Norderhaug.CHI@Xerox.com
http://www.ifi.uio.no/~terjen/
Juliet M. Oberding, J.D.; Media Design inProgress; San Diego, USA.
74634.1166@compuserve.com
The purpose of this article is to analyze the potential impact of Intellectual Property Law on current web technical innovations and encourage the design of alternate technology to overcome emerging legal obstacles. Legal principles are drawn primarily from U.S. law and International Conventions. The article will focus on the paralyzing effect current copyright law has on the creation of new technology and the need for designers to develop innovations which temper the impact of copyright law. It encourages workable solutions before infringement litigation. A brief discussion of legal principles will follow. Thereafter, the paper will target legal/design issues with respect to derivative work, HTML and other aspects of web technology. Finally, this paper will suggest the use of a revenue model to combat copyright implications.
Design of new media technology has a large impact on the section of the social system that has to do with information and communication. The design of a medium determines its economics and thereby its content, even if many developers seem to be unaware of the larger consequences of their decisions. The development of the web technology is not deterministic, but is a result of the design decisions of those involved in production. The potential large-scale consequences of design decisions suggests an increase in responsibility of web technology designers, as well as expanded participation in the decision process.
Intellectual property law is often described as being an archaic device most appropriately suited to the protection of tangible goods. In recent years, the legislators have attempted to stretch the application of copyright to cover on-line media. Recent efforts to control copyright infringement in the U.S. has resulted in legislation which fails to grasp the essential nature of the electronic medium. John Perry Barlow [bar94] is right on the point when he states 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.
The rapid progress of technology creates holes which current law cannot fill. This imposes responsibility on the technologists involved in the definition of standards for the web, as well as creators of browser software. The human interface of technology defines how people can use it. In a progressive world, those that understand and design technology have increasingly taken over the role of legislators in shaping the future of society, controlling the behavior of people and the way in which they think. The web technology designers must make serious decisions based on personal values and the designers new role in determining the legislation of technology. Further, this role forces the designer to take a close look at the nature of authorship and the impact of intellectual property
Previously, legislation has been the main way in which society has controlled the inner working of the social system for information exchange. Now it is time that media designers take their responsibility seriously and create technology not limited by current legislation, but rather, if appropriate, built on the traditional purpose of intellectual property protection.
Branscomb [bra91] pointed out that as computer networks become more ubiquitous in the daily life of ordinary computer users, lawyers will try to adapt existing laws to fit the digitized world. No previous technology has made information sharing on computer networks as user-friendly as those related to the web. Currently, some nations are attempting to change the law to adapt to the new technology. This is of interest to the web designer and user in that new legislation may curtail current freedoms. The designer must understand the law in order to develop new technologies that protect the current environment for change. As a result, it is useful to review the ideas behind copyright, as well as the general principles of the current copyright law. The examples are taken from US legislation, but the general application is similar to the approach in other countries.
It is appropriate to review analysis on intellectual property protection in order to understand the proposed alternatives to current regulation.
Stallman [sta93] describes how laymen and politicians often believe that copyright is a natural right to restrict the public use of information. He found that the US legal system fundamentally rejects this view. According to the US constitution, the stated purpose of copyright is to promote the progress of science and the useful arts. The laymen, thus, often mistakes the means from the actual end.
Society uses intellectual property protection to maximize progress. The paradox of maximal progress in information-based products is important. Lack of copyright regulation and ownership protection of information would create a situation where no one would want to produce a medium or its content, because they would either not receive revenue or would receive inadequate revenue. On the other hand, too much regulation would chill development, because it would lead to the monopoly of specific products, and the inability to build on previous ideas.
As explained by Goldstein [gol92], information property protection is based upon the philosophy that information is property. The goal of protection is to secure the interests of the owner of the information. Such interests include ensuring an income from intellectual work, as well as protecting personal information. Some factors also include the interests of society to secure progress and control the individuals. The most delicate balance in copyright is between protecting the products of authors and at the same time giving authors some freedom to borrow from the work of others.
The concept of authorship is deeply interrelated with the principles of intellectual property. In order to decide the future of authorship on the web, it is important to investigate the origin and development of the concept.
Ong's review of the oral traditions [ong82] describes the oral age before the invention of writing. In the oral tradition, content was created from other content, and pieces were adapted and built together without reference to the source. People in the oral age were little concerned about the origins of their sayings. McLuhan & Fiore [mlf67] suggest that authorship was practically unknown even when written manuscripts appeared. Many small texts were compiled, often loosing authorship in the process.
McLuhan and Fiore [mlf67] suggest that the advent of print technology lead to authorship, and this is also supported by Goldstein [gol92]. Ong [ong82] explains that print lead to a sense of ownership of words, paralleled by a drift in human consciousness toward greater individualism, and followed by development of the resentment of plagiarism. He further suggests that it was the print culture that gave birth to the notions of originality and creativity, two central concepts in today's copyright regulations. There is thus a strong connection between media technology and the idea of authorship and copyright.
McLuhan and Fiore [mlf67] believed that technologies that allow for copying from one work to build another will result in a society which is less and less convinced of the importance of self-expression. Lange [lan92] observed how new technologies undermine intellectual property as a constraint in creative play, and proclaimed the end of authorship as we know it. If authorship was initiated by a media technology, then it will come as no surprise that new media technology such as the World Wide Web may have large impact on the future sense of authorship. The web has a potential for blurring authorship, by making it easier to intertextualize and compile information and ideas from multiple sources into a new product.
The application of digital author signatures in web documents can easily affect McLuhan's and Fiore's prediction. Authorship information does not need to get lost, despite inlining or even repeated cutting and pasting. If web technology standards includes digital signatures in documents, the result can be higher creativity because each author gets credit (or revenue) for his work. Such decisions by web technologists may potentially affect the need for intellectual property protection laws. As Palmer [cited in lan92] wrote:
If laws are dependent for their emergence and validation upon technological inventions, might not succeeding innovations require that those very laws pass back out of existence?
The challenge for web technology designers is to create such an innovation, and it is not obvious that they should support the traditional notion of authorship. Lack of better ways to secure the interests of information owners slows progress. Web technology designers must explore ways to secure revenue on the web which does not hinder creativity and reuse of existing ideas.
The United States Copyright Act provides owners of original works with the right to control activities which create commercial value. The owner of a copyright has the exclusive right to reproduce the copyrighted work, prepare derivative works, distribute copies of the work and, in the case of images, display the copyrighted work publicly [usc17/106].
In order to establish copyright infringement, one must show ownership of the copyright, copying, and that one of the rights guaranteed to the copyright owner is implicated by the violators actions [usc17/106]. The Copyright Act provides protection for original works of authorship fixed in any tangible medium. Specifically, section 102(a) of the Copyright Act provides that:
Copyright protection subsists...in original works of authorship fixed in any tangible medium of expression, now known or later developed, from which they can be perceived, reproduced or otherwise communicated, either directly or with the aid of a machine or device [usc17/102a].The U.S. Copyright Act [usc17/101-] and the Berne Convention [ber71] only provide protection for fixed works. A work which is not reduced to a tangible form is not a writing within the meaning of the statutory requirements [nim93a]. Therefore, fixation is a necessary prerequisite to copyright protection.
The Courts in the U.S. recognize three requirements for protection based upon the statute. These are originality, creativity and fixation. Protection attaches from the moment the work is sufficiently fixed. The work may be fixed in any tangible medium of expression [usc17/102a]. Works be fixed by any method which is currently known or later developed [usc17/101]. In the U.S., computer programs have generally been regarded as fixed works because they can be perceived with the aid of a machine or device [usc17/102a]. However, according to Nimmer [nim93a], embodiment in the tangible form must be sufficiently stable. For example, live radio and television broadcasts, while held in a tangible medium, are not fixed because the sounds and images are ephemeral. The same can be said of a reproduction captured momentarily in the memory of a computer.
Recently, courts have held that a program stored only in RAM was sufficiently fixed to constitute a copy [mai94]. In the Mai case, a manufacturer of computers which also serviced computers and owned a diagnostic software program brought an infringement suit against an unlicensed service company which used the manufacturers diagnostic program while servicing computers. The district court held that loading a software program from the computer's hard drive to RAM was copying.
The technology of the web creates interesting discrepancies when it comes to the issue of fixation. It appears that some transmissions will not be sufficiently fixed to merit intellectual property protection. An important recent report prepared by a federal committee [leh94] held the position that U.S. intellectual property law must be expanded to cover all transmissions via electronic medium. The Working Group noted that technological innovations could be used to provide intellectual property protection, and such technology is contemplated by the report, but the Group chose to recommend additional legislation based on the belief that no technology is failsafe. This position is typical of legislators attempt to rationalize amendments to law in order to extend existing notions to new media. As argued above, any attempt to retrofit existing law to apply to current technology is a dead end. and will not fulfill the purpose of copyright. Additional law will only succeed in slowing future development.
Copyright infringement occurs when a copy is made by one other than the owner. Current web browsers facilitate the creation of copies. Users and authors may create a copy of a browsed document in several ways, such as saving the source on a hard-drive from the browser, printing a rendered version of the document on paper, or cutting and pasting a section of a document for use in their own document production. Such behavior has consequences under the current legislative environment, and the original author may claim copyright infringement.
The question of fixation will also play a key role in determining whether there is copyright infringement on the web. This question appears to have been answered by the Mai decision regarding images contained in RAM [mai94], and this has important implications on the web. The key question answered by Mai is whether a transmission is sufficiently fixed to be considered a copy in a infringement action. On the web, merely following a link will create one or more copies. A duplicate of a document on the web may become fixed in a variety of ways. In addition to user or author created copies, copies are created merely as a function of web technology. The following is a list of such copies:
Following this, plain use of the web may cause copyright infringement in today's legislative context. In the moment that an individual fetches a document and views it on their screen for more than a few seconds, a copy has been created. This is concrete evidence of how inappropriate the current interpretation of the law is, and the paradox that is created when applying the law to the creative use of modern technology.
The creation of copies is an inherent feature of the technical design of the web, and necessary for the system to function appropriately. Thus, attempts to change the web technology to adapt to current copyright regulations will have negative side-effects on the technical design. However, one should not instantly assume that ordinary use of the web will subject the reader to suit. Current U.S. copyright analysis requires a review of the facts and law to determine if there is fair use.
Fair use is a legal defense to a claim of copyright infringement. Fair use has been described as being limited and useful forms of copying and distribution that are tolerated as exceptions to copyright protection [cab90a]. Fair use is often used as an argument for re-publishing works taken from the net, for example, by computer artists that have been known to enhance images from the net without permission from the creator [ros93]. In recent bulletin board sysop cases, the court has denied the defense of fair use. In the Playboy case [pla93], the sysop was held liable because users of the bulletin board were allowed to download copies of Playboy photographs at a profit to the sysop. The court found that lack of knowledge on the part of the sysop did not create fair use. According to the Playboy court, there are four factors that must be considered in determining fair use:
The first factor mainly deals with commercial versus non-commercial use. It is generally held that every commercial use of copyrighted material is presumptively an unfair exploitation... [son94]. Since most current use of the web is non-commercial, this would gravitate toward a finding of fair use. However, as the web becomes more commercialized, this situation may change.
The second factor relates to factual works versus fiction or fantasy works. The application of fair use is greater in the case of fact-based works [nim93b]. Copies of works classified as entertainment will probably not be considered fair use. Copying of academic and other factual documents on the web will more readily be considered fair use.
The third factor involves a qualitative evaluation of the copyrighted work. If what is copied is the essential part of the copyrighted work, then even a small amount of the taking is sufficient to preclude fair use [cab90b].
The fourth factor is the most important. This factor concerns whether unrestricted and widespread conduct of the type engaged in by the potential defendant would result in substantially adverse impact on the potential or current market [nim93c]. The sheer number of users on the web could have a definite impact on any market. For copying of a document on the web to be considered fair use, it should not materially impact the marketability of the copyrighted work.
In the Sony Betamax case, the court dealt with the issue of whether consumer copying of broadcast movies to video was a copyright infringement [sam93]. This is distinguishable from the Playboy case [pla93] or the MAI case [mai94] because those cases dealt with a taking for a commercial purpose. The court in the Sony Betamax case found fair use primarily because the taking was considered ephemeral even though the entire movie was copied [sam93]. The court relied on the fact that the copies were made for time shifting purposes and not as part of a video library. This may be similar to information held in RAM or temporarily stored on a computer for non-commercial purposes. This factor has an application to the web where information is stored for a limited period of time.
The Sega case [seg92] is important to consider with respect to fair use and the technology of the web. In that case, Accolade reverse engineered the Sega program to determine how to make compatible but original games to run on the Sega machines. While Accolade appeared to have breached every factor in the fair use analysis, the court found that there was fair use, and that the action contributed to the achievement of the main goals of the copyright system by enriching the culture through the creation of new works [sam93]. Thus, based on fair use, web authors that create derivative work that promote the purpose of copyright may be immune to infringement claims.
Although fair use may create an argument for the legal reuse of information by duplication, the technology itself should make it possible to minimize the need for content creators to make copies. The use of hyperlinks on the web is one way to reduce the need for duplication. Improved abilities to include parts of other documents without copying them can further the goal of minimizing the need for copying. With proper design, the technology can eliminate the need for claiming fair use of information when deriving documents from other information and services, at least when the originals also are on the web. This proposal will be discussed in further detail in a later section.
Branscomb [bra91] reminds us that the rigors of the market economy are such that it is not a viable economic policy to give away the results of intellectual labor without a fair and equitable compensation. Information is one of the few things humans can create without diminishing resources available to others. However, there are still economic obstacles related to information. The economists Cooter and Ulen [cited in gol92] describe the dilemma of private information property rights: without a legal monopoly not enough information will be produced, but with the legal monopoly too little information will be used. If we want to move towards a web of information with a quality worth paying for, an appropriate revenue model for information trade on the web must be facilitated.
The technology behind the web has important consequences for the notion of copyright, and the need for intellectual property protection. This section will clarify and discuss copyright issues concerning inlining and hyperlinking, because these are radically altering the way derivative work is produced.
Many content providers seems to find a difference between inlining and hyperlinking when it comes to copyright. This feeling is more based on the common behavior of browsers rather than a rational evaluation about how the underlying technology works. Many apply misunderstood versions of current copyright laws to the web medium. These practices can lead to behavior and etiquette that may not be beneficial for the construction of the WWW information space. A rational and legal analysis is thus in it's place.
It is easy to reverse engineer any HTML-document to find the locations of inlined images. These locations can then be used to inline the images when authoring other documents. This way providers can reuse graphics created by another producer without having to create or store any graphics themselves. Does such inlining violate the intellectual property of the creator of the image?
The initial analysis must always evaluate whether there is any copyright protection for the inlined work. The next step is to determine whether any copy is made. Each of these questions is driven by the facts of the particular case at hand. If you are dealing with a functional work or ideas, there is no protection. If one inlines an image, but there is no copy made at any point, there is no copyright infringement. Each case must be analyzed according to the facts.
Looking at the HTML specification [markup; specs-2.0], there is no difference between making a hyperlink and inlining information. In both cases, the markup used consists of an Uniform Resource Locator that addresses the other document, together with a tag and other attributes that give some hints to the browser as to what is located at the particular address. This information is used by the browser to determine how the link or document should appear. Eventually, the browser automatically fetches the document or presents an option to do so for the reader. Thus hyperlinks and inlines are technically the same as a cross-reference as published in a paper article, which arguably does not fall under copyright protection. Web content providers have suggested that copyright laws be updated to cover links and inlining. We find this to be a harmful and unnecessary idea.
Even on today's web, the technology gives the author high control of documents that are inlined or hyperlinked. The author can at any time change or remove information from the server, as well as restrict access to a document so that other documents have to be retrieved before the one requested. Therefore, no additional legislation is required to increase the control of the author.
Derivative work is built on previous work to create a new product that adds value by structuring previous information and/or adding new ideas. Except from the most simple products, all human creations are in some way derived from previous ones. As explained by Lange [lan92], copyright inhibits or suppresses the creation of derivative work. Before hypermedia technology, derivative work was created through enhancing previous work by copying and altering it to create new products.
The web supports better ways to create derivative work than any previous medium. By making hyperlinks to earlier created works, authors can efficiently build on previous ideas without repeating what has already been done. Inlined images provides a mechanism through which authors can create documents that bind together segments of information created by other authors without copying the information. This is an important progression from previous media. Inlining by reference is revolutionizing the problem of copyright in an electronic medium. No unauthorized copying is needed on a globally interlinked information space such as the web, where the original can rest at a server in control of the author.
When creating content for media that requires copying of information for making derivative work, designers often waste significant time negotiating for rights to material. Recent examples are Compton, which spent more than eight months negotiating for rights to photographs and music for one of their CD-ROMs [cox93]. Producers of interactive media on CD-ROM have encountered many problems because creators are protective of their rights. According to Daubert [dau92] the uncertainty in level of protection for computer applications decreases innovation and discourages reuse of ideas and expression even if the origin is not harmed. Before appropriate revenue models for the web are created, copyright will prevent information from being added to the web and, thus, hinder the creation of derivative work.
The web content providers are affected by similar confusion regarding unclear guidelines, leading to less progress and diversity. The web is rapidly changing and diverse. This makes it almost impossible to expect any clear guidelines, because they are immediately outdated when published. What is needed is a dynamic and flexible system that can be adapted to the situation. The web providers and media designers should not wait for legislators, but create own solutions and norms. What previously would have been legislated can instead be built into the technical design. Enhancements of the method through which documents may be distributed on the WWW can limit the need for developers to copy chunks of documents authored by other individuals. Information creators must be provided with the necessary tools for exact linking and inlining of other documents. Rather than further inlining at the server [ncsa-inc], advanced inlining should be implemented on the client/browser. Documents which include information about it's origin as part of the technical solution may further clarify who is entitled to claim recompense for any portion used in the production of another work.
Web content providers can only look to current legislation for protection of their content to a limited degree. Rather they need to use the medium itself if they want to limit the usage of the content. Web technology designers can provide a framework for making this possible. The web itself can be used to create most of the necessary intellectual protection. For example, in the case of re-publications on the web, web robots combined with pattern matching can automate the searching for identical or substantially similar work. The ability to overcome the limitations of legislation is now in the hands of web technology designers.
At the same time, the web makes it easy to make a copy of other's work and claim authorship. As shown previously, applying today's copyright legislation to the web is a dead end. However, the web desperately needs some alternative protection for authors. The revenue model can be built into the web technology, bypassing the need for legislative protection. The role of legislation should be to cover eventual holes that the technology leaves behind when it comes to protecting intellectual property, and to maximize the large-scale creation of ideas and progress.
Intellectual Property is, of course, related to the content providers revenue. A mediums revenue model is the format that allows content providers compensation for their work. In this section, we argue that there is a strong connection between the underlying technology and the revenue model.
The design of a medium strongly affects the revenue model that is applied to that medium. Entrepreneurs will seek an appropriate way of creating a revenue constrained and encouraged by the media technology. One example of this relationship can be discovered through examining the different variations of the television medium. The traditional technology of broadcast made it difficult for content creators to charge individual viewers for selecting a program. This lead to revenue sought via advertising, based on making the viewers the product rather than the customers, and resulted in bland programming in an attempt to get as large an audience as possible to sell to advertisers. The technology of cable television and scrambling made it easier to charge for subscription or pay-per-view, causing more directed programming towards smaller audiences.
Current economic revenue models on the web include subscription and use of commercials. Lack of better ways to identify origin of content on the web leads to providers creating their own technical fixes, such as forcing readers to access a document with credit before other information can be accessed. Such private initiatives are a product of the web technology standards, in that these encourage certain actions by the content provider and limit more appropriate solutions.
The revenue model has implications for what content will be made available on a medium as well as the quality of that content. Web technology designers can strongly impact which revenue models will be used on the web. Creating alternative revenue models is important to avoid the application of potentially harmful revenue models due to a lack of alternatives.
Ted Nelson [nel90] has suggested a method of securing income for material in his Xanadu project. This project is in many ways similar to the web, and his proposals are worth looking into when investigating potential revenue models for the web. Xanadu documents can include parts of many documents and integrate them into a single identity. Rather than quoting from a document, a composed document will pull the included text from its original source whenever it is read, and will automatically ensure that the original author of the quote is correctly credited and allotted royalties for the use of the quote [swa91a; swa91b].
The web needs an improved revenue model which covers both the payment of the author and credits. The web technology designers should be aware that they have a fundamental affect on the revenue model that will be applied. Thus it may be important to integrate the creation of revenue models for the web as a part of the technological design. This may argue for the inclusion of specialists from other areas, such as economics and legislation, in the technology design process.
Litigation is not the solution to author revenue protection. Indeed, litigation will do nothing but slow the pace of any technological developments. Currently, the cry of copyright infringement on the web prevents individual developers from pursuing some interesting work. This tendency to threaten others with litigation is both irresponsible and, in some jurisdictions, illegal. Unfortunately, lawsuits will be filed and judges with limited knowledge or understanding of technology will make rulings that impact the future. Technologists must take action now to prevent improper precedent. Do not depend on the legislators to create new law: there is no great movement underway to radically rewrite current copyright laws.
Ability to inline different media-types and parts of documents is a driving force for improving the way we handle information on the net, and should be furthered in markup standards. Additions like inlining whole or parts of other documents by reference should be added to our markup language. At the same time, the technology needs to support accreditation and revenue of authors.
Certainly, it is not appropriate to adapt or contain technology to fit current laws and regulations even if it were possible. Rather the law must change to fit the growth of technology. Until judges, lawyers and legislators catch up with the digital age (if ever), technologists must create solutions to bridge the gap. Legislation should be secondary to web technology when it comes to being a means for shaping the future. The technology should be designed to minimize the need for legislation, as well as give as much decision power as possible to the individual users.
The design of the web technology has large impact on individuals and the social system. The notion of intellectual property is one of the issues that is strongly affected, and is of high relevance for the shape of an information-driven society. The flexibility of digital media technology gives large control to the design of web standards. This suggests a high level of responsibility of the designers, as well as widespread involvement in the design. The implications of the web are all too important to leave solely to technology-optimizing decisions.