Introduction
Historically intellectual property has generally consisted of literature protected by copyright, inventions protected by patent, trade secrets protected by state and recently federal law and trademarks protected originally by common law then by federal statute. Infringement was usually rather visible as it would involve the public use or sale of another's intellectual property. Private infringement was rare because of the costs involved in creating just a few copies of a machine or book for personal use. It was quite sufficient to protect intellectual property through legal means as the cost of infringement inhibited casual infringement and professional infringement could be reasonably discovered through diligent investigation. However with the advent of personal computers and the Internet this model has radically changed.
Not only has a new type of intellectual property, software, and a new market for intellectual property arisen with the growth of the computer market but the cost of infringement has dropped to nearly zero for every consumer in the world. Traditional intellectual property still exists but whether it is a book, picture or a trademark it can be easily replicated in an electronic format. And with the Internet it can be sent around the world to any number of consumers instantly and with almost no cost. Although traditional mechanistic inventions are not endangered they have a unique counterpart that is at the very core of the new model.
Software is the mechanism which allows computer users to produce and organize graphics, text, data, etc. With so many users of computers who are also on the Internet there has been a corresponding demand for software and data. This has dramatically increased the value of software and content producing industries but because of their ease of replication and distribution the problem becomes one of protecting the intellectual property so that the producer can benefit from their investment in its creation. However, this is a simplistic observation and further analysis reveals other ramifications that may lead to a reshaping of traditional incentive and market structures for intellectual property.
Intellectual property in the electronic environment can be protected with law and with technology. Legal protection can be either public law created by government such as statutes or private law created by individuals such as contracts and licenses. The problem with legal mechanisms is enforceability. Technological protection attempts to solve this problem through hardware and software which prevent unauthorized use of intellectual property.
Technological Protection
Superdistribution is the concept of using the Internet to distribute content and applications on a pay per use basis. The very nature of electronic data allows it to be distributed via the Internet and reproduced on a computer at practically no cost. This characteristic has been the bane of software and content copyright holders as there intellectual property can be sent to anyone in the world with the click of a mouse button without any compensation to the copyright holder. These unauthorized copiers can give away hundreds and thousands of copies yet run a very low risk of being caught. Superdistribution uses this characteristic as the basis of its approach. Instead of fighting the human nature to share, especially when they can do so freely, this method encourages dissemination of copyrighted material to as many people as possible. The catch is that the material is only usable on some variation of a pay per use scheme or as it may be called, meterware.
Superdistribution technology was first developed by the Japan Electronics Industry Development Association in 1987. Their model is based upon the premise that it is easier to monitor usage of electronic objects than it is to monitor how often they are copied. Software based upon this model contains instructions that only allow it to be used only upon specially equipped computers.
The Japanese prototype of this system involved an integrated circuit which plugged into a computers coprocessor slot. Electronic objects designed for Superdistribution would run only if they could access the special circuit which in turn kept track of the usage of the object. The usage reports would then be periodically sent to an administrative organization which would bill the user's account.
Payments by the consumer for usage of electronic objects whether they are software, a book or a music video are then automatically divided between all those entitled to royalties for usage of the particular object. Implicit in this scheme is the idea that in the future software programmers may create a chunk of software code that performs a certain function and that is designed to be combined with other compatible software objects to create a customized multifunctional application. These applications can then be either used by the assembler or distributed without permission of the original object creator as that creator will be automatically reimbursed for all usage of their object.
Superdistribution based upon the Japanese hardware model has not yet penetrated the world market but other technologies based upon the superdistribution concept have been announced by several companies. In the Spring of 1996 IBM announced its new cryptographic containers they call Cryptolopes which allow for secure packaging of digital information. As a representative example of these emerging technology schemes a further exploration of how Cryptolopes work is provided below.
A Cryptolope container allows for content and control information to be distributed together. The control information provides the licensing terms including price and other conditions of usage as granted by the copyright holder. The container also includes network addresses and the means for communicating usage data to clearance centers which handle the billing and royalty distributions to the copyright holders.
On a practical level the distribution process requires that consumers download a free plug in that enhances their Web Browser to use Cryptolopes. Then the consumer identifies the information they want from the IBM "Information Marketplace" or "infoMarket" database website and clicks on the appropriate link. The information is then placed inside a Cryptolope along with the necessary usage information and is sent to the user via normal Internet transfer protocols. The user's enhanced browser identifies the Cryptolope and allows viewing of the data while taking care of the usage reports in the background. The envelope can also be redistributed by the consumer without limitation since each one is self contained to provide for usage by multiple users.
When a user wishes to use data in a container they are presented with the pricing and terms of use for that data. Upon their consent to the terms and price the usage payment is implemented through a clearance center as described infra. An intriguing aspect of this technology is its ability to provide for modified usage and payment specifications for a given piece of data. The owner of data may leave modified terms with the clearance center so that when the request to open a Cryptolope is received at a future date the container can then be given new usage or pricing specifications. An example of this would be for data which due to its age or competition should be priced lower so as to entice the consumer to use that data as opposed to some other producer's data. It is also possible for the data to be given an internal schedule of payment decreases or other term changes that do not require it to phone home.
The Cryptolope Container
The Cryptolope container consists of several layers of data:
The outer layer is the Bill of Materials which includes a list of the container contents and the digital signature which is used to verify the integrity of the container. The next layer contains a clear text abstract or description of the contents and clear text metadata such as the author, size and date of preparation.
The encrypted content of the container may consist of any number of individually encrypted files containing any form of digital information. A public/private key pair is initially generated by a Clearing Center which distributes the public key to the person wishing to enclose information in the Cryptolope. When the content is placed inside the Cryptolope it is encrypted with a randomly generated one time symmetric key which is in turn encrypted with the public key and placed in the key record file along with the id of the public key.
The Rights Management Language layer provides all of the terms of use for the encrypted information. The license can specify any terms including price, printing rights, intended use, etc. As with any contract the licensor can make the material available on any terms. That being said, the legal limitations on contract on valid contract terms can be significant and will be discussed in greater detail infra.
Information to control unauthorized use of the information once decrypted is in the forms of fingerprinting and watermarking. Watermarking is the inclusion of a faint background image indicating the owner of the content. Fingerprinting is the use of an invisible algorithmic tag which can allow tracking of the path of any unauthorized distribution. A unique fingerprint or watermark is generated by the clearance center when a request for access to a Cryptolope is granted and is included in the license Cryptolope. The mark is then applied by the opener application to the files being extracted from the Cryptolope. Each mark provides information as to the person to which that copy of the data was licensed.
The digital certificate contains the appropriate certificates including the public key used to check the digital signature in the Bill of Materials layer thus verifying the integrity of the Cryptolope.
The Cryptolope commerce model is initiated by the distributor of information who inserts it into a Cryptolope and distributes it via the Internet or on other electronic media. A consumer wishing to use Cryptolopes must procure an opener program which is free from IBM. Once the opener has verified the Cryptolope integrity it presents the abstract and license terms to the consumer. The abstract or teaser information should be designed to provide a free sample of the encrypted information so that the consumer can decide whether to proceed with the purchase. The opener program handles all the necessary functions on the consumers end and is designed to operate invisibly. If the consumer agrees to the license terms then the opener application generates a 'Buy Cryptolope' which has the same structure as described above. The relevant information including the terms agreed to, the keys for the individual files which the consumer wishes to access and any required information from the user such as the date, etc. are placed in the Buy Cryptolope which is then digitally signed and sent to the Clearance Center.
The Clearance Center verifies the consumer's digital signature, analyzes the licensing terms which the consumer is requesting, records them, and returns to the consumer's opener application the necessary information to implement the terms. Prior to this step the consumer must have registered their digital signature and provided a payment method most likely a credit card number. The Clearance Center determines which rights to grant the consumer based upon what is requested and what restrictions or special grants might be available depending on the information included in the consumer's registration file. Upon determination of what permissions to grant the consumer a record is generated which will provide the basis for periodic billing of the consumer's account and will likely be used to develop marketing information about the consumer. The usage information is then inserted in a container called the License Cryptolope.
The Clearance Center will decrypt the symmetric session keys which were encrypted with the distributors public key and re-incrypt them with the public key of the consumer. These keys are then inserted in the License Cryptolope and sent to the consumer. The consumer's opener application then makes access to the information available according to the instructions created by the Clearance Center.
Legal Protection
Copyright management systems (CMS) are technological methods of allowing rights holders to control the dissemination and meter the use of their intellectual property (such as described supra). However any technology is susceptible to tampering especially in this realm where either hardware or software can be created to foil the protective system. To this end there have been recent moves to make any such CMS thwarting application illegal. The emerging legal schemes focus on prohibiting tampering and prohibition of removal of "rights management information" (RMI)
The first domestic step toward new legal protection for intellectual property on the Internet was through the National Information Infrastructure Copyright Act of 1995. This proposed act incorporated all of the recommendations made by the Clinton Administrations Information Infrastructure Task Force in their White Paper on Intellectual Property and the National Information Infrastructure. Opposition by many groups including librarians, Internet service providers and computer manufacturers resulted in Congress adjourning before it could be brought to the floor for a vote. Some of the more significant elements of this proposed act were an anti circumvention and a rights management clause.
Notwithstanding the failure of the 1995 Copyright Act the United States proposed basically the same language to the December 1996 meeting of the World Intellectual Property Organization (WIPO). This meeting was specifically to draft a protocol to the Berne Convention regarding copyrights in digital works. In regard to CMS the final language agreed upon was somewhat different than the U.S. proposal but with similar substance.
The final language in regard to technological protection calls for member nations to provide adequate legal protection and legal remedies against the circumvention of technological protection for copyrighted material. The RMI clause requires that removal or alteration of any rights management information or distribution, importation, broadcast, etc. of works with knowledge of such deletion or alteration be given effective legal protection. If the U.S. ratifies the treaty then implementing legislation for these aspects, similar to that in the Copyright Act of 1995 will be required.
The greatest concern is that the implementing legislation will focus on classes of technology to regulate or prohibit. Such legislation would meet the requirements of the treaty but would represent a significant difference from the treaty focus on conduct in specific act. Such though is the draft language being promulgated by the Patent and Trademark Office. This proposal would prohibit "importation, manufacture and distribution of any device, product, or component incorporated into a device or product, or offer or perform any service, an effect of which is to avoid, bypass, remove, deactivate or otherwise circumvent any process, treatment, mechanism, or system which prevents or inhibits the infringement of any of the exclusive rights of a copyright holder under section 106 (of the U.S. Copyright Statute) with reckless disregard for facts demonstrating that the device, product, component, or service primarily enables such infringement or with intent to primarily enable such infringement." The problem with such law is that it could easily encompass technologies with substantial benefit that are also of use in circumventing copyright protection. Software for encryption, deciphering, and reverse engineering as well as hardware such as the personal computer all may have a primary purpose or effect which runs afoul of this proposed law.
Then there is the case of lawful tampering. Copyright holders are given certain rights in §106 (U.S. Copyright Statute) but there are several enumerated exceptions and limitations. Most notably is the fair use allowance of §107 otherwise there are special provisions for libraries and for nonprofit uses. As well the copyright term might expire or works or facts which are not protectable under copyright might be encrypted. The significance is that in each of these cases any tampering with the CMS would not be for the purpose of infringing copyright. But in each case the tools required to extract the data from the protected environment would most likely be unlawful under the proposed law.
More generally the question of whether to protect intellectual property on the Internet via law or technology revolves around the type of incentive and market model we want to utilize for our electronic marketplace. Public law regarding the protection of intellectual property in the U.S. is based upon the U.S. Constitution, (The Congress shall have power) "to promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries" . The primary purpose of U.S. intellectual property rights is then to provide an incentive to individuals to invest their time and energy into creating works that provide benefit to the public. This is a very utilitarian approach which places the benefit to the public (promoting progress of science and art) above the interest of the intellectual property owner.
Many nations though are based upon a Civil Law tradition where the concept of 'Author's Rights' is the basis for their copyright protection. Author's Rights imparts an inalienable right to authors for their creations that is superior to any notion of public benefit. This basic concept that what someone creates is their property and should be subject to their complete control forever is also widely held among those producing intellectual property today. The result is that we are at a crossroads where the rights claimed by intellectual property creators is on a collision course with the Constitutional purpose of those rights.
The issue is whether technological protection of intellectual property (here after IP) will be the enforcement mechanism for a regime of public IP law such as currently in place in the U.S. or will it be the implementing tool for a world of private law to which public law serves only as a foundation. Under the current IP law the bundle of rights given to the IP owner are expansive but not infinite. The rights which are given are for the purpose of providing an incentive to the IP creator to invest in the creation by assuring an opportunity for private benefit by not allowing another to usurp the IP. The limitations are that the rights extend for only a limited time, public disclosure of the IP and certain fair use rights which are reserved for the public. These limitations can not currently be avoided through contract.
Federal copyright law preempts any licensing term which might attempt to extend the IP owners control of these limited areas.
However, the scope of contract is being pushed outward by IP owners. Recent court decisions are mixed as to the validity of various terms in mass market software licenses both under contract and copyright law. But there is a new provision of the Uniform Commercial Code being promulgated to the states. This proposed section 2B provides for statutory validity being given to mass market licenses. Additionally it provides for the transfer of contractual terms between subsequent parties to the contract. Traditionally if two parties agree to a contract and one party transfers the subject of the contract to a third party the third party is not bound by the terms of the original agreement. Part 5 of section 2B would alter this and provide for the original license to run with the subject of the license (i.e. software or data) and thus bind any persons to whom the IP was transferred. Although these are not preemption issues 2B contains many more aspects which provide a complex matrix of legal provisions which avoid preemption conflict only "insofar as possible".
There is also proposed domestic and international law to provide copyright protection of databases. The international aspect of the Internet and perceptions around the world of rights to IP leads to another fundamental aspect for change.
The U.S. Constitution provides for one other equal as the law of the land, a treaty with another nation. Thus it is possible to alter the constitutional purpose (public benefit) of IP through International treaty without the difficulties of amending the Constitution. As well Congress can alter the IP statutes at any time through normal legislative actions. Further, the courts may begin to redefine the frontier between IP owners rights and public rights provided by Federal law especially when confronted with cases based upon aspects of UCC 2B.
The point of this is that there is a growing momentum behind giving the greatest possible control of IP to its creators or owners. The law is amendable in a number of ways and what system of rights will prevail is not predictable. But analysis can be given as to the pros and cons of each system. Assuming that technology will provide a high level of physical protection and that the laws will provide both protection from circumvention and causes of action against those who do succeed in circumvention, what then is the outcome of a world where full legal effect is given to IP owners licensing terms.
Great concern has been voiced in the legal community that such a world would lead to high costs, limited access and a skewing of the incentives to create IP and the ability of others to build upon what has been created. But with the Internet connecting the world, data being instantly transferable at little or no cost and electronic objects being infinitely reproducible at little or no cost it is not clear that such ominous warnings are well founded.
Constitutional IP protection is based upon a world with a significant market failure. Until the Internet and personal computers IP was for the most part expensive to disseminate and to reproduce. The cost structure provided protection against small scale private infringement and the law provided remedies against those who infringed on a large scale. The extremely important aspect of incentive has also changed. The proper level of incentive is very difficult to quantify. IP protection was given for the purpose of providing a significant incentive to produce IP while also allowing it to be used by the public in limited ways until the period of protection ran out. Whether the level of incentive was the proper level has always been a subject of great contention. But the Internet model with strong technological protection provides an incredible possibility, an extremely free flowing marketplace where not government but actual demand is the principle factor in resolving the proper level of incentive.
If IP producers are allowed to create and disseminate IP under any terms they wish the constraints of competition and public willingness to pay will inhibit inordinate licensing terms. Since there will be a world full of potential IP producers with access to the Internet market if there is great demand for data or software then many competitors may quickly enter a market segment. A provider who does not license competitively will lose out to other producers. Such a fluid marketplace will allow for excellent determination of the proper level of reward and thus incentive for producing IP.
Neither does it seem that such a model will inhibit the collective building upon earlier ideas. IP law today limits the use of ideas and expression for a considerable period of time. In the private law model the ideas or expressions will have limited protection as well. Once a product is released how exactly it works may not be published but competitors observing it may produce their own substitute in far less than the 20 year monopoly which would be granted if it were patented. Producers of proprietary systems will have difficulty in the market just as they are having today. Open systems develop economies of scale from compatibility, larger numbers of users and ancillary producers. Proprietary closed systems will continue to face such competition and thus will not benefit from the private law model. Ultimately a private law model would of course have pitfalls but then the current model is not perfect either. The likely implementation would be to base our electronic economic model on private law but with legal limits place upon anomalies which may arise.
The obvious problem with relying on law to provide protection is that of enforcement. The most stringent laws will do no good unless there is a mechanism to ensure that the protection they provide actually exists. Technology can provide this enforcement mechanism. But the greater question is whether technology should only be an enforcement mechanism or the stepping stone to a new electronic economy based upon private law as opposed to public intellectual property law. Such an economy might take advantage of the fluid market dynamics presented by the Internet and the electronic medium resulting in greater efficiencies than are available under the current model.
© Copyright Jay McKinsey, 1997
http://www.pleasurepoint.com/~jay
Download RTF version with footnotes.