTADT: Thank you very much, Mr. Pavlovich.
25 Nothing further.
977
1 THE COURT: Anything else for Mr. Pavlovich,
2 Ms. Miller?
3 MS. MILLER: Nothing, your Honor.
4 THE COURT: Thank you, Mr. Pavlovich. You are
5 excused.
6 All right. In view of the hour, let's get to the
7 housekeeping stuff you people wanted to raise, because I have
8 another matter at 4:30.
9 MR. ATLAS: I don't think we need to do it on the
10 record unless your Honor wishes.
11 THE COURT: I think in this case I want to do
12 everything on the record.
13 MR. ATLAS: Fair enough. Should we come up or do it
14 from here?
15 THE COURT: You can do it from there.
16 MR. ATLAS: Really it's just an issue of how your
17 Honor wants the deposition designations and some of the
18 documents.
19 My thought, and I have talked about it with Mr. Sims,
20 would be to move into evidence the depositions. What we have
21 done is we have marked the actual transcripts, we have
22 bracketed them, and I believe at the front of each deposition
23 is sort of a designation page that we have designated these
24 pages, they have objected, or they have counterdesignated and
25 we have countered the counters in some instances.
978
1 THE COURT: I am almost 100 percent sure about two or
2 three weeks ago I issued an order of a page or two explaining
3 exactly how this was to be done. Did that somehow escape
4 everybody's attention?
5 MR. SIMS: I thought Mr. Atlas wasn't clear.
6 THE COURT: I want to know what the question is
7 exactly.
8 MR. ATLAS: I assume we will move those kings into
9 evidence, and the Court as it goes through the evidence in
10 this case will decide on the objections as it is necessary.
11 THE COURT: Sure.
12 MR. ATLAS: The second thing is with respect to the
13 documents, Mr. Sims has provided me with a list of the
14 exhibits he wishes to move wholesale into evidence. I am
15 going through that now. I think a lot of them I don't have a
16 problem with, some of them I do and I am noting those
17 objectives. I am going to do the same thing.
18 Now, we will do the same process, move all those
19 documents into evidence, and the Court will as its going
20 through the material decide whether my objections or
21 Mr. Sims's objections are valid and decide whether the
22 documents stay in evidence.
23 THE COURT: I am not sure how we are going to do the
24 documents. It is going to depend on volume.
25 MR. ATLAS: My goal this weekend is to significantly
979
1 narrow the volume.
2 THE COURT: Well, I would certainly hope you both do
3 that. Plaintiffs identified a hundred and some.
4 MR. SIMS: We have provided a much smaller list, less
5 than a third.
6 THE COURT: And you provided well over a thousand.
7 MR. ATLAS: I promise you it will be less than a
8 third.
9 THE COURT: Yes, I can't even begin to imagine that
10 all of that stuff is necessary.
11 MR. ATLAS: That's what we are going to do. Assuming
12 you want the process to be handled the same way, which is we
13 give you everything and you rule on the objections, the only
14 wrinkle to that is the documents marked privileged and
15 confidential, attorneys eyes only. Do you want to handle
16 those any differently?
17 THE COURT: How many of those are there going to be?
18 MR. ATLAS: I don't know. I'm going to go through
19 that this weekend.
20 THE COURT: "It depends" is the short answer.
21 MR. ATLAS: I anticipate coming on Tuesday with all
22 this stuff ready to go and at that point we can decide
23 whatever we want to do and we will provide it.
24 THE COURT: What is left on defendants' case?
25 MR. ATLAS: It looks like we have four, possibly five
980
1 witnesses left. My best guess is it would be hard to imagine
2 we would go past Wednesday on this. Again that's just an
3 estimate of what I know from these witnesses and how long the
4 direct is. Obviously it would be subject to whatever cross.
5 But we have been moving at a pretty fast clip unless it slows
6 down significantly.
7 THE COURT: Who are the remaining witnesses?
8 MR. ATLAS: I will give you the possible choices.
9 THE COURT: This sounds look a trade for a pitcher.
10 MR. ATLAS: Mr. Einhorn, Mr. Touretzky, Mr. Appel,
11 Mr. DiBona and Mr. Abelson are sort of the potentials, and
12 over the weekend we are going to work out exactly who we feel
13 we need, and we will advise plaintiffs' counsel as soon as we
14 know.
15 THE COURT: I assume -- Mr. Cooper?
16 MR. COOPER: I just want to know for our own
17 preparation over the weekend that there is not anyone else
18 they are contemplating so that we won't fail over the weekend
19 to prepare for somebody and then find on Tuesday that they are
20 being offered.
21 MR. ATLAS: Just give me a second.
22 THE COURT: Look, I am taking that until I hear some
23 spectacularly wonderful reason nobody other than these five
24 are going to get on the stand.
25 MR. ATLAS: Possibly Olegario Craig. Definitely him.
981
1 THE COURT: I assume the plaintiffs have some clue as
2 to what these people are all about, right?
3 MR. SIMS: These people have all been deposed.
4 THE COURT: Okay. So, if, as seems entirely likely,
5 this is the rest of the defendants' case, is there likely to
6 be a rebuttal case?
7 MR. GOLD: Your Honor, we have been discussing that
8 for several hours last night, and today's testimony threw some
9 question on it, so if there is a rebuttal witness it will be
10 short. He personally is fairly tall, but he will only testify
11 for less than an hour.
12 THE COURT: I am glad we are beginning to get some
13 humor into this case.
14 MR. ATLAS: Is there a name, or you don't know who it
15 would be?
16 THE COURT: Whoever this is is not testifying before
17 Wednesday, right?
18 MR. GOLD: Obviously.
19 MR. ATLAS: It sounds like it may be somebody who has
20 not surfaced in the case. If that's the case, then I would
21 request an opportunity to depose them beforehand.
22 MR. GOLD: Well then that's reasonable, and we will
23 tell you and you can depose this person on Monday or Monday
24 night.
25 THE COURT: All right. That sounds cooperative and
982
1 reasonable. All right. Anything else?
2 MR. GOLD: No, your Honor.
3 THE COURT: Okay. Thank you folks.
4 (Trial adjourned to July 25, 2000 at 9:00 a.m.)
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
983
1
2
3 INDEX OF EXAMINATION
4 Witness D X RD RX
5 LARRY PETERSON...........856 882 883
6 PETER RAMADAGE...........884 932
7 MATTHEW PAVLOVICH........940 962 973
8 DEFENDANT EXHIBITS
9 Exhibit No. Received
10 BDC .........................................862
11 BBD .........................................863
12 BDE .........................................889
13 BDF and BDG .................................929
14
15
16
17
18
19
20
21
22
23
24
25