Michael B. SaphersteinfnA
October 28, 1997
Intelligent agents, sophisticated computer programs that act on behalf of their users and adjust themselves to users' behaviors and preferences, may answer the prayers of people who are increasingly overwhelmed by the sheer volume of information available to them on the Internet.2 Instead of spending frustrating hours "surfing the 'Net" in search of elusive information, users may soon employ intelligent software agents that gather information efficiently and without need for further human assistance, thereby freeing the user to spend time on more productive, or more leisurely, activities.3 Still in its infancy, agent software "launches" itself into a computer system, a local-area-network, or the Internet, in order to perform a task or set of tasks requested by the user, such as retrieving information on a particular company or purchasing plane tickets at the lowest price.4 If sufficiently sophisticated, intelligent agents may be able to "negotiate" with software agents resident in other computer systems, to coordinate, for example, a teleconference between two executives which does not conflict with the schedule of either person.5
Although agent technology is neither ubiquitous nor yet fully realized, it is nearly certain to play a major role in the way consumers and businesspeople deal with information in the years to come.6 Indeed, agent software may cause people to rethink their entire approach to receiving information, creating distinctions between mere data and true information.7 Don Norman, a research fellow at Apple Computer Inc., goes even further, arguing that we do not want data or information - we want knowledge.8 Norman and others believe that intelligent agents will perform the tedious, time-intensive tasks of data collection and information retrieval, ultimately providing the user with readily-applicable knowledge and perhaps rendering obsolete presently "state-of-the-art" search paradigms, such as the unwieldy World Wide Web.9
Not surprisingly, any technology that promises to change fundamentally the way information is gathered will raise the eyebrows of the artistically or intellectually creative individuals (and their lawyers) who possess legally granted ownership rights in expressions of this information.10 To be specific, intelligent agents may retrieve for users, via the Internet, documents, images, video or sound files that are the copyrighted material of another, thus subjecting the user, and/or the user's employer, to potential legal liability.11 This seemingly inevitable clash between copyright holders and the users of the Internet, compounded by the injection of intelligent agent technology, presents several novel copyright issues.
The core issue, and the issue in which the federal government, industry, academia, and individual computer users are all keenly interested at present, is when, or perhaps whether, the viewing, retrieval, transmission, or transfer of copyrighted files, data or images by users of the Internet constitutes copyright infringement.12 This issue, until definitively resolved by the courts or legislature, will almost surely lie at the crux of numerous Internet-related copyright lawsuits that will be filed in the months and years to come.13 Therefore, this paper must, at the outset, explore in depth copyright protection for files on the Internet.
Only after addressing this fundamental issue can we delve into our more creative investigation; that is, whether the retrieval of copyrighted computer files via the Internet by an intelligent software agent can be considered infringement by its owner, operator or initiator. Several ancillary issues will also be explored. For example, would the user's employer, whether a corporation or academic institution, also be exposed to liability for copyright violations, under theories of contributory or vicarious liability? Additionally, would the company who designed, programmed, or owns the copyright to the source and object code for the intelligent agent software face potential liability when its software agents infringe upon copyrights?
The Fundamental Issue - Copyright and Users of the Internet
This paper is not intended to provide an exhaustive treatment of the uncertain legal implications of copyright law on the Internet. Instead, a more confined legal and policy analysis of the copyright exposure of Internet users will be undertaken in order to set the stage for the subsequent discussion of the copyright implications of intelligent agent software in the Internet context.
Although laypeople often mistakenly interchange the World Wide Web, the newest, fastest-growing, multimedia-ready subset of the Internet, with the Internet itself, the Internet is actually an interlinking of thousands of different sizes and types of networks from all over the globe, originally established by the Department of Defense (as "Arpanet").14 The Internet is composed not only of the World Wide Web, but also of search engines and facilities including FTP, Usenet, Gopher, WAIS, Telnet, and E-mail.15 These functional elements of the Internet enable the user to download, upload, transfer, and view files, messages, data, images, sounds, and video.16
The early culture of the Internet, a culture which remains steadfast in many users today, stressed that the Internet is a forum for the free exchange of information, unhindered by the traditional legal and practical impediments inherent in the non-digital world.17 Technology visionary John Perry Barlow, a co-founder and chairman of the Electronic Frontier Foundation, argues that "intellectual property law cannot be patched, retrofitted, or expanded to contain digitized expression any more than real estate law might be revised to cover the allocation of broadcasting spectrum."18 Indeed, opponents of regulation of the information highway argue that a hands-off approach by the government in terms of copyright protection may be more profitable in the long run, because it will attract more writers and artists to enter the electronic marketplace with the lure of more users eager to access material that is unhampered by usage restrictions.19
Conversely, authors, copyright holders, and groups advocating on their behalf, vigorously argue that just because the Internet allows users to transfer, view, and copy files easily, valuable copyright protections entitled to authors should not become nullities, particularly in light of the ascendancy of the Internet as a legitimate publishing realm.20 The legal, computer, and publishing worlds are scrambling to formulate and anticipate the strategies that authors will employ in order to establish a royalty system by which they, as copyright holders, may collect fees. Such fees may come either from large-scale Internet Service Providers (ISPs) who provide users with the ability to link to the Internet or from individual users of the Internet who download, transfer, and view copyrighted works.21
Although the common law has not yet addressed the copyright implications of the Internet with any appreciable authority, the federal copyright statute (Title 17 U.S.C.), in conjunction with existing copyright case law, can serve as useful guides in predicting the development of Internet-related law. Unless stated otherwise, we will assume, for purposes of this analysis, that the potential plaintiffs in Internet-related copyright infringement lawsuits are the proper copyright holders of the software at issue. That is, based on § 102(a) of the Copyright Act, plaintiffs have "produced an original work of authorship fixed in a tangible medium of expression, which can be perceived either directly or by the aid of a computer (machine)."22 Although courts have indicated that a certificate of registration from the Copyright Office is prima facie proof that a claimed copyright is valid and that the registrant is its legitimate owner, authors, particularly of files created exclusively for Internet publication, often do not register their works with the Copyright Office.23 This does not prevent these authors from holding a valid copyright in their original works, as long as the statutory requirements of § 102(a) are met and the statutory restrictions of § 102(b), discussed below, are absent.24
Of course, copyright protection has never accorded the copyright owner complete control over all possible uses of his work.25 Rather, the Copyright Act grants the copyright holder "exclusive" rights to use and to authorize the use of his or her work in five qualified ways enumerated in § 106 of the Act, most notably for our purposes, reproduction of the copyrighted work in copies.26 It is these enumerated, exclusive rights from which a copyright infringement claim under § 501 derives; that is, an individual has infringed upon one or more of the copyright holder's exclusive rights.27 So, assuming that an author holds a valid copyright in material contained in a software file residing on an Internet-accessible server, he or she must then prove the putative infringer unlawfully appropriated protected portions or the entirety of the copyrighted work.28
Does the Internet User's Copying of the Author's Work Constitute Infringement of the Author's Copyright?
In copyright infringement cases, the plaintiff must prove similarity of his or her work to that of the defendant in two ways and for two distinct reasons. First, the plaintiff must prove that the defendant, as a factual matter, had access to plaintiff's work and thus, based on proofs submitted into evidence, actually copied portions of this work - a process of proof which has been termed "probative similarity."29 Second, the plaintiff must prove whether, as a mixed issue of fact and law, those elements of the work that have been copied are protected expression and of such importance to the copied work that the appropriation is actionable - i.e., the "substantial similarity" analysis.30
Assuming that the court finds the defendant's and plaintiff's works to be sufficiently similar to warrant further inquiry into copying, the plaintiff ordinarily must also prove based on a preponderance of circumstantial or direct evidence that the defendant had access to the plaintiff's work.31 Without access, there could not logically be a copying, no matter how similar the two works appear.32 In cases of infringement of music, such as Selle v. Gibb or Bright Tunes Music Corp. v. Harrisongs Music, Ltd.,33 the proofs of actual similarity and access may become quite involved and require expert testimony.34
In proving actual copying and probative similarity in cases of copying off the Internet, the plaintiff has comparatively little to worry about.35 In Internet cases, the defendant will have ordinarily "copied," via computer, the exact software file that the author has created. Therefore, the factual question of probative copying will most often be answered in the affirmative; indeed the defendant's "work" will most likely bear more than a probative or even "striking" similarity to plaintiff's, it will bear exact identity.36 In the software case of Atari Games Corp. v. Nintendo of America, Inc., the Court of Appeals for the Federal Circuit commented that "[e]ven for works warranting little copyright protection, verbatim copying is infringement."37 This statement implies two things: first, that the plaintiff in an Internet case will have no difficulty in proving probative similarity. Second, and more profoundly, the Atari courts comment hints that the more difficult "substantial similarity" test may be reduced, as we will see, to a virtual nullity in cases of verbatim software copying.38
Ordinarily, the second inquiry of the unlawful appropriation test, determining by a "substantial similarity" whether the elements copied are protected expression, derives from § 102(b) of the Copyright Act, which provides that "[i]n no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work."39 The 1976 House Report on the Copyright Act affirmed that copyright protection does not preclude others from using the ideas or information inherent in the author's work; rather, copyright protects only the expression of the work adopted by the author.40
The less than clear-cut process of distinguishing unprotectable ideas or methods of operation from protectable expression in software is, not surprisingly, a significant unresolved issue in the continuing struggle to apply copyright law to the fast-moving target of computer software.41 In many contemporary software copyright cases such as Computer Associates v. Altai, Apple v. Microsoft, and Lotus v. Borland, defendants were not primarily accused of copying the actual software files (source or object code) owned by the plaintiffs; rather, they were accused of copying "nonliteral" aspects of the plaintiff's original expression, e.g., the particular on-screen menu structure or user interface generated by executing the plaintiff's object code in a computer.42 In each case, defendants relied, at least in part, on § 102(b) of the Copyright Act to argue that the plaintiff's work of authorship was a method of operation, idea or process exempt from copyright protection, thereby demonstrating that the defendant's own work of authorship did not constitute an infringement of copyright.43
Unfortunately for potential plaintiffs in Internet-related copyright cases, the § 102(b) argument is less valuable, because courts seem reluctant to embark on an in depth filtration-abstraction-comparison analysis44 in order to address the issue of substantial similarity in cases of literal copying.45 Instead, the Second Circuit in Altai, foreshadowing the sentiment of the Federal Circuit in the Atari case decided just two months later, flatly stated that "[i]t is now well settled that the literal elements of computer programs, i.e., their source and object codes, are the subject of copyright protection.46 Future courts may be confronted with defendants who admittedly have copied from the Internet the literal object code (rather than nonliteral aspects of the plaintiffs work) owned by the plaintiff. The plaintiff in that case will assert, and rightly so, that his or her exclusive right under §§ 102(a) and 106 of the Copyright Act to reproduce the software as a whole literary work has been usurped by the defendant's infringement.47 Because defendant-users of the Internet must, in honesty, admit to copying verbatim the files of plaintiffs, courts will be hard-pressed to find for defendants on grounds of either probative or substantial similarity, even if some of the plaintiff's software were to contain unprotectable ideas or methods of operation.48
The strongest § 102(b) argument that a defendant could proffer would probably be that the plaintiff's file is entirely in the public domain and thus does not merit copyright protection; i.e., that the files are composed of expressions of ideas or expressions too old to retain copyright protection, that they are composed entirely of ideas or facts with virtually no original expression overlaying this information, or that the potential range of expression is so narrow as to preclude copyright.49 Although it seems unlikely that the plaintiff would have injected no original expression alongside the public domain facts or ideas in the file that he or she created (unless the file is expressly intended to be in the public domain, in which case the author would not be suing), the Tenth Circuit does leave the door slightly ajar for defendants to argue under § 102(b).50 In Gates Rubber Co. v. Bando Chemical Industries, Ltd., the 10th Circuit Court of Appeals cites to § 102(b), stating that "[s]ource and object code, which are the literal elements of a program, will almost always be found to be protectable expression [rather than the nonprotectable idea itself] unless the doctrines of merger and scenes a faire come into play."51
The merger doctrine states that when the possible ways to express an idea are limited, the expression "merges" with the idea and is therefore uncopyrightable; when merger occurs, identical copying may be permitted.52 Copying of this merged idea/expression is not barred by copyright law, because protecting such expression would, in essence, confer a patent-like monopoly of the idea upon the copyright owner free of the conditions and limitations imposed by patent law.53 Similarly, the doctrine of scenes a faire ("scenes which must be done"54) removes protection from both the elements of a work necessary to perform a task and elements which are obvious, and has been applied to software.55
Even if, as the Tenth Circuit in Gates Rubber Co. implies, defendants can employ either merger or scenes a faire arguments in copyright infringement cases involving the Internet, these defenses are viable only in the software development arena where courts may be concerned that the plaintiff's work of authorship, if granted copyright protection, would confer upon him or her a de facto monopoly in an idea essential to efficient software development.56 That is, these defenses are employed ordinarily in cases where the defendant has created a new, arguably original, work of software authorship, based in part on the ideas of the plaintiff, which will provide market competition against the plaintiff's software product.57 Courts are understandably reluctant to confer upon plaintiffs copyright monopolies, particularly when doing so significantly retards the Constitutionally encouraged progress of science and the useful arts as well as stifling the free market.58
Users of the Internet, however, face a situation which seems difficult to reconcile with the merger or scenes a faire exceptions. First, many, if not most users of the Internet, when downloading, viewing, or transferring potentially copyrighted files, generally do so with no intention of creating a new work of authorship they intend to bring to market.59 The copyright implications of intelligent agents, explored infra, will emerge in the situations where potentially copyrighted images, programs, or other files are downloaded for their intrinsic educational or entertainment value.
In fact, most users "merely" seek to download software off the Internet for personal, educational or entertainment purposes. This noncreative, albeit common, use in itself frustrates merger or scenes a faire defenses, because in typical cases where courts accept the merger defense, defendants have created a work of authorship competitive with that of the plaintiff, and which the plaintiff seeks to bar from the marketplace of products or ideas.60
Second, in recent software infringement cases such as Altai, courts have accepted these defenses only where it was demonstrated that defendants endeavored to create software sharing common functional elements with plaintiffs, without copying source or object code.61 Where, as in the Internet scenario, a defendant copies an entire software file which, we are assuming, is the original creation of the plaintiff,62 it seems highly unlikely that either the merger or scenes a faire defense would be accepted by a court.
As for the Tenth Circuit's statement in Morrissey v. Procter & Gamble Co. that identical copying is permitted in situations of merger, the particular situation on which the court was commenting was non-software-related; a defendant appropriated a set of copyrighted rules for a sweepstakes.63 The court found these rules so straightforward and simple as to require application of the merger doctrine, because there was available at best a very limited number of forms of expression of the rules.64 Based on the three decades of copyright law since Morrissey, particularly in the software arena, it is all but certain that, absent affirmative defenses (particularly the fair use exception we will explore shortly), no court would accept defenses of merger or scenes a faire to absolve a defendant of liability for identical copying, particularly via download, of an entire copyrighted software file.65
"I Didn't Do Anything Wrong, Did I?" - Ignorance is No Excuse
Users of the Internet who hope to escape legal liability for viewing or downloading copyrighted materials might believe that a sincere defense of ignorance would absolve them of liability. If they asserted this defense in court, these well-meaning souls would discover to their dismay that this defense is unavailable in cases of copyright infringement.66 No requirement of scienter or infringing intent is required by the Copyright Act or has been inferred by the courts in finding innocent infringers guilty of copyright infringement and liable for damages to a copyright holder.67 A finding of copyright infringement requires simply that the infringer had access to the copyright holder's work, and copied the work such that the resulting work is probatively and substantially similar to the holder's.68 Indeed, the Southern District Court of New York in the Harrisongs Music case reasoned that so long as the two works pass the similarity tests, even "unconscious" infringement is sufficient to impose liability on the author of the infringing work.69 George Harrison, ironically, likely "sealed his own fate" in terms of copyright liability by admitting, to the agreement of other music experts, the substantial similarity between his work and the plaintiff's.70
The additional component critical to the Southern District Court of New York's finding of infringement was Harrison's admission that he had heard, and subconsciously remembered fondly, the copyright holder's work years before recording his own song, and thus composed his song entirely, albeit unknowingly, based on the original.71 This proof of access, though lacking in malicious or even competitive intent, satisfied the original composer's burden in proving Harrison's copyright infringement and ultimate liability.72 Users of the Internet cannot, then, absolve their own liability by averring that in downloading or viewing the plaintiff's computer file, they did not knowingly or intentionally infringe upon a copyright holder's exclusive right to reproduce his or her work.73
Certainly, the Copyright Act provides that statutory damages in infringement cases may be reduced when the "court finds that such infringer was not aware and had no reason to believe that his or her acts constituted an infringement of copyright."74 This allowance would be of little comfort to innocent users of the Internet after spending the time, expense, and anguish in defending an infringement suit, only to lose the case and face injunctions, the impoundment of computer hardware and media, destruction of the reproductions of the plaintiff's work made in violation of his rights, a recovery of (reduced) damages, and potentially attorney fees.75
Fair Use - The Users' Saving Grace?
Based on our analysis thus far, it is clear that many users of the Internet face potential copyright liability given their countless reproductions, whether by viewing or downloading, of copyrighted software files. However, the "saving grace," so to speak, for the millions of Internet users quivering with fear after reading this paper thus far, may be § 107 of the Copyright Act - the equitable "fair use" provision limiting the copyright holders exclusive rights.76 Section 107 allows reproduction in copies of a copyrighted work for purposes such as "criticism, comment, news reporting, teaching, . . . scholarship, or research."77 Moreover, other uses may be "fair" based on the courts consideration of four factors: "(1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes; (2) the nature of the copyrighted work; (3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and (4) the effect of the use upon the potential market for or value of the copyrighted work."78
The fair use case most favorable to the copyright plight of Internet users may be Sony v. Universal Pictures, in which the U.S. Supreme Court ruled that Sony's sale of videocassette recorders enabling users to "time-shift" the personal viewing of television programs did not constitute a contributory infringement by Sony of the exclusive rights of the copyright holders of selected television programs.79 It must be noted that the Second Circuit's decision in American Geophysical Union v. Texaco Inc. presents a competing fair use analysis founded on the presumption that unless the defendant shows a transformative use of the original work, such that a new contribution of intellectual value is added to the marketplace, the court will more readily rule that the defendant is not protected by the fair use defense.80 Nevertheless, the fact situation of an individual Internet user copying computer files for personal use is highly dissimilar to the Texaco situation of an employee of a large commercial actor photocopying copyrighted materials for a business purpose.81 The individual Internet user is, instead, much more akin to the television viewers considered by the Sony court, and thus the Sony analysis seems more applicable as well as equitable.82
The Supreme Court paid particular attention to the first and fourth factors of Sony's fair use defense. The Court agreed with Sony that the purpose of their customers' use of the copyrighted material - i.e., commercial TV shows - was not of a commercial or educational nature, but it was for the nonprofit entertainment purpose of watching TV shows at a time more convenient for the viewer.83 Based on the viewers' lack of commercial purpose in time-shifting, factor one favored the defendants.
Quickly discussing the second and third factors, the Sony court concluded that when one considers the nature of a televised copyrighted audiovisual work84, and that time-shifting merely enables a viewer to see such a work which he had been "invited" to witness in its entirety free of charge, the fact that the entire work is reproduced85 does not have its ordinary effect of militating against a finding of fair use.86
In analyzing the critical fourth factor, the Supreme Court also sided with Sony. The Court agreed with the District Courts finding that because time-shifting enlarges the television viewing audience, a significant amount of television programming may be used in this manner without objection from the owners of the copyrights on the programs.87 The Court noted that although copying for noncommercial purposes may impair the copyright holder's ability to obtain the rewards that Congress intended him or her to have, a use that has no demonstrable effect upon the potential market for, or the value of, the copyrighted work need not be prohibited in order to protect the author's incentive to create.88 The Court concluded that the prohibition of such noncommercial uses would merely inhibit access to ideas without any countervailing benefit.89
So, although the Supreme Court absolved Sony, the manufacturer of copying technology, of contributory liability, the Court's reasoning is applicable to both the potential direct liability of the Internet user and the contributory liability of those providing the means by which users may copy files via the Internet. Indeed, the four-stage fair use analysis may well be framed along similar lines to those drawn by the Sony defendants.90 The major difference, we will see, is that, unlike the Sony situation, where viewers copied a legitimate broadcast authorized by copyright holders on the TV shows, Internet users often download materials such as photographs or sound clips for which copyright holders have given content providers no permission whatsoever to post on the Internet. This difference severely complicates the users' fair use defense.
In terms of factor one, the purpose of most Internet users use of the copyrighted material on the Net is generally not of a commercial nature, but ordinarily is for entertainment or educational purposes.91 Therefore, factor one favors Internet users even more strongly than it does the defendants in Sony, because it is arguable that legitimate educational use of the Internet is more prevalent than that of network television.92
The nature of copyrighted work on the Internet is admittedly difficult to classify - some works are highly expressive and original, while others are composed largely of uncopyrightable facts. And, as noted above, the most legally tricky material on the Internet, for purposes of the fair use defense, may well be data, particularly digitized images or sounds posted on Web pages or FTP/Gopher sites for users to download freely, that content providers have converted from copyrighted media such as magazines or TV shows without the copyright holders' permission. In these cases, we will assume that the fair use defense is unavailable to the content provider who violated the copyright holder's rights by converting, for example, copyrighted photographs of Cindy Crawford or sound clips of Bart Simpson into Internet-ready computer files. Assuming that the copyright holder of the original photos or sounds never intended them to be available on the Internet, and that these are the people most likely to bring Internet-related lawsuits, factor two appears to fall in favor of the plaintiff. 93
As in Sony, the Internet context also represents a situation in which the entire work is reproduced; therefore the amount and substantiality of the portion of the copyrighted work used in relation to the copyrighted work as a whole is a one-to-one relationship.94 However, the Sony court reasoned that because viewers had been invited to witness the entire work free of charge at the time it was originally televised, copying the entire work would not harm the author. 95 Here, we must distinguish between copyright holders who intentionally post their works on the Internet and those who have their work reproduced without permission from magazines, TV or other media, via scanner, video or frame capture hardware, or digital recording equipment, and posted for users freely to download from the Internet. The copyright holders who post their own works on the Internet in downloadable form would likely be viewed by a court as inviting Net users to view and download the material for their own private, noncommercial use free of charge. In that case, the Sony analysis would push the third factor in favor of the defendants.
However, the copyright holders whose works were copied and digitized without permission likely had no intention of "inviting" Internet users to view or download their images, sounds, full-motion-videos, data or software files. Although Internet users might not know that these files were posted without the authors permission, we have already seen that ignorance is no defense in copyright cases, and therefore these users probably should have thought twice before downloading. So, with these aggrieved copyright holders as plaintiffs, factor three would lie in favor of them.
In cases involving the Internet, factor four will prove critical in the court's decision as to whether defendants should face infringement liability. This is also the most difficult inquiry, because it requires the court to estimate the effect of the download, transfer, or viewing of an Internet-available file on the potential market for or value of the copyrighted work, a valuation that will be extremely difficult to estimate given the vastness and lack of centralized knowledge of the Internet.96 Unlike time-shifting, which arguably enlarges the television viewing audience, allowing 'Net-surfers to download and transfer files97 does not per se enlarge the audience of Internet users. The question then emerges, what do copyright holders of information want? Again, the analysis diverges depending on the nature of the plaintiff.
If the plaintiff is an author who has voluntarily posted his or her work on the Internet, then these copyright holders must want as many users as possible to download their works of authorship, and thus do not want these users to be legally liable upon doing so. Obviously, in deciding to post their works on the 'Net, these authors felt that the value of public exposure outweighed the cost of freely distributing their work to Internet users.98 Authors who have voluntarily posted copyrighted material to the Internet are, however, unlikely plaintiffs.
Instead, future plaintiffs will likely be authors who did not intend their works to be downloadable off the Internet. The copyright holder in a Cindy Crawford photograph or in a Bart Simpson video clip will argue that the effect of the free availability of their authorship will reduce both the potential market for and value of the work.99 An individual defendant might counterargue that his or her solitary download of an admittedly unauthorized copy would do little to decrease the market. In response, plaintiffs would point out that the Internet is composed of millions of individual users, each performing "solitary" downloads. These downloads, in the aggregate, may well have a material impact on the market for or value of a copyrighted work.100
Defendants may alternatively argue that photographs converted to computer form do not decrease the value of the original, because computer images are of lower quality, in terms of resolution and clarity, than those in a hard copy magazine. Therefore, the people who would have bought a magazine for a photograph of Cindy Crawford would not decline to do so because her photo was available on the computer. In fact, defendants might argue that Cindy Crawford photos on the Internet, albeit unauthorized, would stimulate sales of magazines featuring photos of the supermodel, just as time-shifting in Sony increased the TV viewing audience.101 Based solely on this Cindy Crawford example, factor four appears to lie slightly in favor of defendants. Likewise, the overall result of the fair use defense is very close, perhaps also slightly favoring defendants.102
However, the above factor four hypothetical should be confined to its facts, because, unlike copyright holders in photographs, copyright holders of retail software products, such as Microsoft® WindowsTM 95 may be profoundly economically damaged should their products be posted to the Internet without authorization.103 In cases such as this, if Internet users were to argue fair use as a defense to downloading software, factor four would certainly favor the plaintiffs, the fair use defense as a whole would likely fail, and plaintiffs ultimately would be entitled to damages.104
Although the above hypotheticals assumed the Internet users as defendants, the companies who provide access to the Internet for home users as well as personal computer manufacturers themselves would likely be joined as defendants in such cases. As with the Sony Corporation in the Sony case, the contributory liability of these larger actors may depend on the success or failure of the consumers' fair use arguments.105 Based on the Supreme Courts reasoning, it seems more likely that an Internet Service Provider would be held contributorily liable than would a PC manufacturer such as Apple, IBM, or Compaq. Personal computer manufacturers, like Sony in the Betamax context, are not contributorily liable if their products are capable of substantial noninfringing uses. 106 Needless to say, personal computers are capable of substantial noninfringing uses. It remains to be seen, however, whether Internet service providers will face contributory liability, or if copyright issues concerning the Internet will be resolved sufficiently to avert a judicial conflict.
The Sony decision portends unpredictable outcomes in future copyright lawsuits involving the Internet, particularly given Justice Stevens's observation that because the fair use doctrine is an equitable rule of reason, no generally applicable definition is possible, and thus each case raising the question must be decided on its own facts.107 Courts hearing an Internet-related copyright case in which defendants invoke a fair use argument will have to balance two valid positions, the prevailing Internet culture of the easy, free exchange of ideas and data against the concern of copyright holders that if copyright protections are not maintained in the cyberpublishing world, then authors will eventually refuse to produce high quality works for this vitally important new medium.108
It seems fairly certain that the resolution best protecting the rights of authors publishing on the Internet along with the freedoms of the individual 'Net surfer will not be judicially created in a piecemeal fashion; rather, a combination of legislative, industry, and consumer groups must convene to resolve this important issue with uniformity.109 We will not predict the resolution here. Instead, we will attempt to further "complicate" this unresolved quagmire of copyright and the Internet!
Intelligent Agents - The Next Copyright Quandary?
Although most computer-literate people are familiar with the Internet and "surf" from time to time, relatively few would be able to describe intelligent agents, let alone claim to use them. This is quickly changing. Consumer-level PC-based software has already begun to incorporate some of the agent technology discussed earlier. For example, Quarterdeck Corporations WebCompassTM (currently in Version 2.0), an add-on for existing web browsers, is billed as "the first search agent with built-in intelligence . . . like having a personal assistant who spends 24 hours a day, 7 days a week, surfing the Net for you, bringing back just the good stuff."110 Described as such, WebCompassTM sounds strikingly similar to the prototypical intelligent agent described in the Introduction to this paper. However, delving beneath the marketing hype reveals that WebCompassTM does not actually "surf the net" or copy files from Internet servers.111 Instead, it automatically and periodically accesses web search engines such as Yahoo, Lycos or WebCrawler and returns to the user the descriptions and URL addresses of web pages which are generated in response to the user's search request.112 While revolutionary in its own right, WebCompassTM, in its present form, does not possess the software functionality to copy and deliver actual files to the user. Therefore, current versions of WebCompassTM do not hold significant implications for copyright infringement.
Quarterdeck's product release undoubtedly heralds agent technology which will retrieve actual files for the user. Until that time arrives, however, we must continue our analysis on a more theoretical plane. Therefore, for the remainder of this analysis, we will refer to the "imaginary" intelligent agent software able to automatically surf the Web and retrieve actual files for the user as "SuperWebCompass."113
The first question we must explore is whether the retrieval of computer files by SuperWebCompass would constitute breach of copyright. To answer this question, we must discern the differences, if any, between a user downloading files from the Internet and an intelligent agent doing so on the user's behalf. Users are able to read if copyright notices or warnings are posted on screens adjacent to files to be downloaded, and then proceed based on, or in contravention of, these notices. But, as discussed earlier, most software files and images posted on the Internet do not have attached copyright notices. More to the point, many home page content providers actually convert and post without permission the copyrighted authorship of others, e.g., Star Trek® or The Simpsons® images, digitized photos of supermodels, or copyrighted software programs, for the "benefit" of their fellow Netizens. Everyone who surfs the Web to any appreciable extent knows, or at least suspects, that many home pages are probably assembled without the blessings of respective copyright holders. Further, in presenting these files in downloadable form, home page authors are encouraging further infringements of copyrights which may already have been infringed multiple times through successive generations of file reproductions and transfers.114 Millions of times per day Web surfers download these images or files for their own use, but these users are consciously choosing to do so with the knowledge that they may be infringing the rights of the true copyright holder. Assuming that downloading an image that may be copyrighted does not fall within the fair use exception, users who do so may be analogized to speeders on the highway - although many speed, the few who are caught cannot argue that they were unaware that the speed limit applied to them.
This analogy is more troubling, however, in a situation where a user is sued for copying one or more files actually retrieved by SuperWebCompass. As we know, innocent infringement is still infringement, per Harrisongs, but in this case it would not be the user who performed the copying at all.115 The likely scenario would be as such: in the morning a user would provide a general search query to SuperWebCompass, which would proceed to launch itself throughout the day into the Net to search for files meeting the users criteria. Upon finding matching files, images, etc., SuperWebCompass would "return" to the users computer and copy the files to the users hard drive. The only input of the user into the actions of the agent was the initial generalized query. When the user sits in front of his or her computer at the end of the day, and discovers that SuperWebCompass delivered several screen captures from a copyrighted TV show, for instance, he or she then has a problem. A copyright infringement has been committed. The user may delete the offending material from his or her hard drive, but the infringement has already occurred (upon the download) and the user may be liable, perhaps doubly, if the copy to RAM also constitutes an infringement.116 It appears highly doubtful that any court would conceptually divorce a computer user from his or her intelligent agent (until and unless intelligent agents truly live up to their moniker) when assessing copyright liability. Therefore a user will not be able to cry, "my agent did it" and hope to avoid copyright liability.
Moreover, based on our line of discussion earlier, it does not appear that the unfortunate user facing copyright liability as a result of SuperWebCompasss actions can rely with any more certainty on a fair use defense than can his counterpart in the days prior to agent technology.117 The same arguments with respect to factors two, three, and four explored above will militate in favor of a plaintiff suing a user who infringed upon the holders copyright by means of intelligent agent technology.118
More disturbingly, under theories of contributory infringement and vicarious liability, Internet service providers, future developers of intelligent agents, as well as the users employer or school, may all be implicated as joint defendants in situations where intelligent agent technology is used to copy protected material.119 Courts already have acknowledged that the doctrines of contributory and vicarious liability can be applied to cases involving computer bulletin board services.120 Contributory infringement has been held as to bulletin board operators based on the their role of providing facilities, direction and knowledge for the copying of a plaintiff's copyrighted software.121 Future plaintiffs might rely on these precedents of BBS systems operator liability in order to persuade courts to extend contributory liability to schools or businesses who provide computers, Internet access, and agent software to students or employees who commit copyright infringements.122
With perhaps less legal difficulty than in the contributory liability scenario, plaintiffs might also subject schools and businesses to vicarious liability in the Internet/intelligent agent context. Schools and business will be subject to vicarious liability if they have the right and ability to supervise the infringing acts of their students or employees, as well as a financial interest in the exploitation of the copyrighted materials.123 In copyright law, there is no requirement that the direct infringer be acting within the scope of employment for vicarious liability to attach.124 In a case involving copyright infringement of music in a dance hall setting, the U.S. District Court for Massachusetts applied a low standard of proof to satisfy the criteria for vicarious liability.125} The court indicated that the plaintiff need only prove that the copyrighted pieces of music were performed at a place that was open to the public, operated for profit and under the control and management of the defendant.126 As opposed to cases where plaintiffs attempt to impose contributory liability, the Massachusetts District Court declared that the dance hall owner would be vicariously liable even if he or she had no knowledge of the infringement and even had expressly warned the band not to perform copyrighted works.127 Such a case clearly exposes businesses, and to a slightly lesser extent, schools, to vicarious liability if employees or students are caught on school or business premises copying, either by way of intelligent agents or using traditional techniques, protected works off the Internet.128
An interesting question is whether developers and manufacturers of SuperWebCompass-type agent software might also be liable under theories of vicarious or contributory liability. The U.S. Supreme Court ruled that Sony's Betamax machines were capable of a substantial noninfringing use - authorized noncommercial private time-shifting. 129 Conversely, agent software may well be seen by a court as designed purely with the goal of copying software from the Internet for users, and thus incapable of a substantial noninfringing use.130 If the economic fair use analysis of the Internet discussed earlier is plausible, and agent software is found to be incapable of noninfringing uses, then developers like Quarterdeck may well face contributory or vicarious liability.131
Of course, a court may choose to absolve one or more of these ancillary defendants of infringement liability, but it is clear that a court-by-court judicial remedy for the copyright implications of intelligent agent technology will be inadequate to delineate the rights and responsibilities of all parties involved in some capacity with the Internet.
Does Agent Technology Itself Hold the Solution to the Copyright Quandary?
An obvious question, and perhaps at its core the seed for a solution, is how a copyright holder would know that a user through intelligent agent technology infringed his or her copyright. The hypothetical lawsuit conceived above is only possible if software contained some sort of copyright encoding which identified its source and likewise identified to the copyright holder whenever an unauthorized reproduction was made. Perhaps naively, we may assume for sake of argument that authors of home pages who "wrongfully" digitize copyrighted materials would not be able to encode these files with valid copyright ownership codes. Assuming as such, a potential scheme for copyright protection might be that every file on the Internet must be encoded with the copyright owners electronic signature, and in the case of public domain material, files would be encoded to reflect that status as well.132 Then, when software such as SuperWebCompass attempted to transfer the file to the users machine, upon "intelligently sensing" the electronic copyright ownership code, the agent would either automatically charge a "copyright debit account" of the user which is reflected in the user's monthly service bill, or the agent would electronically notify the copyright holder's account in order for royalty payments to be debited to the service provider's account. Or, if customers do not wish to retrieve copyrighted files or pay extra monthly fees, SuperWebCompass might be encoded with a "directive" to avoid retrieving or copying any non-public domain files.
One real-world solution has recently been developed by BMI, a licensing organization which ensures that individuals and companies comply with music copyrights and associated royalties. BMI has developed a "Web robot" (akin to an agent) to monitor transmission and sales of music on the Internet.133 With this technology, BMI is hoping to find potential copyright violations on the Internet. BMI estimates that 26,000 Web sites contain music, and says only a handful have licensing agreements with the agency, which represents more than 180,000 songwriters and song publishers.134
Of course, technology-based solutions such as BMI's are often relatively unsophisticated and thus bypassable by computer experts. Nonetheless, such solutions represent noteworthy preliminary attempts to enforce copyrights, often to the outrage of consumers.135 Indeed, there was considerable consumer uproar when the serial copy management system (SCMS) was first suggested in the early 1990s in conjunction with the release of consumer Digital Audio Tape recorders. Hot on the heels of the launch of consumer technology capable of exact digital reproduction of compact discs, copyright holders demanded technology which would prevent individuals from making endless generations of perfect digital copies of a single original copyrighted recording.136 In the face of consumer hostility, the Audio Home Recording Act was enacted in 1992 and amended to the Copyright Act as Chapter 10, §§ 1001-1010.137 Chapter 10, entitled Digital Audio Recording Devices and Media, requires, inter alia, all digital audio recording devices to incorporate SCMS, which allows a first generation digital-to-digital copy, but prevents subsequent digital copies.138 Therefore, users are allowed one digital "copy" from a CD onto a DAT, a Minidisc, Digital Compact Cassette, and eventually recordable CD, for archival purposes or for listening in one's car. Significant elements of Chapter 10 of the Copyright Act include mandated royalty payments, a distinct Chapter 10 cause of action, and most importantly, the imposition of legal liability on technology itself, rather than on the uses to which that technology may be put.139
The lesson that copyright holders, Internet users and the legislature might take from the Audio Home Recording Act is that as long as the law is consistent and equitable, consumers will eventually accept, if not actually embrace, the law, as evidenced by the cessation of concern over SCMS. Although digital home recording technology is still as far from ubiquitous as agent technology currently is, the law is prepared for the eventual replacement of analog cassettes by recordable CDs or DVDs.
Similarly, the law can prepare for the ascendancy of agent technology by devising technology and legislation-based measures which can attempt to satisfy authors needs for copyright protection as well as users desires for fast, low-cost access to knowledge. We shall see how successfully such measures protect the interests of both sides. That this analysis has raised many additional questions and identified numerous but only hypothetical solutions, however, suggests that the existing copyright law is a poor fit to resolve the novel issues presented by the Internet.
fnA. Michael B. Sapherstein is a 1997 graduate of Boston College Law School and a founding member of the Intellectual Property and Technology Forum. This article is © 1997 Michael B. Sapherstein, published under license.
2. See, e.g., David Bottoms, Agents of Change, Industry Week, September 5, 1994, at 49.
5. Vanessa Houlder, Special agents - Intelligent gophers and experts are here, The Financial Times, August 16, 1994, at 12.
7. Id. at 51.
9. See id.
10. See id.
11. See, e.g., Mary Holden, Intellectual-property disputes flare on the electronic frontier, Chicago Daily Law Bulletin, April 22, 1995, at 1.
12. See Holden, supra note 10.
13.See, e.g., George E. Christodoulo and Anthony J. Fitzpatrick, Electronic Theft Could Create Vicarious Liability, The National Law Journal, April 17, 1995, at C44.
14. See Patrick Suarez, The Beginner's Guide to the Internet, 1993.
15. See id.
16. Marina Roesler and Donald T. Hawkins, Intelligent Agents; Software Servants For An Electronic Information World, Online, July 1, 1994, at 18.
17. See, e.g., Ron Coleman, Copycats on the Superhighway, 81 A.B.A. J. 68, 69 (1995).
18. Id. at 70.
22. 17 U.S.C. § 102(a) (1988 & Supp. V. 1993); Computer Assocs., Int'l. v. Altai, Inc., 982 F.2d 693, 705 (2d Cir. 1992) (citing H.R. Rep. No. 1476, 94th Cong., 2d Sess. 54 (1976), reprinted in 1976 U.S.C.C.A.N. 5659, 5667).
23. 17 U.S.C. § 410(c); Dynamic Solutions, Inc. v. Planning & Control, Inc., 646 F. Supp. 1329,1337 (S.D. NY 1986).
25. See Sony Corp. v. Universal City Studios Inc., 464 U.S. 417, 432 (1984).
26. Id.; 17 U.S.C. § 106.
27. 17 U.S.C. § 501.
28. See Gates Rubber Co. v. Bando Chemical Indus., Ltd., 9 F.3d 823, 831 (10th Cir. 1993).
29. See Raymond T. Nimmer & Patricia Ann Krauthaus, Software Copyright: Sliding Scales and Abstracted Expression, 32 Hous. L. Rev. 317, 341 (Symposium 1995).
30. See id. Note: Although the proof of actual copying has also been termed by some courts as 'substantial similarity,' I will employ Prof. Nimmer's terminology of probative and substantial similarity to emphasize that the proofs represent two distinct burdens on the plaintiff.
31. Selle v. Gibb, 741 F.2d 896, 900 (7th Cir. 1984).
33. Bright Tunes Music Corp. v. Harrisongs Music, Ltd., 420 F. Supp. 177, 179 (S.D. N.Y. 1976).
34. Gibb, 741 F.2d at 898; Harrisongs, 420 F. Supp at 178, n. 3.
35. See, e.g., Atari Games Corp. v. Nintendo of America, Inc., 975 F.2d 832, 840 (Fed. Cir. 1992) ("Even for works warranting little copyright protection, verbatim copying is infringement."); Williams Elec., Inc. v. Artic, Int'l, 685 F.2d 870, 876 & n.7 (3d Cir. 1982).
36. See, e.g., Williams Elec., Inc. v. Artic, Int'l, 685 F.2d 870, 876 (3d Cir. 1982).
38. See Atari, 975 F.2d at 840.
39. 17 U.S.C § 102(b); see, e.g., Baker v. Selden, 101 U.S. 99, 100 (1879); Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1443 (9th Cir. 1994).
40. H.R. Rep. No. 1476, 94th Cong., 2d Sess. 57 (1976).
41. See Lotus Development Corp. v. Borland Int'l., 49 F.3d 807, 807 (1st Cir. 1995), aff'd, 116 S.Ct. 804 (1996); Altai., 982 F.2d at 705.
42.See Lotus Development, 116 S.Ct. 804; Apple Computer, 35 F.3d 1435; Altai, 982 F.2d at 705. But cf. Apple Computer, Inc. v. Franklin Computer Corp., 714 F.2d 1240, 1253 (3d Cir. 1983) (Franklin did copy portion of Apple's literal object code in older iteration of OSCAR operating system), cert. denied, 464 U.S. 1033 (1984).
43. See id.
44. The Altai court devised a three-step filtration-abstraction-comparison test which provided an intellectual framework by which the Second Circuit, and subsequent courts in software copyright cases, could first sift protectable expression from nonprotectable material in software programs, and then compare the protectable remainder of the plaintiff's work with that of the infringing program. Altai, 982 F.2d at 710. This comparison would yield a result on the question of substantial similarity between plaintiff and defendant's works, and ultimately on the question of infringement. Id. Although the results in cases where defendants have invoked § 102(b) have not been uniformly predictable across the circuits, the tactic itself is nevertheless appropriate in most situations where defendants are borrowing, rightly or wrongly, from plaintiffs in order to create a new work of authorship. See Lotus Development, 116 S.Ct. 804 (1st Circuit ruled in favor of defendant-Borland that Lotus' menu command hierarchy was an unprotectable method of operation, and U.S. Supreme Court, by an equally divided court, affirmed the appellate decision, providing no additional guidance on issue of copyrightability of menu command hierarchy); Apple Computer, 35 F.3d 1435 (9th Circuit ruled in favor of defendant-Microsoft because limited number of expressions of graphical user interface rendered Apple's interface unprotectable); Altai, 982 F.2d at 705 (2nd Circuit ruled in favor of defendant-Altai on rewritten OSCAR operating system program because nonliteral elements of programs not substantially similar).
45. See Nimmer, supra note 30.
46. Altai, 982 F.2d at 702-03. See also Whelan Associates, Inc. v. Jaslow Dental Laboratory, Inc., 797 F.2d 1222, 1233 (3d Cir. 1986).
47. See Coleman, supra note 17. See also Williams Elec., Inc. v. Artic, Int'l, 685 F.2d 870, 876 (3d Cir. 1982).
48. See id.
49. See, e.g., Feist Pubs., Inc. v. Rural Tel. Serv. Co., 499 U.S. 340, 359-60 (1991) (copyright does not protect sweat of the brow in assemblage of public domain information for telephone book).
50. Gates Rubber Co. v. Bando Chemical Industries, Ltd., 9 F.3d 823, 845 (10th Cir. 1993)
51. Id. (italics added); see also Control Data Systems, Inc. v. Infoware, Inc., 903 F.Supp. 1316, 1323 (D.Minn. 1995). But cf. Marshall & Swift v. BS & A Software, 871 F.Supp. 952, 962 (W.D.Mich. 1994) (declining to apply merger doctrine where numbers on which copyright was claimed, unlike those in Gates, were not "facts" and thus the merger defense could raise no genuine issue of material fact).
52. Morrissey v. Procter & Gamble Co., 379 F.2d 675, 678-79 (1st Cir. 1967).
53. Herbert Rosenthal Jewelry Corp. v. Kalpakian, 446 F.2d 738, 742 (9th Cir. 1971).
54. Schwartz v. Universal Pictures Co., 85 F. Supp. 270, 275 (S.D. Cal. 1945).
55. Altai, 982 F.2d at 709.
56. See, e.g., Baker v. Selden, 101 U.S. 99, 99 (1879).
57. See id.; Lotus Development, 116 S.Ct. 804.
58. See U.S. Const. art I, § 8; See Lotus Development, 116 S.Ct. 804.
59. It must be noted that some 'surfers' of the 'Net do 'collect' the home page files of others, embodied in HTML code, in order to appropriate interesting ideas, techniques, or graphics for use in their own web sites. The issues raised by such actions, while intriguing, are outside the scope of this paper.
60. See id.; Baker, 101 U.S. at 99.
61. Altai, 982 F.2d at 693.
62. Note: Later in the paper we will explore cases in which defendants have downloaded copyrighted material posted on the Internet without the holder's permission - e.g., photographs originally appearing in magazines. The legal defenses of merger or scenes a faire are obviously inapplicable to these situations.
63. See Morrissey, 379 F.2d at 678-79.
64. See id.
65. See Altai, 982 F.2d at 693; Atari, 975 F.2d 832, 840.
66. Harrisongs Music, Ltd., 420 F. Supp. at 179.
69. Id. at 180.
73. See id.
74.17 U.S.C. § 504 (c).
75. 17 U.S.C. §§ 502-505.
76. 17 U.S.C. §§ 107.
79. Sony Corp., 464 U.S. at 456.
80. American Geophysical Union v. Texaco Inc., 60 F.3d 913, 923 (2nd Cir. 1994).
81. See id. at 918.
82. Sony Corp., 464 U.S. at 456.
83. Id. at 448.
84.See 17 U.S.C. § 107(2).
85.See 17 U.S.C. § 107(3).
86.Sony Corp., 464 U.S. at 450.
87. Id. at 421.
88. Id. at 450.
89. Id. at 451.
91.See Sony Corp., 464 U.S. at 448.
92.See id.; 17 U.S.C. § 107(1).
93.See Sony Corp., 464 U.S. at 450.
94. See id. at 449; 17 U.S.C. § 107(3).
95.See id. at 449.
96.17 U.S.C. § 107(4).
97. Note: For the sake of this line of argument, I must dismiss, for the time being, the possibility that a copyright holder will sue a user merely for 'viewing' his material a la Mai Systems Corp. v. Peak Computer, Inc., 991 F.2d 511 (9th Cir. 1992).
98.Karl Olson, Chasing Money Down the Information Highway, The Recorder, July 31, 1995, at 8.
99.See Sony Corp., 464 U.S. at 421.
100.See 17 U.S.C. § 107(4).
101.See Sony Corp., 464 U.S. at 421.
102.Note: This analysis is limited to the situation (currently the rule) where there is no equivalent commercial market to the product in on-line form; were the copyright holder to operate an on-line store for selling the images, digitized for computer use, the proper comparison would no longer be to the sale of magazines, and would as a result be more favorable to the plaintiff.
103.See 17 U.S.C. § 107(4).
104.See id. Cf. United States v. LaMacchia, 871 F. Supp. 535 (D. Mass. 1994) (court refused to extend liability under wire fraud statute to student's copyright infringement of retail software).
105.See Sony Corp., 464 U.S. at 450.
106.See id. at 442.
107. Id. at 450 n. 31.
108.See Olson, supra note 120.
109.See, e.g., Report of the IITF Working Group on Intellectual Property, White Paper, September 1995.
110.Internet World, January 1996, at OBC.
111.Quarterdeck Unveils WebCompass at Comdex, Due to Ship to the Market this Month, PR Newswire, November 13, 1995.
113.Note: The author has no relationship with or interest in Quarterdeck Corporation whatsoever. Quite simply, that company's new product provides a timely real-world example by which to amplify the discussion.
114.See Coleman, supra note 17.
115.See Harrisongs, 420 F. Supp at 179.
116.See Mai Systems, 991 F.2d at 518; Atari, 975 F.2d at 840.
117.See 17 U.S.C. § 107(4).
118.See 17 U.S.C. § 107(2-4).
119.Sony Corp., 464 U.S. at 435.
120.E.g., Sega Enterprises Ltd. v. MAPHIA, 857 F. Supp. 679, 679 (N.D. Cal. 1994); Playboy Enterprises Inc. v. Frena, 839 F. Supp. 1552, 1552 (M.D. Fla. 1993).
121.Sega Enterprises, 857 F. Supp at 686-87.
123.ITSI T.V. Productions California Authority of Racing Fairs, 785 F. Supp. 854, 861 (E.D. Cal. 1992).
124.3 David Nimmer & Melville B. Nimmer, Nimmer on Copyright Sec. 12.04[A],at 12-70.
125.Cass County Music Co. v. Vineyard Country Golf Corp., 605 F. Supp. 1536, 1536 (D. Mass. 1985).
129. See Sony Corp., 464 U.S. at 442.
132.David Bauder, BMI Develops Robot to Monitor Online Music Sales, CyberTimes: The New York Times on the Web (Oct. 16, 1997) http://search.nytimes.com/books/search/bin/fastweb?getdoc+cyber-lib+cyber-lib+17669+0++BMI, CNN Online (October 20, 1997) http://www.cnn.com/DIGEST/.
133.CNN Online, October 20, 1997.
135.See White Paper, supra note 108.
136.17 U.S.C. §§ 1001-1010 (1992).
137.17 U.S.C. § 1002.
139.17 U.S.C. §§ 1001-1010 (1992).