16 June 2000. Thanks to Anonymous.
See related files:
http://www.eff.org/pub/Intellectual_property/DVD/
http://cyber.law.harvard.edu/openlaw/dvd/
http://www.2600.com
http://jya.com/cryptout.htm#DVD-DeCSS
Contents
Defendants' Reply Brief in Support of the Motion to Vacate the Injunction
Reply Declaration of Frank Stevenson
Declaration of Edward Hernstadt
Reply Declaration of Chris Moseng
Reply Declaration of Olegario Craig
Exhibit 1 to Reply Declaration of Olegario Craig
Reply Declaration of Matthew Pavlovich
Reply Declaration of Emmanuel Goldstein
| UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK -----------------------------------------------------------------------X
UNIVERSAL CITY STUDIOS, INC., - against -
ERIC CORLEY a/k/a "EMMANUEL GOLDSTEIN"
|
00 Civ. 0277 (LAK) |
DEFENDANTS' REPLY BRIEF IN SUPPORT OF THE
MOTION TO VACATE THE INJUNCTION
Plaintiffs' opposition to the motion to vacate the preliminary injunction is based on groundless speculation about possible future harms rather than a demonstration of the elements required to obtain a preliminary injunction: actual irreparable harm and of a likelihood of success on the merits. Plaintiffs concede two crucial facts that thoroughly undermine their claim of irreparable harm: (1) despite months of investigation and the continued presence and visibility of DeCSS on the Internet, plaintiffs have not discovered a single instance of piracy or copying of related to DeCSS; and (2) the technology required to upload and move on the Internet the massive files containing decrypted DVD movies does not now exist and will not exist for at least the next several years.
Plaintiffs' attempt to justify a claim under Section 1201 of the Copyright Act by treating it like an infringement claim for damages and contributory infringement purposes, but not for the fair use doctrine, is unconstitutional. Congress did not intend the unconstitutional results that flow from plaintiffs' attempt to separate 1201(a)(2) from the rest of Section 1201.
DeCSS is not a commercially viable utility for piracy; its only "commercially significant" (or other) purposes are legal. DeCSS permits the public to access the copyrighted materials on a DVD in order to make fair use of them. It enables users to play legally purchased DVDs on any operating system and with virtually any DVD player, regardless of whether it is equipped with CSS, as part of reverse engineering a software DVD player for general distribution or a personal DVD player to view legally purchased movies. Plaintiffs ask this Court to bar providing DeCSS to anyone, ever, despite the fact that 1201 cannot constitutionally operate to silence harmless, expressive speech. This interpretation of 1201 would prohibit any and all access to (as opposed to viewing of) the copyrighted material on a DVD, and thereby obliterate fair use and the public's other well-established rights to non-infringing uses of copyrighted materials on DVDs.(1)
____________________
1. Under this position, an open-source DVD player that does not utilize CSS, such as like the Linux LiVid DVD player here at issue, would be every bit as illegal as DeCSS itself.
Defendants' cross motion is based upon facts put forth after the January 20, 2000 preliminary injunction hearing, including admissions by the plaintiffs and the MPAA. These new facts include plaintiffs' acknowledgment that DeCSS has caused no damage, and plaintiffs' resulting new-found claim of irreparable harm based on the (false) danger posed by the DivX-compressed DVD movies on the Internet.(2)
____________________
2. See Huk-a-Poo Sportswear Inc. v. Little Lisa, Ltd., 74 F.R.D. 621 (S.D.N.Y. 1977). Plaintiffs' conduct, and the relief sought in their complaint, also violates anti-trust law and the First Sale doctrine, as well as 1201(c)(3)'s prohibition against encryption technology that, like CSS in home DVD players, requires a particular design. Plaintiffs believe that the licensing regime plaintiffs and the DVD CCA constructed permits them to dictate that members of the public who purchase their movies on DVDs (but not on VHS or VCD) must also purchase an approved, CSS-licensed DVD player on which to view that movie. Indeed, Fritz Attaway, the MPAA's Washington DC General Counsel and major Congressional Lobbyist, testified that "[i]n order to obtain authorized access to a DVD, the consumer has to in effect make two purchases. He or she has to buy . . . a DVD disk . . . and also has to purchase a DVD display device." Hernstadt Declaration, Exh. E (Attaway Depo., 30:1-11).
None of plaintiffs' references to the legislative history of Chapter 12 supports the contention for which it is cited. The legislative history clearly establishes that in making and passing Chapter 12, Congress was concerned with creating a law that would control piracy/copying while preserving rights such as fair use. Congress ensured that 1201 would not ban devices that employed circumvention as an incident to lawful use or access. See 144 Cong. Rec. at H7097 (daily ed. Aug. 4, 1998) (letter from Rep. Coble (chairman of the House Subcommittee on Courts and Intellectual Property) to Reps. Boucher and Campbell) (emphasizing that 1201's anti-circumvention were considerably narrower than those in the previously rejected "NII Copyright Protection Act of 1995": "under [ 12010], it is not enough for the primary effect of the device to be circumvention"). Even Jack Valenti was clear that "'[f]air use remains unchanged. Legitimate devices are unaffected." S. Rep. No. 105-25 (1998). Congress did not pass a law that imposes a blanket prohibition against the public's having any access whatsoever to the copyrighted material on a DVD.
Plaintiffs have failed to point to any instance of infringement related to DeCSS. Where there is no possibility of wrongful copying(3) in the past, present or imminent future, there can be no presumption of irreparable harm. Plaintiffs cannot as a matter of constitutional law and statutory interpretation argue that they must be given the presumption of irreparable harm recognized in copyright infringement cases (even though they make no claim of infringement), and that defendants cannot interpose fair use as a right, limitation or defense because plaintiffs do not assert copyright infringement in their complaint.(4)
____________________
3. In this regard, plaintiffs' contention that there is "copying" because DeCSS "copies" the DVD files as it decrypts them is deceptive. First, if all DeCSS did was copy files to a hard drive where the files sat and moldered, this lawsuit would not have been filed. Rather, plaintiffs direct their case (and claim of irreparable harm) to what then is done with the "copied" files, and the potential danger of piracy. Plaintiffs do not contend that the provision of DeCSS is barred by 1201 because of the post-"copying" non-infringing and legally protected uses it enables. Second, even licensed DVD players engage in some "copying" as they decrypt and play the DVD movie. See Reply Declaration of Frank Stevenson.4. In this Circuit, the presumption of irreparable harm in infringement cases is related to the threat of copying or confusion; it is not derived from the availability of injunctive relief under the Copyright Act. Ironically, the same 1203(b)(1) to which plaintiffs cite for the proposition that irreparable harm is presumed whenever an injunction is possible, concludes by directing that "in no event shall impose a prior restraint on free speech or the press protected under the 1st amendment to the Constitution." The cases plaintiffs cite are not to the contrary. For example, in RealNetworks v. Streambox, Inc., No. 2:99CV02070, 2000 WL 127311 (W.D. WA Jan. 18, 2000), the Court came to the opposite ruling than that for which plaintiffs cite it, finding that there is no presumption of irreparable harm under the Digital Millenium Copyright Act. And in Fisher Price, Inc. v. Well-Made Toy, 25 F.3d 119 (2d Cir. 1994), the Court did no more than confirm the rule that a presumption is permitted for infringement, but that waiting too long to act destroys that presumption. Id. at 124.
Plaintiffs acknowledge that they have found hundreds of websites posting DeCSS all over the United States and Europe, yet they have neither presented any evidence that DVDs are being copied at all, let alone any evidence that DeCSS is being used to do so, nor rebutted defendants' technical and factual showing that distribution of DVD movies on the Internet (with or without the use of DeCSS) is not taking place and that it is not now, and will not be, technically feasible for several years. It is not a question of weeks or months. There is no copying, nor any imminent threat of copying, to support any finding of irreparable harm; and plaintiffs and their witnesses have admitted this on numerous occasions at deposition and to the public. See Hernstadt Declaration, Exhs. A and B.
Plaintiffs now rely instead on accusations of file-sharing technology, intra-campus 100 Mps ethernet connections and LANs, the Napster, and the DivX compression utility. These have nothing to do with the issues of this case, and have no connection to DeCSS. When not testifying about their lack of knowledge regarding the use and/or dissemination of DeCSS, or any copying or piracy connected to DeCSS, plaintiffs have repeatedly asserted that all of their investigations into the use of DeCSS, copying of DVDs, and the source and distribution of DivX-compressed movies on the Internet are "privileged." See Hernstadt Decl.. And while "file-sharing" technology, which plaintiffs incorrectly and misleadingly suggest is being used to disseminate DVDs on the Internet, may be relevant to the exchange of very small files containing music (MP3, the Napster), it is irrelevant to DVDs; networks like Gnutella are not being used on the Internet to move huge files of decrypted DVD movies.
DVD files are far too large to be posted on or down-loaded from the Internet like songs. The time required to transmit DVDs is prohibitive even in the ideal world of college campuses discussed by Robert Schumann. Schumann's theoretical conclusions regarding the speed at which large files can be transferred on via 100 Mbps ethernet connections are incorrect; they cannot be achieved even between two connected computers. See Reply Declarations of Frank Stevenson, Chris Mesong and Olegario Craig. Mr. Schumann, who is plaintiffs' expert on these matters, also grossly exaggerated the transmission speeds available to a hypothetical college student with a 100 Mbps ethernet connection. See Reply Declarations of Olegario Craig, Chris Mesong and Matt Pavlovich. Mr. Schumann's untested guesswork is even more off-base when compared to how comparatively slow (far slower than a T1 line) the actual transmissions on such college and University LANs usually turn are. Plaintiffs' assertion that "end users can use DeCSS to make exact digital reproductions" of DVDs and send them "throughout the world," rhetoric which relies solely on Mr. Schumann's erroneous theorizing, is simply false.(5)
____________________
5. Defendants request that the Court strike the declaration of Franklin Fisher in its entirety. Mr. Fisher frankly acknowledges at paragraphs 2 and 3 of the Declaration that his experience and expertise as an economist does not in any way include the habits and customs of the Internet's users, and that his knowledge about such subjects is anecdotal and derived upon unnamed articles in the press. It is also without value. First, Mr. Franklin relies on the assumption that perfect DVD copies are presently or will soon be available on the Internet, which plaintiffs have admitted is simply not technologically possible. Mr. Franklin ignores the technical impossibility of Internet distribution of perfect DVDs, the fact that the source of the overwhelming majority of movies on the net now are cam-cordings, and that such movies are poor quality. Similarly, Mr. Franklin's speculations about the mindset of college students on the Internet are complete speculation, Finally, Mr. Franklin seriously misstates the opinion of Lewis Kurlantzick, whose expert report was cited by the 9th Circuit, who never states that "the availability of unauthorized product "for free" on the internet will not have a serious financial market impact. Rather, Mr. Kurlantzick's two points are that there is not one-to-one loss relationship between someone's paying two dollars for a pirated DVD and failing to pay eight dollars for a legitimate one; the higher price may be unacceptable; and that this is particularly true if the pirated DVD were more costly than a legitimate one.
Plaintiffs admit that they know of no Internet proliferation of DVD movies enabled by DeCSS at all, and in the eight months since DeCSS was released the MPAA with its vast worldwide resources has not uncovered a single instance of copying related to the utility. The DivX compression utility does not change this fact. As explained in the Reply Declaration of Frank Stevenson, DivX permits the already compressed files on a DVD to be further compressed by a maximum of about 30%. DivX could compress a seven gByte file to five gBytes, with some loss of data and quality of picture. That much data is useless on the Internet unless it is further reduced, with the loss of some 70-90% of the remaining data that was the movie, including some or all of the menus and navigational functions that make DVD special. The quality of the resulting movie is quite degraded and far from plaintiffs' hyptothetical but non-existent perfect digital master instantly distributable to millions of people over the Internet. See generally, Reply Declarations of Frank Stevenson and Matthew Pavlovich.
Plaintiffs are left solely with the speculative claim that there is a possibility of some future harm two or three years from now. This cannot justify a finding today that 1201 bars any and all access to the information on a DVD. Given the huge sales of DVDs and the increasing likelihood that it will become the only format on which movies will be distributed to the public, plaintiffs' claim that 1201(a)(2) has nothing to do with fair use because the public has access to alternate formats (like VHS), is wrong. Not only are some materials unique to the DVD format today, and the quality of DVD vastly superior to that of VHS, but any movie that will only be available on DVD will be forever and completely unavailable to the public for every constitutionally mandated use (other than viewing). This cannot be squared with the Copyright Act or the doctrine of fair use. See, e.g., Sony Corp. of America v. Universal City Studios, 464 U.S. 417 (US 1984); Sony Corp of America v. Connectix, 203 F.3d 596 (9th Cir. 2000).
Plaintiffs claim that 1201(a)(2) must be read in isolation from other provisions in 1201 and, indeed, the Copyright Act itself (Hernstadt Declaration, Exh.E, Attaway Depo., pp 106-07), and understood in that context as a blanket prohibition against the provision of any utility that circumvents CSS, for any purpose.(6) Plaintiffs strip the language of 1201(a)(2) of its obvious complexity and ambiguity and reduce it to a simplistic equation: the provision of any device or program that circumvents a technological measure controlling access to a copyrighted work is prohibited. Making 1201(a)(2) wholly unconnected to any other provision of Chapter 12 of the Copyright Law also creates a fatal constitutional and statutory defect: members of the public are permitted to make non-infringing uses of copyrighted materials on DVDs, but are prohibited access to any utility that would enable them to exercise those rights.
____________________
6. To this end, plaintiffs also attempt to simplify DeCSS itself by repeatedly claiming that all it does, that DeCSS' only function, is to decrypt and copy. Given That is like saying that Netscape's only function is to connect a computer to the Internet.
Section 1201 was Congress' response to concerns raised about copyright infringement and consumers' rights (including fair use) in the digital age. It expressly preserved the public's existing rights by permitting the circumvention of technological protections in order to engage in activities such as fair use and reverse engineering. See, e.g., 1201(a)(1), (c)-(k). Congress did not intend to render these rights and express provisions nullities by denying individuals access to utilities that were necessary to the exercise of the rights set forth. Section 1201(a)(2) must be read in conjunction with 1201(a)(1) and 1201(c) to prevent plaintiffs from blocking the availability of utilities that enable persons to circumvent CSS for permitted purposes, including fair use and reverse engineering, and viewing legitimately purchased DVDs on a consumers' computer and operating system of choice.(7)
____________________
7. Plaintiffs' efforts to ignore the complex interpretative issues presented by definitions such as "effectively controls access to a work," which requires a case-by-case examination of the facts of each situation, are similarly flawed. In this regard, plaintiffs incorrectly relate the basis for defendants' contending that CSS is not "effective." It is not because CSS is weak (which it is) that it is not "effective," but because it overreaches. "Effective" control must be legal control.
The statute's exceptions to the anti-circumvention provisions supports this view. In the face of 1201's listed exceptions listed, plaintiffs propose an interpretation that would bar any provision of a utility that circumvents CSS, even to make fair use of the copyrighted materials on a DVD. Section 1201, however, contains several provisions that expressly forbids this destruction of the fair use doctrine. In particular, the language of 1201(c)(1) -- "Nothing in this shall affect rights, remedies, limitations, or defenses to copyright infringement, including fair use, under this title" -- compels the conclusion that "the phrase 'including fair use,' as set off in commas, modifies not 'defenses to copyright infringement,' but 'limitation...under this title.'" Jane C. Ginsburg, Copyright Use and Excuse on the Internet, 24 Colum.-VLA J.L. & Arts (forthcoming 2000).(8) Section 1201 is part of Title 17 of the United States Code, and fair use applies to all of the rights set out in Title 17, as well as other intellectual property rights such as trademark, see New Kids on the Block v. News America Publishing, 971 F.2d 302 (9th Cir. 1992). Thus1201(c) indicates Congress' oft-stated intent to guarantee fair use, which Congress and the courts have always considered must be flexible enough to change with advances in technology, see H.R. Rep. No. 94-1476, 94th Cong. 2d Sess. 66 (1976), in connection with the anti-circumvention provisions as well. Accordingly, fair use "defenses" to 1201(a)(2) claims must be broad enough to encompass the fair use defenses available in infringement cases, or the statute is unconstitutional.
____________________
8. See Hernstadt Decl., Exh. G. Professor Ginsburg has also expressed this argument in stronger terms, stating that "without an appropriate fair use limitation," the copyright holders' control of access "becomes . . . an Über-copyright law. See, From Having Copies to Experiencing Works: the Development of an Access Right in U.S. Copyright Law, Public Law & Legal Theory Working Paper Group (http://papers.ssrn.com/paper.taf?abstract_id=222493), Paper Number 8 at 16-17.
It is not illegal to provide anti-circumvention devices in connection with reverse engineering. That is not the law as applied to the Copyright Act, and Congress in enacting Section 1201 did not and did not intend to override Sony Corp of America v. Connectix, 203 F.3d 596 (9th Cir. 2000) and Sega Ents. Inc. v. Accolade, Inc., 977 F.2d 1510 (9th Cir. 1992). The absence of a right to provide reverse engineering tools would damage development projects like international LiVid Project, which need tools to engage in necessary reverse engineering. The issue here is not whether the defendants engaged in reverse engineering, but whether 1201(a)(2) is mutually exclusive with 1201(f), which permits the use of a device to circumvent of access controls "for the purpose of enabling interoperability of an independently created computer program with other programs." 1201(f)(2); Sony v. Connectix; Sega v. Accolade. Congress did not, as plaintiffs' imply, intend to permit conduct but prohibit the means. The statute is incoherent unless its provisions are read as a whole.(9)
____________________
9. Plaintiffs' factual contention that DeCSS is not reverse engineering, or that it was irrelevant to Linux DVD player is wrong. Defendants' declarants, who were directly involved in the International LiVid project to reverse engineer an open system DVD player, explained how and why DeCSS is an important tool in the project to create interoperability between an open-source operating system and a CSS-free, open-source DVD player. Plaintiffs' submission of a handful of emails posted on LiVid sites that were carefully culled from hundreds of widely varied contrary messages does not constitute evidence to the contrary. In this regard, it is clear that this argument would apply to an individual attempting to play a DVD she purchased on her Linux operating system. Contrary to plaintiffs' statements, no CSS -licensed DVD player is yet available for Linux. Http://www.wired.com/news/business/0,1367,35311,00.html. Linux and other open source systems are used by millions of consumers -- so many that hardware manufacturers who make CSS chips are helping the development of a Linux DVD player by making available source code regarding non-CSS-related hardware on computers that also contain CSS-related hardware, in order to increase the efficiency of the Linux player.
Fair use must be in Section 1201, or it must be read into Section 1201, or the statute is unconstitutional. With respect to the constitutionality of the preliminary injunction itself, the Court in Junger v. Daley, 209 F.3d 481 (6th Cir. 2000) does not distinguish between source code and object code, and that object code warrants the same scrutiny as source code. It is one click away. Both constitute expressive speech that communicates the same thing. The Junger Court's reasoning applies equally well to both. Of course, DeCSS exists in both forms.
Finally, Plaintiffs' attempt to undermine defendants' proof that the preliminary injunction constitutes and unconstitutional prior restraint on the basis of the holding in Pittsburgh Press Co. v. Pittsburgh Commission on Humans Relations, 413 U.S. 376 (8th Cir. 1973) is also without merit. That case did not involve expressive speech, but commercial speech. Additionally, the key to the Court's holding in Pittsburgh Press was that there was no preliminary injunction (prior restraint) in place. Here, the expressive speech being restrained by the preliminary injunction was published by a media defendant in the course of covering a breaking story of significance to its audience, and neither 2600 nor Mr. Goldstein ever urged anyone to use DeCSS to copy DVDs.(10) See Goldstein Reply Declaration, Exhibit A.
____________________
10. Many of the mirror sites linked to by 2600 contain a great deal of expressive, political as well. See, id., at Exhibits B-C.
The denial of plaintiffs motion and the granting of defendants motion to vacate the injunction is required by the Copyright Clause of the Constitution, the First Amendment and the applicable law.
Dated: New York, New York
June 14, 2000
FRANKFURT, GARBUS, KLEIN & SELZ, P.C.
By:____________________________________
Martin Garbus (MG 6261)
Edward Hernstadt (EH 9569)
488 Madison Avenue
New York, New York 10022
(212) 980-0120
Attorneys for Defendants Eric Corley a/k/a
Emmanuel Goldstein and 2600 Enterprises, Inc.
[Reply Declaration of Frank Stevenson]
Martin Garbus, Esq. (MG 6261)
Edward Hernstadt, Esq. (EH 9569)
FRANKFURT, GARBUS, KLEIN & SELZ, P.C.
488 Madison Avenue
New York, New York 10022
(212) 826-5582
Attorneys for Defendant Eric Corley a/k/a
EMMANUEL GOLDSTEIN
| UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK -----------------------------------------------------------------------X
UNIVERSAL CITY STUDIOS, INC., - against -
ERIC CORLEY a/k/a "EMMANUEL GOLDSTEIN"
|
00 Civ. 0277 (LAK)
REPLY DECLARATION OF FRANK ANDREW STEVENSON IN SUPPORT OF DEFENDANTS' CROSS-MOTION TO VACATE THE PRELIMINARY INJUNCTION |
I, FRANK ANDREW STEVENSON, declare under penalty of perjury under the laws of the United States of America, that the foregoing is true and correct:
1. My name is Frank A. Stevenson and I live in Oslo, Norway. If called upon to do so, I would competently testify to the following facts:
2. As stated in my previous declaration submitted in support of this motion, I am a computer research programmer employed by Funcom Oslo AS. I have 18 years experience in programming, and more than 6 years experience with cryptography. I have been practicing research and development in DVD encryption in particular for six months. My main interest in this endeavor is cryptographical, but as a participant on the LiVid project mailing list (a forum dedicated to providing Video and DVD playback capabilities to the Linux computing platform) I also have a great deal of experience in the mechanics and technology of playing DVDs on computers that are not equipped with DVD players licensed by the DVD CCA, in part by using programs that negate CSS. A copy of my Curriculum Vitae is attached as Exhibit A to my previous declaration.
3. The compression utility DivX is a program that compresses video and audio files. Compressed files are smaller, and may therefore be transmitted faster and take up less computer storage space. DivX is, however, far from the threat that plaintiffs and the MPAA make it out to be. DivX is actually just an implementation of MPEG-4, an standard for video/audio encoding made by the Moving Picture Experts Group ("MPEG"). MPEG is a working group of ISO/IEC in charge of the development of standards for coded representation of digital audio and video. Established in 1988, the group produced MPEG-1, the standard on which such products as Video CD and MP3 are based; MPEG-2 the standard on which such products as Digital Television set top boxes and DVD are based; and MPEG-4, the standard for multimedia on the web. For more information, see http://www.cselt.it/mpeg/. DivX is far from a revolution in compression technology, but actually a mere extension of long-existing technology. See, http://www.cselt.it/mpeg/standards/mpeg-4/mpeg-4.htm.
4. Specifically, DivX does not, and cannot make a perfect copy of a DVD at the file sizes that plaintiffs suggest. Indeed, the video quality -- that is, the quality of the images that a viewer actually sees on his computer -- of a DVD-based DivX is significantly degraded from an actual DVD. Such a "copy," if made, is not a perfect digital master from which further copies could be made and distributed.
5. This is the case because with respect to the video computer files in question, there is a mathematical relationship to the tradeoff between the quality of the image which results from playing the files, and the size of the files themselves. The tradeoff occurs because MPEG utilizes a "lossy" compression scheme, where loss of image quality is traded for reduction in file size. However, DivX does gain some coding efficiency by stripping some of the data needed for navigating within a movie, and employs a improved compression scheme (ACE); these gains come at the cost of some of the functionality that makes DVD a compelling format. Any gains beyond this are controlled by the fact that modern compression techniques are so efficient that a proportional relationship can be shown mathematically to exist between the size of the video files and the quality of the resultant images. Thus, for example, a file one-half the size of a DVD file will look one-half as good (and one-half worse) than the original. Since a DivX file is approximately one-fifth (1/5) the size of an average DVD file, a DVD-based DivX will have about 1/5 the image quality, as measured in information content of the source DVD.
6. Comparing relative quality between different compressed streams is a very subjective matter. A viewer who knows how to identify the artifacts of a heavily compressed file -- i.e., what to look for -- may find a particular file unacceptable, while an untrained eye may very well ignore these artifacts. Stated numerically, in compressing the video and audio, the following comparisons was made by MPEG when evaluating MPEG-4 by showing samples to a test audience. For more information, see http://www.cselt.it/mpeg/standards/mpeg-4/mpeg-4.htm#E15E24 . (a) For the object based case, the quality provided by the ACE Profile at 256 kb/s is equal to the quality provided by the Main Profile at 384 kb/s; (b) For the frame based at low bit rate case, the quality provided by the ACE Profile at 128 kb/s and 256 kb/s is equal to the quality provided by the Main Profile at 256 kb/s and 384 kb/s, respectively; (c) For the frame based at high bit rate case, the quality provided by the ACE Profile at 768 kb/s is equal to the quality provided by the Main Profile at 1024 kb/s.
7. These figures support an increased coding efficiency of about 40% for DivX. If this figure is correct, however, a 7 GB DVD can only be converted to a 5 GB MPEG-4 stream, with little degradation, with DivX. This buys the "copier" -- or anyone wishing to utilize DivX to enjoy DVDs in any way -- virtually nothing at all. In order to fit the newly DivX compressed bitstream onto even 2 CDs, it would be necessary to discard 74% of the recorded data. Whether the reduction is made in the form of reduced resolution, or degraded image quality, the result is the same: a perfect DVD copy cannot be made. Moreover, once data is discarded from a datastream, it is impossible to recover, except by re-reading the original source; the compressed stream has no memory of what has been lost.
8. Another misconception is that DVD players only "play" DVDs. Actually, an authorized, duly licensed computer-based DVD player must, in order to operate, decrypt and "copy" the DVD files. DVD players copy the decrypted information into RAM, as opposed to hard disc, and thus copy and decrypt in a different manner than DeCSS by copying somewhat less and storing it for a shorter period of time. Nevertheless, the files are decrypted and copied, and for this very reason programs like PowerRipper are able to latch themselves on to licenced players and extract the decrypted data.
9. With regards to using DeCSS as a means to bypass region codes I would
like to note that the great majority of DVD drives installed in computers
today are RPC-1 drives. RPC-1 drives leave it to the player software to enforce
region codes, they merely report which region the current disc is coded for.
DeCSS.exe pays no attention to the reported region code, and will be able
to access a DVD disc, regardless of which region it is coded for. In the
output from DeCSS.exe these region codes are stripped, so there is no regional
coding information available to enforce, even for a licenced player, in which
case the movie will be played without regards to the region code of the original
disc. The licensors of DVD drives are aware of this loophole in the region
codes enforcement, and for this reason the newer drives of type RPC-2 enforce
regions codes internally. However, I am aware of firmware upgrades to RPC-2
drives that will make them RPC-1 compatible. This means that DeCSS.exe will
be useful for evading region codes restrictions for many years, be it with
original RPC-1 drives, or 'downgraded' RPC-2 drives. Any suggestion to the
contrary by the plaintiffs or the MPAA would be disingenuous.
Dated: June 14, 2000
Oslo, Norway
_________________________________
FRANK ANDREW STEVENSON
[Declaration of Edward Hernstadt]
Martin Garbus, Esq. (MG 6261)
Edward Hernstadt, Esq. (EH 9569)
FRANKFURT, GARBUS, KLEIN & SELZ, P.C.
488 Madison Avenue
New York, New York 10022
(212) 826-5582
Attorneys for Defendant Eric Corley a/k/a
EMMANUEL GOLDSTEIN
| UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK -----------------------------------------------------------------------X
UNIVERSAL CITY STUDIOS, INC., - against -
ERIC CORLEY a/k/a "EMMANUEL GOLDSTEIN"
|
00 Civ. 0277 (LAK)
REPLY DECLARATION OF EDWARD HERNSTADT IN SUPPORT OF DEFENDANTS' CROSS- MOTION TO VACATE THE PRELIMINARY INJUNCTION |
I, EDWARD HERNSTADT, hereby declare under penalty of perjury that the foregoing is true and correct.
1. I am associated with Frankfurt, Garbus, Klein & Selz, P.C. I have personal knowledge of the matters to which I attest herein, except where otherwise stated.
2. I was present at the depositions of Robert Schumann, Kenneth Jacobsen, Daphne Gronich, Jack Valenti and, most recently, Fritz Attaway. Relevant excerpts from the transcripts of these depositions are annexed hereto as Exhibits A - E.
3. As set forth more specifically below, these deponents readily admit, and plaintiffs must therefore concede, that neither the plaintiffs nor the MPAA have any knowledge of: (i) any unauthorized copying or distribution of DVD movies connected to, or arising out of, DeCSS, (ii) any amount of unauthorized copying or distribution, i.e. "piracy" of DVDs that is occurring over the Internet, (iii) any DVDs that have been compressed with the utility DivX in connection with the unauthorized copying or distribution of DVDs on the Internet, or even (iv) any use of the utility DivX that is in any way connected with any actual downloading, copying, distribution, or other use of DVDs, with or without DeCSS. Excerpts from transcripts related to these matters are annexed hereto at Exhibits A - C.
4. Defendants only learned of the extent to which plaintiffs were speculating as to the "irreparable" and imminent harm they alleged would be caused by proliferation of DeCSS as a result of taking depositions. We learned that plaintiffs have no factual basis for this claim, and that the injunction now in place is based on an unsupportable finding of irreparable harm - As a result, even through the exercise of extraordinary diligence, defendants could not have presented this evidence to the Court in time for such evidence to have been considered at the hearing on the preliminary injunction.
5. Neither the MPAA, nor any of the plaintiffs (all of which the MPAA essentially represents in investigating "piracy") are aware of any copying or distribution of DVD movies attributable to DeCSS. See, Deposition Transcript excerpts annexed hereto at Exhibit A (E.g., Jacobsen Depo, 7:23 - 8:4 (no personal knowledge of anyone who has used DeCSS to hack a DVD and make an unauthorized copy), 10:5 - 11:14 (only knowledge of DeCSS being used at all is from claim of reporter in newspaper article; MPAA investigation has not gone beyond responding to "posting and linking" with cease and desist letters)), 44:20 - 47:9 (MPAA has not determined, and has no knowledge concerning, whether any website claims regarding the use of DeCSS to hack DVDs have any validity at all), and [Text Omitted; Designated "Confidential"]; Schumann Depo, 65:10 - 66:5 (only knowledge of DeCSS "being used" to decrypt DVDs since date he was retained by plaintiffs is from article in Toronto Star describing failed experiment to use DeCSS, and from defendants' declarations), 28:3 - 16 (Prior to January 14, only "knowledge" of the use of DeCSS to copy DVDs was taken from review of claims made in "Linux development logs"); Valenti Depo, 14:25 - 19:1 (has no knowledge of whether a DVD has ever been decrypted by DeCSS, whether a DVD has ever been viewed via DeCSS over the internet, and is unaware of the existence of any documents which would "refresh his recollection" as to any of these matters)).
6. Further, neither the MPAA nor the plaintiffs are actually aware of any amount of unauthorized copying or distribution, i.e. "piracy" of DVDs that is occurring over the internet. See Deposition Transcript excerpts annexed hereto at Exhibit B (E.g., Attaway Depo, 111:4 - 22 (admits that there is not, as of today, a sufficiently extensive broadband network coupled with high enough available rates of compression to permit massive motion picture piracy on the internet); Jacobsen Depo, 321:20 - 322:11 (MPAA admits that no lost sales have been attributable to the use of DeCSS), 240:3 - 18 (Of all "pirated" media, pirated DVDs account for only between 1% and 2%), and [Text Omitted; Designated "Confidential"]; Schumann Depo, 106:7 - 109:12 (he has no specific knowledge whether any of the plaintiffs or the MPAA even claim that there has ever been a pirated DVD enabled by DeCSS actually sold), and 172:2 - 22 (he has no personal knowledge of anyone who has tried to transmit a DVD via the internet, including mere downloading, but rather has "just seen people talking about doing it or having it done"); [Text Omitted; Designated "Confidential"]).
7. Neither the MPAA nor the plaintiffs are actually aware of any DVD files that are being compressed with the utility DivX in connection with any copying or distribution of DVDs, or of any use of DivX that is in any way connected with any actual downloading, copying, distribution, or other use of DeCSS. (See, Deposition Transcripts annexed hereto at Exhibit C (E.g., Jacobsen Depo, [Text Omitted; Designated "Confidential"], 263:4 - 19 (Jacobsen has only seen one DivX "exhibition"; it is too new a technology, and the MPAA is unaware of any specific instances in which any VCDs have been created using DivX), [Text Omitted; Designated "Confidential"]; and Schumann Depo, 100:8 - 19 (His only knowledge of DivX being used in conjunction with DeCSS is derived from an article he read in the Toronto Star (this article is annexed hereto at Exhibit F)).
8. Despite their inability to adduce even a scintilla of actual evidence that DeCSS has been, or will be, used to pirate DVDs, plaintiffs and the MPAA concomitantly insist that most, if not all, of any knowledge they might have, including information concerning past or ongoing investigations into such matters, is the subject of privileged attorney-client communications. Plaintiffs thus infer that information concerning use of DeCSS and related piracy is out there and available, but they cannot divulge it. See, Deposition Transcript excerpts annexed hereto at Exhibit D (E.g., [Text Omitted; Designated "Confidential"]; Schumann Depo, 65:10 - 25 (after testifying that he had no specific personal knowledge about the actual use of DeCSS to decrypt DVDs, was not permitted to respond, on grounds of attorney-client privilege, to whether he had acquired knowledge about actual use of DeCSS through any means other than reviewing the Toronto Star article and defendants' declarations); Attaway Depo, 36:10 - 14 (Attaway has had no contact with the plaintiffs concerning DeCSS in a non-privileged context); [Text Omitted; Designated "Confidential"]).
9. Lastly, the deposition of Fritz Attaway discloses that is the plaintiffs' and the MPAA's position with regards to fair use and other non-infringing uses of DVD material that there can be absolutely none. That is, noone is permitted to "access" as opposed to "view" the copyrighted material contained on a DVD at all, under any circumstances. See, excerpts from Attaway Depo annexed hereto at Exhibit E (E.g. 27:12 - 29:3 (the Fair Use Doctrine has never permitted unauthorized access; "unauthorized access" means access without the permission of the copyright holder); 29:21 - 32:6 (a consumer does not get authorized access to the material on the DVD by purchasing the DVD; "access" is only gained by purchasing the DVD and a licensed player); 68:18 - 69:7 (Attaway testifies that a consumer who purchases a CSS encrypted, authorized DVD and a DVD CCA-licensed DVD player thereby acquires the right only to view the copyrighted material contained on the DVD; such consumer is not thereby authorized to access the material on the DVD to make a fair use, or any non-infringing use, of such material); and 136:25 - 137:16 (prohibiting all access to DVD materials does not impact fair use because works on DVD are available in other formats which permit fair use by those with a "strong interest" in exercising fair use).
10. In addition to the deposition testimony referred to above, several articles have been written about DeCSS, DivX, and the Internet, and the potential impact of such technologies on the movie industry. One conclusion that can be drawn by these articles, including the Toronto Star article frequently named by plaintiffs' deponents as a source of information concerning the alleged proliferation of DeCSS and DVD copying, is that the technology simply does not exist to permit DeCSS or DivX, used separately or in conjunction with each other, to be used via the Internet for commercial piracy. Another conclusion that can be drawn is that DivX is extremely difficult to use, and produces substantially degraded and inferior images compared to DVD. See Articles annexed hereto at Exhibit F.
11. Annexed hereto as Exhibit G is a copy of Jane C. Ginsburg, Copyright Use and Excuse on the Internet, 24 Columbia - VLA J. L. & the Arts xx (May 31, 2000).
Dated: New York, New York
June 14, 2000
__________________________________
EDWARD HERNSTADT
[Exhibits not provided]
[Reply Declaration of Chris Moseng]
Martin Garbus, Esq. (MG 6261)
Edward Hernstadt, Esq. (EH 9569)
FRANKFURT, GARBUS, KLEIN & SELZ, P.C.
488 Madison Avenue
New York, New York 10022
(212) 826-5582
Attorneys for Defendant Eric Corley a/k/a
EMMANUEL GOLDSTEIN
| UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK -----------------------------------------------------------------------X
UNIVERSAL CITY STUDIOS, INC., - against -
ERIC CORLEY a/k/a "EMMANUEL GOLDSTEIN"
|
00 Civ. 0277 (LAK)
REPLY DECLARATION OF CHRIS MOSENG IN SUPPORT OF DEFENDANTS' CROSS-MOTION TO VACATE THE PRELIMINARY INJUNCTION |
I, CHRIS MOSENG, declare under penalty of perjury under the laws of the United States of America, that the foregoing is true and correct:
1. I am Christopher Moseng, an Assistant Manager of Technology for Kinko's in Coon Rapids, Minnesota. My position involves the configuration and management of a cross- platform network that consists of 3 Macintoshes, 12 IBMs, and 6 printers. I am responsible for the 24-hour operation of the rental and full-service computers and network, our connection to the internet and our link to the Kinko's corporate network in California. Currently I pursue a Bachelor's of Network Administration from the University of Minnesota. In this declaration I wish to address claims made in the Second Supplemental Declaration of Robert W. Schumann regarding the facility and speed with which a computer user can transmit the large files that result from using DeCSS on a DVD movie over a 100Mbps network. I also wish to address his statements regarding the availability of 100Mbps networks in college dorms.
2.In order to analyze his claims, I performed a series of time trials over my private 100Mbps, full-duplexed, switched network. There are three computers on this network, and the only network traffic during the trials were the trials themselves.
3.This network configuration is not representative of a standard consumer home network, and certainly not representative of a standard consumer internet connection, but reflects optimal conditions in a high-end consumer or a standard institutional network.
4.Nodes on 100Mbps and 10Mbps Ethernet networks must be within 100 meters of their switch; therefore, all computers on the same network must be within 200 meters of one another. In most cases, the computers are much closer to one another: often in the same room or on the same floor.
5.I attempted to copy the movie The Matrix from one computer to another over the network. Both computers run Windows 98 Second Edition and use Windows Networking to share files. The only special preparation taken to perform the trials was to make room on the computers' hard drives to accommodate the files.
6.The 20 files produced by DeCSS total 7,746,538kB in size. Among the trials that completed successfully, the fastest time in which all files produced by DeCSS were copied was 64 minutes. In comparison, removing the DVD from the first computer and putting it into the second computer took 15 seconds. This second method of transfer will consistently move data between computers on a 100Mbps network at orders of magnitude faster and with greater ease than using the network itself because of the network's relative lower bandwidth, distance limitations, and the significant available storage space necessary.
7.With regards to Mr. Schumann's claim that most colleges provide 100Mbps network connections in their dorm rooms, my experience and knowledge suggests otherwise. While attending Grinnell College from 1996 to 1998, they completed retrofitting their dorms with a 10Mbps network. Before this time, there was no network in the dorms and internet access was provided in computer labs. The primary use for the in-room network connections is internet access. A campus-wide 10Mbps network is beyond adequate to saturate the Grinnell's T-1 connection to their ISP, because all the computers on the campus network need to share this (1.544Mbps) internet connection. Installing a 100Mbps network would not speed internet access for any user, and so would not be cost effective.
8.A person attempting to transfer large files such as those in my time trials over a 10Mpbs network would need to wait at least 10 times as long under optimal conditions. In an environment such as a college dorm, speeds on the network are never optimal because of other users' network traffic and network overhead. Moreover, while one switch may provide full 10Mbps connections to its clients, it only communicates with other switches at 10 or 100Mbps. Again, all users on the same switch must share this linking bandwidth, so moving files beyond the network boundary will be further speed restricted.
9.The net result of these trials demonstrates the significant inefficiency of transferring the files of a DVD over a 100Mbps network, in contradiction with statements in Mr. Schumann's Second Supplemental Declaration in which he states it can be accomplished in seven minutes, and suggests that this method may be an efficient avenue for movie piracy.
10.In short, borrowing a DVD is more efficient and desirable than contributing the time, effort and resources necessary for copying it over 10Mpbs or 100Mbps networks.
11.Additionally, while Mr. Schumann claims that colleges operate 100Mbps networks in their dorms, this runs counter to the purpose of installing the networks in the first place. Their primary use is to provide internet access to students, and a campus-wide 10Mpbs network will saturate most colleges' internet connections; this provides a dis-incentive to undergo the significantly higher cost of providing a 100Mpbs network in the dorms.
12.Ultimately, such networks are not a viable medium to copy DVD movies either casually or deliberately. It takes too long, requires too much storage space, consumes too many computer and network resources, and abandons the attractive features of the DVD itself, not the least of which are portability and affordability.
13.The following is a description of the computers used and the trial test results:
Computer L: AMD K-6 300, 128MB RAM, IDE-33 HD, Netgear FA-310 10/100 Ethernet adapter, Creative Labs 2x DVD drive, Windows 98 SE with dynamic virtual memory and in "Network Server" mode (otherwise unoptimized).
Computer K: AMD Athlon 600, 128MB RAM, IDE-66 HD, Netgear FA-310 10/100 Ethernet adapter, Toshiba 2x DVD drive, Windows 98 SE with dynamic virtual memory and in "Network Server" mode (otherwise unoptimized).
Trial 1 Methodology: At computer K, use DeCSS to decrypt The Matrix in the DVD drive on L and have the results put into an empty folder on the desktop.
Trial 1 Results: DeCSS crashed. No files were decrypted or copied.
Trial 2 Methodology: Map the C: drive on K to drive G: on L. Use DeCSS to decrypt the contents of The Matrix, and put them directly into in an empty folder on K. Perform all operations from computer L in the DeCSS user interface.
Trial 2 Results: This operation took 80 minutes. At the operation's completion, all data files from the DVD were in the folder on K.
Trial 3 Methodology: Copy the new decrypted files from drive C: on K to drive C: on L.
Trial 3 Results: The operation took 64 minutes.
Dated: St. Paul, Minnesota
June , 2000
___________________________
CHRIS MOSENG
[Reply Declaration of Olegario Craig]
Martin Garbus, Esq. (MG 6261)
Edward Hernstadt, Esq. (EH 9569)
FRANKFURT, GARBUS, KLEIN & SELZ, P.C.
488 Madison Avenue
New York, New York 10022
(212) 826-5582
Attorneys for Defendant Eric Corley a/k/a
EMMANUEL GOLDSTEIN
| UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK -----------------------------------------------------------------------X
UNIVERSAL CITY STUDIOS, INC., - against -
ERIC CORLEY a/k/a "EMMANUEL GOLDSTEIN"
|
00 Civ. 0277 (LAK)
DECLARATION OF OLEGARIO L. CRAIG IN SUPPORT OF DEFENDANTS' CROSS-MOTION TO VACATE THE PRELIMINARY INJUNCTION |
I, OLEGARIO L. CRAIG, declare under penalty of perjury under the laws of the United States of America, that the foregoing is true and correct:
1. My name is Olegario L. Craig. Since 1995 I have been employed as a network administrator and software specialist for the Department of Computer Science at the University of Massachusetts. In the course of my work I regularly maintain, evaluate, optimize, and troubleshoot the performance of networked computers and associated hardware and software. I have extensive experience working with many computer and network operating systems, file sharing protocols, and client-server applications.
2. I submit this Declaration in support of defendants' cross-motion to vacate the preliminary injunction issued in this case on January 20, 2000, and to counter certain factual errors committed by Robert W. Schumann in his Second Supplemental Declaration, specifically with respect to errors made by Mr. Schumann in dealing with compression technologies and data transfer speeds over computer networks.
3. Mr. Schumann claims that: (a) A 5 Gigabyte file can be transferred in under 7 minutes over a 100Mbps ethernet connection; (b) Most colleges and universities provide 100Mbps ethernet connections to all dorm rooms; and (c) A DVD image file can be compressed (using technologies such as DivX) to a much smaller size than the original.
4. First, in order to assess Mr. Schumann's transfer-speed claims, I performed the following tests (precise details are attached hereto as Exhibit 1):
a. Transfer of a 1.5 Gigabyte file over the Internet, using two computers connected via multiple T1 (minimum 24 megabit-per-second) infrastructure to the Cable & Wireless (formerly BBN) Internet backbone. This took 7 hours and 46 minutes for the 1.5GB transfer. Extrapolation to a 5 Gigabyte file yields an estimated time of 28 hours, 29 minutes.b. Transfer of a 1.5 Gigabyte file between two computers over a standard LAN (Local Area Network) using 10 megabit-per-second switched ethernet took 2 hours and 15 minutes. A 5GB transfer would therefore take roughly 7 and a half hours.
c. Transfer of a 1.5 Gigabyte file over a state-of-the-art high-speed LAN built on 100 megabit-per-second fast-switched ethernet took 13 minutes and 6 seconds for the 1.5GB file; the 5GB file would take approximately 48 minutes.
d. Transfer of a 1.5 Gigabyte file over a single direct "crossover" connection between two computers, using 100 megabit-per-second networking. This "absolute-best-case possible" scenario still took 4 minutes and 2 seconds to effect the 1.5GB transfer; so even after cutting all extraneous network devices and IP routing out of the connection, a 5GB file download would take approximately 13 minutes. I must point out that one would never see this type of direct connection on a normal LAN; it is usually done only to connect two otherwise isolated computers or, (as here), to absolutely minimize network overhead during data transfer testing.
5. It should be noted that these tests were performed using a protocol (FTP) that is optimized for fast and reliable file download. Using a network disk scheme (such as Windows(tm) filesharing or NFS) would only increase the protocol overhead, as these methods need to exchange far more information between computers in order to keep the physical disk synchronous with the networked computer's idea of its file system. This can only result in slower file transfer speeds.
6. Based on these tests, I conclude that Mr. Schumann's claims of a seven-minute transfer time (for a DVD imagefile comprising or exceeding 5 Gigabytes) are not reproducible even under the most favorable conditions. Moreover, my trials were conducted using high-speed computers and networking resources available to me in a state-of-the-art facility devoted specifically to computing research. These conditions are unlikely to be available to many outside such an establishment. Internet connections, even on computers (such as those in trial "(a)," above, that could be considered "neighbors" in terms of Internet topology, would be significantly slower.
7. Mr. Schumann mentions compression technologies, and claims that a DVD compressed to 1.2 Gigabytes using DivX or similar technologies would be downloadable in under 2 minutes. While that figure, like his others, does not reflect reality, it is perhaps more important to note that the compressed file would not be comparable in quality to the original DVD. DivX (like any other compression technology capable of such a feat) is a "lossy" compression method; it is not possible to decompress its output and recover the data discarded during the compression process. Doing so would be -- please pardon a pithy comparison -- akin to producing authentic fresh-squeezed orange juice from frozen concentrate.
8. Lastly, the University of Massachusetts is typical of most higher educational institutions in that it provides 10 megabit-per-second (not 100Mbps) connections to selected residence halls. Even Harvard - which certainly has at least as much money to spend on such matters as UMass - uses 10Mbit networking to its dorms. (See for instance http://www.fas.harvard.edu/computing/comp2004/AdmittedStudents04.pdf). This is analogous to the configuration I used for trial (b), above, where I estimated a time of 7.5 hours for a 5GB download under ideal conditions. Those "ideal conditions" will never exist in our dorms; under even a moderate network usage load, I estimate that file transfers of that size would take at least two or three times that long. Moreover, my tests did not reflect the possibility of multiple clients attempting to download at the same time, which would swamp the source computer's network connection as well as putting a heavy drain on its CPU, memory, and disk bus. Three or four simultaneous download attempts could easily stretch the time required for each to complete into a matter of days.
9. It should also be noted that UMass, like other institutions, does not provide the dorms with the full benefit of its Internet bandwidth. The residence halls' network is separated from the backbone that connects the rest of the campus to the Internet; all of the dorms (those lucky enough to be networked) combined cannot use more than 10Mbps of our 24Mbps Internet connection bandwidth. There are no plans to replace the dormitory networking infrastructure with 100Mpbs technology, which is significantly more costly and has more stringent limitations in terms of individual cable lengths and shielding requirements.
Dated: Amherst, MA
June 14, 2000
_____________________________________
OLEGARIO L. CRAIG
[Exhibit 1 to Reply Declaration of Olegario Craig]
Description of hardware and methodology
used for tests mentioned
1. All tests used anonymous FTP to transfer a 1.5 gigabyte file from a computer
on my desktop, hereinafter referred to as the "Source computer." The file
was created by doubling the binary image of a RedHat 6.2 installation CDrom
and then truncating to exactly 1.5 gigabytes (1610612736 bytes). This figure
was chosen in part because (as of the current stable kernel) the Linux ext2
filesystem cannot handle filesizes in excess of 2 gigabytes. 1.5 gigabytes
seems to me to be enough of a sample from which to extrapolate.
2. The latter three tests all used linux, `wget' (an FTP client that allows URL specification on the command line) and GNU `time' to time the wget operation.
3. Source computer hardware:
Intel Pentium II (Deschutes) 500MHz processor 256Mb 100MHz SDRAMAdaptec 2940 Ultra SCSI bus (40Mbyte/sec sustained Western Digital WDE2170-1807A3 HDD, 40Mbyte/sec transfer rate (system disk)
Seagate ST19171W HDD, 40Mbyte/sec transfer rate (data disk)
3Com 3c905B (Cyclone) 100baseTx ethernet adapter
352 Mbytes system swap space on system HDD
4. Source computer software:
RedHat Linux version 6.2Kernel 2.2.14-12
wu-ftpd-2.6.0-3
anonymous FTP area on data HDD
5. For each trial, I'll detail the network topology between computers and the known information regarding the target computer.
6. Trial "a": Internet transfer between two well-connected computers. Source computer is connected via 100megabit/s full-duplex ethernet to a 3Com 3300 SuperStack 10/100 autonegotiating switch, which is in turn connected via gigabit ethernet to a Cisco 7507 router, which connects via another 3Com 3300 SuperStack to the UMass Internet backbone router. Thus at no point between Source computer and the Cable & Wireless backbone is there less than 24Mbps bandwidth.
7. Target computer "a" hardware:
Dell Optplex GX1
Intel PentiumII 400MHz
64Mbytes SDRAM
(disk subsystem speeds are not relevant for this test, as the bottleneck is the Internet itself)
8. Target computer "a" software:
Microsoft(tm) Windows NT Workstation v.4 servicepack 3Microsoft(tm) Internet Explorer 4.0
9. Target computer is connected to the Internet via 100Mbit/sec ethernet, fast ethernet switch, 3 100BaseTx lines to a Cisco router, and a DS3 (45Mbit/sec) line to the Cable & Wireless backbone, consecutively. Thus at no point between Target computer "a" and the Cable & Wireless backbone is there less than 45Mbps bandwidth.
10. Traceroute from Target computer "a" to Source computer:
tracing route to hex.cs.umass.edu [128.119.243.169] over a maximum of 30 hops: 1 <10ms <10ms <10ms 172.16.38.1 2 <10ms <10ms <10ms 172.16.1.2 3 <10ms <10ms <10ms cvextrt1-fe0.classifiedventures.com [172.18.254.1] 4 10ms <10ms <10ms cvexort2-hs4-0.classifiedventures.com [172.18.1.2] 5 10ms <10ms <10ms 172.25.250.2 6 10ms 10ms <10ms exort1-fe4-0.classifiedventures.com [216.33.87.82] 7 <10ms 10ms 10ms 216.33.87.66 8 10ms <10ms 10ms dcr03-g2-0.okbr01.exodus.net [216.33.64.177] 9 10ms <10ms <10ms ibr02-g4-0.okbr01.exodus.net [216.34.183.67] 10 <10ms 10ms 10ms cw-px.exodus.net [216.32.132.182] 11 10ms 10ms 10ms corerouter1.willowsprings.cw.net [204.70.9.135] 12 30ms 30ms 30ms xcore4.boston.cw.net [204.70.150.89] 13 30ms 30ms 30ms cwusa-mci-worldcom.boston.cw.net [204.70.179.122] 14 30mw 30ms 30ms lgrc-rt-106-8-999.gw.umass.edu [128.119.3.238] 15 30ms 30ms 30ms cs-gw-ext-i1.cs.umass.edu [128.119.3.153] 16 40ms 30ms 20ms hex.cs.umass.edu [128.119.243.169] Trace complete.
11. Trial "b": Transfer over a 10Mbit/sec LAN.
Target computer "b" hardware:
Intel Pentium II (Deschutes) 450MHz processor256Mbytes SDRAM
Adaptec 2940 Ultra SCSI adapter (40Mbytes/sec transfer)
Seagate ST39175LW HDD, 40 Mbytes/sec transfer rate
3Com 3C905B (Cyclone) 100BaseTx ethernet adapter
517 Mbytes system swap
12. Target computer "b" software:
RedHat Linux version 6.2Linux kernel version 2.2.14-5.0
wget-1.5.3-6
GNU time 1.7-9
13. Target computer "b" is connected to a 3Com 3300 SuperStack switch; the switched port to which "b" connects has 10/100 Autonegotiation turned off and speed forced to 10Mbit/second full-duplex. Source computer is connected to the same 3Com 3300 SuperStack switch via its normal 100Mbit/sec full-duplex connection. While this arrangement is actually somehwat faster than a typical LAN based on 10Mbit/sec ethernet, it suffices for the purposes of this test. Note that this test is the most relevant for someone wishing to evaluate the situation of the typical college dormitory.
14. Trial "c": Transfer over 100Mbit/sec LAN.
Target computer "c" hardware:
Gateway ALR 9200 Xeon4 Intel Xeon(tm) Pentium II 400MHz processors
15. Target computer "c" software:
RedHat Linux version 6.0Linux kernel version 2.2.10
Adaptec 2940 Ultra SCSI bus (40 Mbyte/sec transfer)
IBM DDRS-34560D HDD, 40 Mbyte/sec transfer rate (system disk)
ICP-Vortex GDT6538RD RAID 5 SCSI bus (3 channels, 80 Mbyte/sec transfer rate for each channel) (data disks)
16. Target computer "c" is connected via 100Mbit/sec full-duplex ethernet to a 3Com 3300 Superstack Switch, which is connected via full-duplex 100Mb/sec ethernet to another 3Com 3300 SuperStack switch, which is connected via Gigabit ethernet to another 3Com 3300 SuperStack switch, which is connected via full-duplex 100Mbit/sec ethernet to Source computer's 3300 SuperStack switch. This is typical of a large high-speed switched ethernet installation.
17. Target computer "d" hardware:
Intel Pentium II (Deschutes) 450MHz processor256Mbytes SDRAM
Adaptec 2940 Ultra SCSI adapter (40Mbytes/sec transfer)
Seagate ST39175LW HDD, 40 Mbytes/sec transfer rate
3Com 3C905B (Cyclone) 100BaseTx ethernet adapter
517 Mbytes system swap
18. Target computer "d" software:
RedHat Linux version 6.2Linux kernel version 2.2.14-5.0
wget-1.5.3-6
GNU time 1.7-9
19. Target computer "d" is connected (via 100Mbit/sec full-duplex ethernet crossover cable) DIRECTLY to Source computer. Target and Source computers in this trial were not otherwise connected to any networking equipment, and were both running in Linux "single-user" mode with networking support; this means that all extraneous demands on CPU time, RAM, and system busses were eliminated.
[Reply Declaration of Matthew Pavlovich]
Martin Garbus, Esq. (MG 6261)
Edward Hernstadt, Esq. (EH 9569)
FRANKFURT, GARBUS, KLEIN & SELZ, P.C.
488 Madison Avenue
New York, New York 10022
(212) 826-5582
Attorneys for Defendant Eric Corley a/k/a
EMMANUEL GOLDSTEIN
| UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK -----------------------------------------------------------------------X
UNIVERSAL CITY STUDIOS, INC., - against -
ERIC CORLEY a/k/a "EMMANUEL GOLDSTEIN"
|
00 Civ. 0277 (LAK)
REPLY DECLARATION OF MATTHEW PAVOLOVICH IN SUPPORT OF DEFENDANTS' CROSS-MOTION TO VACATE THE PRELIMINARY INJUNCTION |
I, MATTHEW R. PAVLOVICH, declare under penalty of perjury that the foregoing is true and correct:
1. I am the President of Media Driver, LLC, a software and services company. This Declaration is submitted in opposition to the plaintiffs' motion to modify the preliminary injunction issued in this case on January 20, 2000, and in support of the defendants' cross- motion to vacate the injunction.
2. In this declaration, I focus on the following points:
a. Clarification of my previous declaration in which there was a typing error;b. The plaintiffs' claims that: (i) defendants' are incorrect in asserting that DeCSS does not result in unauthorized copying. (See, http://cryptome.org/mpaa-v-2600-rso.htm#Gold); and that (ii) that Jon Johansen, an alleged creator and poster of DeCSS, freely admits that DeCSS can be used for "piracy."
c. DeCSS contributes to the advancement of an open source/Linux DVD Player, even though it is a windows program.
d. Commercial hardware vendors (Companies which are CSS licensed, or DVD-CCA members, such as Creative labs, Luxsonor, SiS, Matrox, Zoran, ATI) appreciate the market for an open source/Linux DVD player, and in fact have actively supported the development of a player.
3. Plaintiffs attempt to question statements in my First Declaration by taking an excerpt from paragraph 10 thereof in which there was a typographical error. In paragraph 13 of Robert W. Schumann's declaration, he states:
Thus, in paragraph 10 of Matt Pavlovich's Declaration, there is a statement to the effect that "[a]ny effort to simply play the unencrypted content stored on a hard disc or other large medium would be futile, due to the fact that there is no known player that can play from anything but a DVD disc." This is wrong.
I thank Mr. Schumann for pointing out my error. The sentence he quotes from my declaration should have stated:
[a]ny effort to simply play the encrypted content stored on a hard disk or other large medium would be futile, due to the fact that there is no known player that can play a DVD movie from anything but a DVD disc.
The simple typographical error was using the word 'unencrypted' instead of the correct 'encrypted'.
4. The reason there is no DVD player that can play encrypted content from anything but a DVD disk is due to the fact that the DVD disk contains sectors also known as the "disk key block" (which are normally inaccessible to anyone except to the DVD drive), and contain the necessary keys to decrypt the encrypted portion of the DVD disk. Attempting to play the encrypted content would yield errors, because the DVD player application cannot decode the DVD movie data that is encrypted.
5. The plaintiff's assertion that Jon Johansen has stated that DeCSS can be used for piracy is misleading. What Jon stated was "[w]ell, yes, it can be used for pirating . . . [b]ecause you can decrypt a DVD disk and put it on your hard drive and then you can convert it, say, to VCD and then post it on the Internet." Converting a DVD movie down to a VCD, however, is a tedious process that requires a multitude of utilities. It is very time consuming to convert a DVD to VCD, and there is a substantial quality loss from DVD to VCD. Further, posting the now substantially lower-quality VCD on the internet is still not a simple task. One would need to have off site storage of 650Mb to store the lower quality movie, and downloading 650Mb across, for example, a 56k modem connection would take over 25 hours.
6. Mr. Schumann also states at 16 of his Second Supplemental Declaration that "[t]here are a considerable number of DVD discs that are released without CSS encryption." In reality, the vast majority of DVD discs are encrypted. Furthermore, there is not an official list of the all DVD movie titles that are not encrypted. Sporadic listings of unencrypted titles have only become available since tools have become available that allow people to test movies which they have already purchased. The web page at http://www.videolan.org/freedvd.html, for example, provides a listing of less than 20 movies that are not encrypted, and over 70% of the listed titles have non-US region codes (thus rendering them unable to be viewed on licensed DVD players in the United States).
7. Several CSS licensed companies have a vested commercial interest in the development of the Linux DVD player, and have made traditionally proprietary source code information available so the open-source Linux player can better utilize their products. Just recently, for example, Silicon Integrated Systems Corp. announced that they will be supporting the Linux DVD player efforts, by developing code to support their hardware DVD decoder chips. See, http://www.sis.com.tw/news/linux.htm.
Creative Labs, Inc. made source code for one of their hardware DVD decoder boards available, so it can be fully supported in the Linux DVD player. See their website, http://opensource.creativelabs.com, which serves as a resource for developers wishing to utilize the Creative Labs Dxr2 hardware DVD decoder chip. Luxsonor Semiconductors, Inc. also has provided low level source code to one of their hardware DVD decoder boards that is found in laptops. The source code allows support for their decoder board to be developed for the Linux DVD player. See, http://www.luxsonor.com/pcsupport.html. On their page they state: "Since we are unable to port to Linux or Win2000, we have decided to release our source code to the public."
8. The Linux development model (also known as open source, or free software development) is beneficial to smaller companies, such as Luxsonor, that do not have the resources to develop the necessary code for their hardware to be supported on every platform. For example, Zoran Corporation provided documentation for one of their DVD decoder chips so that the community could provide support for the Zoran chips in Linux systems, because Zoran did not have the resources to develop it themselves. The Zoran chip is found on an upgrade board for Matrox Graphics, Inc.'s video cards. Matrox made available documentation to further accelerate DVD playback on their video cards, by releasing the low level documentation on their video card line.
Dated: Dallas, Texas
June 14, 2000
_____________________________________
MATTHEW PAVLOVICH
[Reply Declaration of Emmanuel Goldstein]
Martin Garbus, Esq. (MG 6261)
Edward Hernstadt, Esq. (EH 9569)
FRANKFURT, GARBUS, KLEIN & SELZ, P.C.
488 Madison Avenue
New York, New York 10022
(212) 826-5582
Attorneys for Defendant Eric Corley a/k/a
EMMANUEL GOLDSTEIN
| UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK -----------------------------------------------------------------------X
UNIVERSAL CITY STUDIOS, INC., - against -
ERIC CORLEY a/k/a "EMMANUEL GOLDSTEIN"
|
00 Civ. 0277 (LAK)
REPLY DECLARATION OF EMMANUEL GOLDSTEIN IN SUPPORT OF DEFENDANTS' CROSS-MOTION TO VACATE THE PRELIMINARY INJUNCTION |
I, EMMANUEL GOLDSTEIN, declare under penalty of perjury that the foregoing is true and correct:
1. As stated in my previous Declaration submitted in support of this motion, 2600 is a quarterly journal with a circulation of approximately 4000 subscribers and 60,000 additional newsstand readers. Our readers range from teenagers to college professors, corporate executives, universities, and intelligence agencies. By reporting on issues concerning the modern, technical era, including internet issues, 2600, which now includes the 2600.com web site, has grown as the public's knowledge of, and fascination with, the internet and all things computer-related has grown. The web site is an on-line supplement to the content published in the print magazine, and does not exist to distribute illegal code.
2. As I've stated previously, 2600.com posted DeCSS, and currently links to DeCSS mirror sites, because I recognize the importance of such a program to a variety of disciplines, including (i) the ongoing efforts to reverse-engineer an open-source DVD player, (ii) cryptography and the science of encryption, and, most importantly, (iii) the legal fair use of materials contained in movies released on DVD.
3. Neither 2600.com nor 2600 has, at any time before or after the issuance of the preliminary injunction in this case, in connection with the posting of DeCSS, linking to mirror sites or otherwise, ever advocated, suggested, or otherwise encouraged the use of DeCSS, or the illegal copying or distribution of DVDs. Annexed hereto as Exhibit A are printouts of each of the stories that 2600/2600.com has published concerning DeCSS. I invite plaintiffs and the Court to read each story carefully to confirm that, indeed, no copying or other "unlawful" activity is advocated.
4. In particular, we have, and continue to state to interested parties that mirroring or linking to DeCSS mirrors is not about copying DVDs, but rather about the right to freedom of information and about educating people about technology. For example, 2600.com's DeCSS mirror page has, since the issuance of the preliminary injunction, stated in part as follows:
While we have every intention of sticking this out to the end, we have to face the possibility that we could be forced into submission. For that reason, it's especially important that as many of you as possible, all throughout the world, take a stand and mirror these files. Don't do this because you just want to copy DVD's - that's not what this fight is about at all. This is about freedom of information - the right we all still have to LEARN how technology works. . . .
A copy of the 2600.com mirror page is annexed hereto as Exhibit B.
5. As I stated in my previous declaration in support of this motion, the sites containing DeCSS mirrors to which we currently link are very diverse in nature. In particular, many of these sites provide a variety of legitimate information concerning topics related to this case and DeCSS, and are thus more than just mirrors setting forth the DeCSS code. Annexed hereto as Exhibit C are copies of the first pages of the mirror sites currently listed at 2600.com.
Dated: June 14, 2000
New York, New York
____________________________
HTML by Cryptome.