New Basics: 12 Principles for Fair Commerce in Mass-Market Software and Other Digital Products

Abstract: This paper gives an overview of the principles project of Americans for Fair Electronic Commerce Transactions (AFFECT), the coalition of business, consumer and library customers of mass-market software products. AFFECT has successfully blocked passage of the Uniform Computer Information Transactions Act (UCITA) since two quick enactments in 2000, before the opposition effectively organized. The law of mass-market digital products is in need of clarification to protect both the consumer interest and the public interest in sound information policy that fosters innovation and competition. Unfortunately, UCITA is poorly drafted and not an improvement over the existing patchwork of common law, Article 2 of the Uniform Commercial Code, and federal intellectual property law. UCITA is slanted toward mass-market producers and against the interests of consumers and the public more generally. AFFECT has now moved beyond opposition to a bad statute and is engaged in a positive search for better alternatives. As a first step, it has produced a statement of principles. These principles can be used as a basis for specific reform efforts, including drafting a model license and writing a more modest and targeted model statute than UCITA. They also can be used by consumer protection officials and consumers’ counsel to choose cases to litigate and by consumer advocates to identify bad practices to bring to the attention of the media and the consuming public. The principles have been written to be accessible to the general public as well as lawyers and policymakers. Many end-users of mass-market digital products are alarmed by unfair terms and practices. They are looking for ways to effect change. The principles identify what is wrong in current practices as they envision better ones.

This paper, in addition to tracing the origins of the principles in the struggle against UCITA and discussing their possible uses, explains their underlying theory. The principles draw upon familiar ideas from the law of contracts, commercial transactions, consumer protection and intellectual property law. They also adapt these ideas to the digital world. Some key concepts are good advance disclosure (easy on the Internet) to maximize shopping for the best terms, meaningful assent,

and substantive limits on terms that are unfair or contrary to public policy, particularly information policy. The principles also reflect the insight that the category “consumer” should include all customers who purchase mass-market products with non-negotiated terms. The minority of customers who read and shop over terms introduce only weak competition in non-salient terms in the massmarket. This is why substantive limits are also needed. In mass-market transactions, general standards such as unconscionability—requiring expensive and time-consuming case-by-case litigation—are not an effective way to deal with identified problems that could be quickly solved by specific rules. The principles address such common issues in the world of digital products as transfer and use rights, disclosure of product defects, and security of systems and data.

Introduction

1 Yet, despite its inadequacies, this body of law is the vehicle for resolving an important policy tension, between private ordering, on the one hand, and, on the other, the public interest in sound information policy, including preserving the culture of innovation that produced the digital technology revolution. Sooner or later, state or federal law reform will be needed, but it may be wise to start with a relatively small project, with a well-defined scope and a clear objective of addressing the most pressing problems.

Currently, the most problematic type of software and digital content transaction is for mass-market digital products, where the market is at best working weakly and does not necessarily do a good job of serving either the interests of customers or of the public more broadly. The lack of fairness in non-negotiated terms for many of these products is symptomatic of market failure and market weakness,

resulting, for example, in terms that provide no usable remedies for product quality defects or that attempt to prohibit study of products or comment on them. 2 The market for digital products is characterized by network effects that push customers to use products that establish an early lead in market share.

To become market leaders, producers tend to rush products to market before they are fully tested, leaving customers to deal with the costs of product flaws. Other market problems include monopolies, niche products with no alternatives, and parallel decisions not to compete on terms.

efforts concerning software and digital content deals is with a focus on massmarket transactions in digital projects. To do this job well, we need to have basic principles in mind. Although law reform could begin at either the state or federal level, ultimately federal legislation will probably be desirable because of the connection to federal intellectual property law and policy involved in addressing the overreaching in non-negotiated terms for mass-market digital products. Some state statutory experiments could lead the way to this outcome.

Americans for Fair Electronic Commerce Transactions (AFFECT), a coalition of consumer, business and library customers of mass-market digital products, has recently released two versions of a statement of principles, a simple version

intended for the general public and a technical version for policymakers, containing more explanatory detail. 6 In this chapter, I will attempt to provide historical, theoretical and practical background on the AFFECT principles project, in which I participated as a volunteer member of the Drafting Committee.

Broadly, AFFECT’s goal is to establish principles for fair transactions in massmarket digital products before taking on the challenges of implementation of law reform. AFFECT seeks to correct for market failure and weakness and also to assure that democratically made information policy is not undermined by nonnegotiated terms, often in fine print. Producers have attempted to use delayed form terms to write their own more favorable intellectual property law. AFFECT shares the widely held view among intellectual property experts that state contract law, especially as applied to mass-market digital product transactions, should not be a vehicle for undermining the public policy balance of intellectual property law.

AFFECT has learned from the largely failed Uniform Computer Information Transactions Act (UCITA), promulgated by the National Conference of Commissioners on Uniform State Laws (NCCUSL).

AFFECT, in contrast, has started with principles and has focused on only one type of deal, the mass-market digital product transaction. The AFFECT principles address specific, identified problems in this subset of deals, where the market is working least well and where intellectual property norms favoring competition, creativity, and innovation are most under threat. The principles are not designed for fully negotiated transactions. They also are not written with open source software in mind.

A core concept in the AFFECT approach is the mass-market transaction in digital products. This concept has two parts, the digital product and the mass-

market transaction. Digital products include software (copies of computer programs), digital content (such as digital copies of books, recordings, or movies), and combinations of the two (such as a copy of an e-book with a copy of a computer program to navigate it), when marketed as finished products or components of or add-ons to finished products. Mass-market transactions involve distribution to a substantial public of end-users under substantially the same terms

with no negotiation except as to price, quantity, method of payment, selection among standard options, or time or method of delivery. 13 The category of massmarket digital product is quite functional because there are many transactions in such products in the marketplace, and most of the problems in these deals can be dealt with well by one set of basic principles.

In the balance of this chapter, Part One reviews the history of the principles and describes the process by which they were produced. Part Two summarizes the content of the principles themselves and explains how they address both the process and the substance of digital product deals. Part Three describes the long tradition in contracts, commercial law and consumer protection scholarship that provides the theoretical underpinnings for the approach of the principles. Finally, the conclusion outlines the uses of the principles and discusses possible follow-on projects.

Part One: History of the Principles

The history of the AFFECT principles is in opposition to UCITA. UCITA’s history, in turn, is in the failed effort to revise or amend UCC Article 2. In 1988, the sponsors of the UCC, NCCUSL, and the American Law Institute (ALI), appointed a study group to explore revising UCC Article 2.

When the Article 2 study group issued its report in 1990, its primary scope concern was mixed transactions in goods and services. Its only reference to the issue of coverage of software was as an example of a mixed transaction in goods and services, specifically when a transaction involves hardware, software, and backup services:

A current example of some interest is a contract for the sale or license of computer systems, which involves hardware, software and various backup services. Are these “transactions in goods” to which Article 2 should apply? Is scope an either-or

proposition, or is there room for a selective application of relevant Article 2 sections to part of the transaction? [Footnotes omitted.] 17 One of the footnotes alluded to the fact that NCCUSL was considering whether to

prepare a uniform act on software contracts. 18 Overall, the study group’s report did not identify any major reason for the revision project

in statutory language. 19 Despite the lack of a compelling reason to pursue the project, the Revised Article 2 drafting project began after the study committee report was published.

The issue whether to cover software remained in limbo in the early years of the Revised Article 2 project. In 1995, the Article 2 drafting committee experimented with a draft covering both sales of goods and licenses of software (a development that might have supplied the missing rationale for the project), but later that year NCCUSL decided to go forward with a separate UCC article, Article 2B, on licenses of “computer information.”

The politics of the two committees were quite different. The majority of the Article 2B drafting committee consistently voted for positions favored by software producers, while the Revised Article 2 drafting committee was more balanced in its treatment of sellers and buyers.

The project was scaled back somewhat from a revision to a set of amendments. The ALI membership narrowly approved proposed Amended Article 2 in May 2003, following final approval by NCCUSL in August 2002. The project remains controversial and does not have good prospects for enactment due to opposition from all affected interests.

The amendments to Article 2 were finally promulgated by ALI and NCCUSL after it had become clear that UCITA, first approved by NCCUSL in 2000, had stalled in the enactment effort. In an unsuccessful effort to minimize controversy, proposed Amended Article 2 punted on the issue of the coverage of software and

27 a change that would throw

in doubt the body of case law applying current Article 2 to software, without

supplying an acceptable alternative and thus perhaps encouraging application of

UCITA by analogy. This is the primary reason that AFFECT opposes proposed Amended Article 2. 28

29 This comment suggests

that downloaded software would be outside the scope of the amended article, but

that it could be applied by analogy to downloaded software to the extent

appropriate. On the other hand, the comment states that the software in hard

goods—such as software in a car—is covered by Amended Article 2. Thus, the

comment leaves open the possibility of a distinction between downloaded and pre

loaded software, a distinction that is not functional or sustainable. Furthermore,

such a distinction could drive engineering decisions about whether to include

digital elements in goods or offer them as downloadable add-ons, to get the producer the more favorable body of law. 30

31 Overall, as applied to software transactions, the amendments would be a step backwards, further confusing rather than clarifying the law.

The bias of the Article 2B project in favor of the software industry ultimately doomed it. In 1999, the ALI pulled out of the Article 2B project, ending its status as a UCC Article. The major reasons included amorphous scope, complex and unclear drafting, overreaching into issues best left to intellectual property law, and

a failure to require pre-transaction presentation of terms even in Internet transactions. 32

34 the opposition

effectively organized. Since then, AFFECT and other opponents of UCITA have

succeeded in defeating it in every jurisdiction where it has been promoted.

AFFECT also has succeeded in persuading four states to enact so-called

“bombshelter” legislation, to protect customers from application of UCITA through non-negotiated choice of law or forum clauses. 35

who brought the perspectives of consumer advocates, large manufacturers, the insurance industry, private law practice, and technology journalism. 46 A simple version of the principles, under the name “Stop Before You Click,” was completed in the fall of 2004. A technical version, with greater detail and more progress toward implementing solutions in its commentary, was finished in January 2005. The first public presentation of the principles was given at the Shidler Center for Law, Commerce and Technology at the University of Washington on March 4, 2005, as part of a session on “New Rules for New Deals.”

Part Two: The Content of the Principles: Addressing Form and Substance of Deals

The 12 principles, without the commentary of either the simple or technical versions, are as follows:

I. Customers are entitled to readily find, review, and understand proposed terms when they shop.

II. Customers are entitled to actively accept proposed terms before they make the deal.

III. Customers are entitled to information about all known nontrivial defects in a product before committing to the deal.

IV. Customers are entitled to a refund when the product is not of reasonable quality.

V. Customers are entitled to have their disputes settled in a local convenient venue.

VI. Customers are entitled to control their own computer systems.

VII. Customers are entitled to control their own data.

VIII. Customers are entitled to fair use, including library or classroom use, of digital products to the extent permitted by federal copyright law.

IX. Customers are entitled to study how a product works.

X. Customers are entitled to express opinions about products and report their experiences with them.

XI. Customers are entitled to the free use of public domain information.

XII. Customers are entitled to transfer products as long as they do not retain access to them.

It should be emphasized that because these are principles, they do not resolve implementation difficulties that law reform will ultimately have to address. Also, the principles do not make trade-offs that could come in an implementation

process, such as, for example, a decision to trade less effective disclosure for more effective substantive policing of terms. The ideal reflected in the principles is attention to both process and substance of non-negotiated deals for mass-market digital products.

The principles can be roughly divided into four categories: (1) disclosure of terms and assent to them (Principles I and II)

Disclosure of and Assent to Terms

Principles I and II call for advance disclosure and active customer acceptance of terms. Advance disclosure and real assent, premises of a working mass market, are also prevailing norms in consumer protection and contract law. For markets to work best, customers must be able to find terms before they make a decision to purchase. Even where customers have no choice of competing products, such as when there is only one company offering a given kind of software, it is still useful to have the terms in advance because some customers will make the choice to avoid using that type of software entirely if the terms are sufficiently undesirable. In addition, substantive limits on terms, discussed below, are needed to police against unfairness in terms. Although disclosure alone is not sufficient to deal with market weakness or failure, it should be preserved as part of an overall strategy.

Advance disclosure of terms facilitates some shopping for the best terms and permits customers to avoid deals that are inappropriate for their needs. When customers only find out the terms after payment and delivery of products, shopping for the best terms is likely to be discouraged. Even a small number of customers shopping over terms can introduce some weak market policing.

To be most useful, disclosure must occur not just before a customer makes an order for a product, but also before the customer makes a decision to order. Thus, terms should be easy to find on the public parts of Web sites, so that they can be researched during shopping, well before submission of a credit card number, something that most customers would not do unless they had already made a

decision to acquire. If a customer has already made a decision to enter into a transaction, it causes cognitive dissonance

Robert Hillman’s pilot study, Online Consumer Standard Form Contracting Practices: A Survey and Discussion of Legal Implications, provides some evidence of shopping for terms on the Internet. His study of his own first-year law students reading and shopping habits revealed that Cornell law students do not seem to enter into many online transactions, perhaps because of the demands on their time and their pocketbooks. For this reason and also because of their high levels of analytical and reading comprehension skills, these subjects probably are not typical Internet shoppers. More research is clearly needed, with a bigger sample that is representative of customers in the mass market, including business customers.

In the mass market generally, business customers are more likely to shop over terms than consumer customers. This is a potentially important effect of treating mass-market transactions as a category that goes beyond consumers in the sense of purchasers for personal, family, or household purposes. When terms are first presented after delivery of digital products, during installation, as a practical matter businesses cannot police technology workers to keep them from clicking through the terms. If the terms were available online prior to order, however, it would be easier for the lawyer of a business customer to review terms before the business made a purchase decision. Shopping by businesses could have spill-over benefits

for consumer customers, at least to the extent sellers are not successful in segmenting these parts of the market for mass-market products. 50 Furthermore, with public online availability of terms, journalists and bloggers can publicize bad terms and get the word out to more customers in a manner that is more readable than the usual standard form.

Active assent is another important principle because non-return or use of a product does not usually mean real assent, especially to delayed form terms. Professor James J. White has called it “coercive” to require return and non-use to avoid delayed terms, using the colorful example of a customer sitting in his den in the winter in International Falls, Minnesota, in his underwear when he first encounters terms and must decide whether to get dressed and go back out into the cold to drive back to the store and return the product in order to avoid bad terms.

Delayed terms, with the seller attempting to define non-return as acceptance, can also be seen as a more virulent form of bait and switch

Principle III calls for easy access to plain language descriptions of known nontrivial product flaws, fixes, and incompatibilities. Sellers should have reasonable time to evaluate and address defects, including security breaches, before disclosing details of any defect. They also should give notice of dangers and possible fixes. The principles do not make specific implementation recommendations concerning liability for undisclosed known nontrivial defects, but one way to create incentives for compliance with this principle would be to make sellers liable for consequential damages when they make decisions not to disclose known flaws that pose significant risks of causing foreseeable harm.

Principle VI states that customers are entitled to control their own computer systems. With software that remains in contact with the Internet while in use, there is new potential for outsiders to interfere with computer systems and to export data without the knowledge or permission of the customer. Security backdoors for selfhelp, discussed below, are one cause of this increased security and privacy vulnerability. Even the presence of backdoors should not be permissible in digital products, absent the strong disclosure and assent involved in a negotiated contract. Furthermore, producers should take reasonable steps to ensure that products are free of viruses, spyware, other malicious code or security problems that compromise computer systems.

Principle VII provides that customers also are entitled to control their own data, which includes having continuing access to it. Customers also should be informed of the purposes of any transmissions of information from their systems before being asked to agree to them.

Remedies and Dispute Resolution

A third focus of the principles is remedies. The principles provide for a minimum adequate remedy and for access to dispute resolution for customers, while protecting against unfair remedies of sellers. Principle IV calls for a basic remedy of refund of the price for products that do not meet reasonably expected product quality in light of the performance claims of product descriptions. The AFFECT principles do not contemplate liability for consequential loss as a routine remedy

for quality defects, absent personal injury, and this change would be an improvement in the law for producers compared to current law. 61

Principles V and VI address potential unreasonable remedies of sellers. A nonnegotiated term providing for a non-local forum, if enforceable, means that a producer may sue a customer in a remote place, claiming violations of surprising use and transfer terms in fine print. With ordinary hard goods not involving software, requiring customers to use a remote forum may mean they have no real remedy for quality problems, because it would be too costly to litigate far from their home base. The situation is even worse with digital products, because customers are more likely to be named as defendants in a distant place. With surprising use and transfer restrictions in many digital product deals, customers could find themselves liable for default judgments for damages with no practical way to defend themselves. Furthermore, it is possible that violation of use and

transfer restrictions in form terms could make a customer an infringer, subject to the strong remedies of the U.S. Copyright Act, 62 but unable to afford litigating remotely. For these reasons, Principle V calls for a local forum in litigation over mass-market digital products deals.

Principle VI takes the position that producer “self-help” involving remote disabling of software products threatens disproportionate damage to customers through loss of access to critical functions and data. Sellers should not be authorized to use this type of self-help absent negotiation for such a term. This is a particular concern of business users. Furthermore, undisclosed backdoors put in software to enable remote disabling are not fair because they create unacceptable security risks. Only by negotiation and clear assent should it be permissible to put into software products technology that permits remote shutdown.

Protecting Fair Use, the Public Domain and Transfer of Copies

This is a reason that implementation of the principles ultimately would probably best be achieved by federal legislation, but states could first experiment with contract regulation to head off use of non-negotiated terms for mass-market products to undercut federal law and policy, which could also be recognized as state policy. Principle VIII states a general principle that fair use of copyrighted material should not be limited by non-negotiated terms for mass-market products. Libraries must be able to lend materials without keeping track of and policing surprising restrictions in mass-market terms. Libraries do not have the personnel to read, understand, and insist on compliance with the huge variety of restrictions written into mass-market product terms. As with books, they need one set of rules governing mass-market digital content products. If producers want to restrict libraries’ lending activities concerning mass-market products, they should have to negotiate for those restrictions and not be able to rely on form terms that could thwart the mission of libraries. Furthermore, users generally should be able to

licensed copies should not be burdened by the need to negotiate for releases. In addition, digital products and digital components of products need to be transferable to avoid eliminating competition from second-hand products, which drives down the cost of new products. Transferability of digital products such as ebooks is also important to permit libraries to fulfill their mission. Much needed is a concept of digital first sale that is not subject to limitation by fine print terms. If enforceable, mass-market terms restricting fair use and transfer would take away the balance of owner and user rights in current intellectual property law.

Scope Does Not Include Open Source Software

68 and it does not prohibit testing, study, comment, copying or transfer. A lack of disclosure of flaws is not a problem with this software. Development to improve the software is a cooperative, open venture. The license’s objective is not to restrict the public domain, but rather to keep open source software in the public domain. Furthermore, the software itself is free, so that a refund of the price is clearly not an appropriate remedy for quality problems. The community that contributes to open source development is too diffuse to hold responsible for quality, and users do not look to this community as warrantors. Service providers sometimes provide support for open-source software on a paid basis and make warranties, which ought to be enforced, but the AFFECT principles have not for now addressed issues in support services for digital products, whether closed or open source. Dealing with service issues is a project for another day. Furthermore, the distribution system for open source software is new and quickly evolving. It may be too soon to develop principles for the changing array of transactions involving open source software.

Part Three: Theory of the Principles

The approach of the principles, addressing both the process and the substance of transactions in mass-market software transactions, has a long tradition in the law of contracts, commercial transactions and consumer protection. Three theoretical points are fundamental underpinnings of the principles: mass-market nonnegotiated terms represent a special problem in contract and commercial law

supplied by law and variable only by negotiation, is a promising possible way to deal with this problem.

As to form terms in general, Llewellyn believed that there is no assent even of a blanket nature to unreasonable terms. This idea was incorporated into the Restatement (Second) of Contracts, especially in a comment stating:

Although customers typically adhere to standardized agreements and are bound by them without even appearing to know the standard terms in detail, they are not bound to unknown terms … beyond the range of reasonable expectation. …. [A] party who adheres to the other party’s standard terms does not assent to a term if the other party has reason to believe that the adhering party would not have accepted the agreement if he had known that the agreement contained the particular term. Such a belief or assumption may be shown by the prior negotiations or inferred from the circumstances. Reason to believe may be inferred from the fact that the term is bizarre or oppressive, from the fact that it eviscerates the non-standard terms explicitly agreed to, or from the fact that it eliminates the dominant purpose of the transaction. … This rule is closely

related to the policy against unconscionable terms and the rule of interpretation against the draftsman. 74

The Restatement makes unenforceable standard form terms that are beyond reasonable expectations, thus explained.

The Restatement concerns the common law, which involves a case-by-case approach to law-making. Furthermore, the reasonable expectations doctrine, like unconscionability, uses a standard, giving courts great discretion. Early in his career, Llewellyn noted the tension between flexibility and predictability in the law and saw statutes as a way to achieve greater predictability. In a book originally published in German in 1933, he wrote in a section on “Desirable Interaction of Precedent and Statute” that case law is a good way to begin to deal with a problem, but that once enough case law and experience has amassed “to make an incisive

diagnosis possible, a statute can move much more directly and efficiently toward its real goal than the pure tradition-bound case method.” 75 He also saw that consumer protection was the type of problem that needs a statute: “Should the task facing the court be simple and not very broad in scope a complete solution can be effectuated entirely within the framework of case law methodology…. For a problem as enormous as consumer protection, however, this is not possible.”

This reasoning is on target as applied to efforts to achieve fairness in the mass market for digital products

Efective Advance Disclosure of Terms

The controversial idea that disclosure of preplanned contract terms used in a massmarket transaction can be delayed until after delivery of the product without

77 It is out of step with strong norms of commercial law and consumer protection. Commercial law disfavors delayed material terms, 78

79 It is also ironic that the information technology industries resist creating market transparency concerning information relevant to deals in their products. Information technology makes it possible to store and retrieve vast amounts of information cheaply.

Of course, disclosure is not a complete or necessarily even a major solution to the problem of unfair terms in the mass market for digital products. Although advance disclosure of terms may have weak effects in stimulating shopping, it is the least interventionist approach to correct for market failure and weakness and worth preserving as part of the set of tools to address unfairness in digital product

terms. Alan Schwartz and Louis Wilde made the argument a generation ago that even a small number of shoppers can create competitive pressure on sellers. 80 However, delaying availability of terms until after delivery puts a heavy, costly burden on a customer who wants to shop

If all customers in the mass market are entitled to advance public disclosure of terms, the shopping by businesses that are included in that category can have spillover benefits for consumer customers. This is a benefit to consumers of addressing the problems of business customers in the mass-market for digital products along with the problems of customers acquiring products for personal, household, and family purposes.

The reason for inclusion of mass-market business customers in the scope of the principles project is the reality that businesses are also subjected to unfair massmarket terms. Businesses have not been able to avoid these terms for multiple reasons. To a great extent, business customers face the same problems as consumers when they acquire digital products. The costs of understanding complex terms, such as the impact of a choice of law clause, can be much greater than the amount at stake in a transaction.

terms, and even if this were possible, it would be exceedingly cumbersome to reverse transactions after delivery in order to avoid bad terms. Increasingly, new terms come with each update, making the burden of policing workers and returning products next to impossible. Sometimes the terms for updates purport to override even negotiated earlier contracts, potentially threatening the ability of business customers to protect the security of transactions.

A lack of competition in terms affects all mass-market customers, not just consumers. Some digital products lack alternatives, or even where there are alternative products, unsuitable and unfair terms appear in the forms of all producers. The market is failing to meet business needs for products that protect the security of their systems and the privacy of their and their customers’ data. Producers use sweeping disclaimers to protect their practices of not addressing or even disclosing product flaws, even though effective disclosure of flaws and features that create vulnerability could help to reduce business customers’ costs of dealing with them.

83 William Whitford has nonetheless defended

disclosure as a technique with potential in some settings, particularly oligopolistic

markets in which sellers make parallel decisions not to compete and wish to

withhold information about disadvantageous terms to avoid convincing some consumers not to make any purchase. 84 This is a promising theory to explain why digital product producers seek to avoid advance disclosure of their terms, not only to consumers but also to business customers.

Given the lack of remedies and the limitations on use in many producers’ terms, delayed disclosure seems to be a good way, from the seller’s perspective, to frame transactions to deemphasize these unattractive aspects of deals. AFFECT, however, takes the position that the law should not permit this framing strategy because of its impact in reducing market pressure for better terms. Advance disclosure in the digital era is cheap, probably cheaper than getting terms onto disks or printed on pieces of paper and into boxes. Thus, advance disclosure need not provide much benefit to be justified under a cost-benefit analysis.

The Need for Specific Policing of the Substance of Terms

The principles project adopts the approach of insisting on advance availability of terms but does not consider effective disclosure a sufficient check on unreasonable terms. Substantive policing is also needed. A key question is what form of law should be used to address substantive unfairness in mass-market transactions. Llewellyn himself recognized the unpredictability of case-by-case policing using the common law and the need for greater precision where possible through

statutory solutions. Arthur Leff built upon Llewellyn’s insight that consumer protection is too enormous a task to achieve effectively by case law and argued that the availability of judicial policing under vague doctrines such as

unconscionability should not distract us from the hard work of particularized legislation or administrative regulation. 87

88 Furthermore, when producers fail to

conform to specific requirements, their noncompliance is clear and it will be easier

for a customer or agency to win a lawsuit. Another key insight from consumer

protection law is that making terms unenforceable may not be enough of a

sanction

customers, making it advisable to have outright prohibitions on putting certain terms in contracts, with sanctions for doing so. 89 Thus, following the lessons of Llewellyn, Leff, and Whitford and applying them to the mass market rather than just to consumers in the narrower sense of purchasers for personal, family and household purposes, the AFFECT principles call for substantive checks on specified unreasonable terms in mass-market digital product transactions. A narrow focus on mass-market digital product deals makes specificity possible. Three substantive areas are addressed: (1) reducing the impact of product flaws and vulnerabilities (by requiring disclosure of known product flaws to reduce customers’ costs of dealing with them, and by prohibiting use of certain product features, such as remote self-help, absent negotiation)

Conclusion: The Uses of the Principles and Possible Follow-On Projects

The very process of writing the principles has been valuable to the customer coalition that is AFFECT. Shared opposition to UCITA drove formation of the coalition, but preparing the principles has solidified the sense that there are common problems among consumer, library, and business customers of digital

products. It is relatively rare that customers of any particular kind of product have enough of a shared sense that the market is not serving them well to lead them to band together to seek solutions through law reform. Furthermore, writing the principles has clarified that solutions to common problems would have benefits to customers generally and to the public interest more broadly. This gives AFFECT a mission that goes beyond economic self-interest to reinforcing important values in our culture. Fair use is an important part of an open, competitive, and innovative society.

Beyond clarifying AFFECT’s mission, the principles have many potential uses. They provide a simple explanation of what customers are seeking, in broad conceptual terms, in the way of solutions to their current problems and to the broader problem of preserving fair use and alienability of copies and the benefits that come from these rights. The principles are an effort to provide public education and to inform policymakers. In addition, the principles may stimulate more public debate and generate better understanding of the stakes involved in non-negotiated terms for mass-market digital products. These terms threaten to dampen competition and constrain follow-on innovation. For example, if terms barring comment, testing, and study are enforceable, this would upset the balance in intellectual property law between giving innovators incentives while permitting fair use that leads to more competition and innovation. Not just fairness in particular deals is at stake, but also sound information and competition policy.

In addition to their public education uses, the principles could be the basis for producer self-regulation. A Fair End User License Agreement (FEULA) is a

possible follow-on project. An early version of such a license, although not formally approved by AFFECT, has already been published online. 90 Producers could choose to follow the principles, gaining customer approval and removing pressure for law reform in the process.

Another possible function of the principles is to identify areas where customers should litigate. The principles may encourage and facilitate more effective challenges to delayed disclosure and fictional assent, as well as to particular terms attempting to restrict fair use and transferability of digital products and access to public domain material. Some of the early litigation concerning digital products was characterized by a dramatic imbalance in the resources and sophistication brought to bear by producers as opposed to customers.

Ultimately, the principles are a guide for a first stage of law reform. Thus, another follow-on project could be model legislation, such as a Model End User Licensing Act (MEULA). UCITA was too broad a project for a first effort at commercial law of “computer information.” The principles, by focusing on one type of transaction, carve out a more realist scope to tackle in a first legislative

assault. The technical feasibility of targeted legislation becomes clearer when one sees that it is possible to state in 12 principles most of the major issues of concern to a very diverse set of mass-market customers, although political feasibility is perhaps a more difficult challenge.

The principles do not address issues already adequately dealt with in current UCC Article 2, suggesting that model legislation could be an overlay on Article 2, which could continue to apply to digital product sales. This approach is possible when one conceives of a mass-market digital product transaction as both a sale of a copy and a license of its use. MEULA would only need to take on issues that Article 2 does not address, such as stating explicitly requirements for effective advance disclosure of terms for mass-market digital products, setting default use terms, requiring disclosure of known nontrivial product flaws, providing for transfer for the same type of use, and prohibiting non-negotiated form contract erosion of fair use. AFFECT’s 12 principles thus set the stage for legislation to address the most obvious problems in mass-market transactions in digital products.

Notes

See Official Text, Uniform Commercial Code (U.C.C.) Article 2 (2002) [hereinafter, current Article 2]. Amendments approved by the sponsors in 2003 have not been enacted anywhere. See Official Text, U.C.C. Article 2 (2005) [hereinafter proposed Amended Article 2). See also U.S. Copyright Act, 17 U.S.C. §§ 101-1332

See Annalee Newitz, Dangerous Terms: A User’s Guide to EULAs, www.eff.org/wp/eula.php (collecting examples of unfavorable, bizarre and onerous terms). See also www.fairterms.org/EULALibrary.htm (collecting links to sites that feature unfair terms) (last visited Jan. 21, 2006). Unfairness and gross inefficiency are treated as synonymous in much consumer protection law. See Jean Braucher, Defining Unfairness: Empathy and Economic Analysis at the Federal Trade Commission, 68 B.U.L. REV. 349 (1988).

See CARL SHAPIRO & HAL R. VARIAN, INFORMATION RULES: A STRATEGIC GUIDE TO THE NETWORK ECONOMY 13 (1999) (concerning network effects).

See id. at 11-17 (concerning lock-in, switching costs and the competition for market share to get the benefits of network effects).

See MARK A. LEMLEY ET AL., SOFTWARE AND INTERNET LAW 539-641 (2000) (concerning competition law in the digital industries).

See http://www.fairterms.org/ (giving two versions of AFFECT’s 12 Principles for Fair Commerce in Software and Other Digital Products) (last visited Jan. 21, 2006).

See id. (listing full drafting committee in technical version of the principles).

See, e.g., Mark Lemley, Beyond Preemption: The Law and Policy of Intellectual Property Licensing, 87 CAL. L. REV. 111 (1999)

See Official Text, Uniform Computer Information Transactions Act (2002) [hereinafter, UCITA]. There was an earlier official text as of 2000 that was the basis for enactments in Virginia, VA. CODE ANN. 59.1-501.1 to 59.1-509.2, and Maryland, MD. CODE ANN. 22-101 to 22-816. For information about the National Conference of Commissioners on Uniform State Laws, see http://www.nccusl.org (last visited Jan. 21, 2006).

See UCITA, supra note 9, at Sections 103(a) (scope is “computer information transactions”) and 102(a)(9), (10), (11) (defining computer, computer information, and computer information transactions in very broad, general terms).

A high level American Bar Association panel, whose concerns led the ABA not to endorse UCITA, supra note 9, criticized this law for “audaciously” trying to cover virtually every issue in every kind of “computer information” transaction. See ABA, Report on the Uniform Computer Information Transactions Act (UCITA) 7 (Jan. 31, 2002), available at http://www.abanet.org/leadership/ucita.pdf (last visited Jan. 21, 2006).

See text infra at note 68, for a discussion of why the principles are not appropriate for open-source software.

UCITA, supra note 9, used a version of the mass-market concept, but defined it very narrowly and deployed it misleadingly, for counter-productive purposes. See Jean Braucher, The Failed Promise of the UCITA Mass -Market Concept and Its Lessons for Policing of Standard Form Contracts, 7 LEWIS & CLARK J. OF SMALL & EMERGING BUSINESS LAW 393 (2003) [hereinafter Braucher, Failed Promise].

See Permanent Editorial Board Study Group, Uniform Commercial Code Article 2, Preliminary Report (1990) [hereinafter PEB Article 2 Report].

See Robert Braucher, The Legislative History of the Uniform Commercial Code, 58 COLUM. L. REV. 798 (1958).

See RESTATEMENT (THIRD) OF TORTS § 19, cmt. (d), supra note 1 (noting that the courts have treated mass-marketed software as goods under current UCC Article 2).

See PEB Article 2 Report, supra note 14, at 39.

Id.

See Richard E. Speidel, Introduction to Symposium on Proposed Revised Article 2, 54 SMU L. REV. 787, 791 (2001) (noting that most commercial sellers and buyers are content with current Article 2, that no interest tried to capture the Article 2 revision process, and that affected interests view current Article 2 as “not broke.”) There is reason for consumer advocates to be dissatisfied with current Article 2, because it relies heavily on vague standards that do not promote seller compliance in the way specific rules do

The first draft of Article 2B is dated Feb. 2, 1996. All drafts are available at http://www.law.upenn.edu/bll/ulc/ulc_frame.htm (last visited Jan. 21, 2006). See also Speidel, supra note 19, at 789-790 (noting that a combined project on sales of goods and

licenses of software was attempted in 1995 before NCCUSL split the project in two later that year).

See Linda J. Rusch, A History and Perspective of Revised Article 2: The Never Ending Saga of a Search for Balance, 52 SMU L. REV. 1683, 1714 (1999) (discussing attempts to harmonize Articles 2 and 2B with each other and with other uniform law projects) [hereinafter Rusch, History]. See also Speidel, supra note 19, at 792 (noting that after the Article 2 and 2B drafting projects were separated in 1995, there was continuing tension over the degree of textual harmony and also over the line between goods and “computer information”).

See Rusch, History, supra note 21, at 1689-1690 (stating that no interest group captured the original Revised Article 2 drafting committee).

See id. at 1686, 1689 (describing the NCCUSL leadership decision to interrupt the Article 2 revision project and appoint a new committee with a presumably more industry-oriented perspective).

See id.

See Linda J. Rusch, Is the Saga of the Uniform Commercial Code Article 2 Revisions Over? A Brief Look at What NCCUSL Finally Approved, 6 DEL. L. REV. 41 (2003) (describing changes in six areas—scope, formation and terms, warranties, performance or breach, remedies for breach, and third-party rights).

Proposed Amended Article 2, supra note 1, was introduced in Kansas and Nevada in 2005. See Kansas H. Bill 2452 (2005)

See proposed Amended Article 2, supra note 1, § 2-103(1)(k).

Other AFFECT objections to proposed Amended Article 2 include the failure to clearly reject enforceability of delayed terms in mass-market deals and the confusing and unnecessary provisions on electronic commerce that largely replicate the Uniform Electronic Transactions Act (UETA), but with different language. See http://www.ucita.com/Legislation.htm#two (listing main AFFECT objections to proposed Amended Article 2, supra note 1) (last visited Jan. 21, 2006). See also Jean Braucher, Amended Article 2 and the Decision to Trust the Courts: The Case Against Enforcing Delayed Mass -Market Terms, Especially for Software, 2004 WIS. L. REV. 753, 753-762 (concerning the problem of the new discretion the amendments would give courts, in particular concerning the scope of proposed Amended Article 2 and concerning enforceability of delayed terms).

See proposed Amended Article 2, supra note 1, § 2-103(1)(k), cmt. 7.

See Philip Koopman & Cem Kaner, The Problem of Embedded Software in UCITA and Drafts of Revised Article 2 (Parts I & II), U.C.C. Bulletin (2001).

See proposed Amended Article 2, supra note 1, § 2-207(b).

See ALI Council Ad Hoc Committee on Article 2B, Memorandum on Proposed UCC Article 2B (Dec. 1998) (on file with the author), and Jean Braucher, Delayed Disclosure in Consumer E-Commerce As An Unfair and Deceptive Practice, 46 WAYNE L. REV. 1805, 1840-1842 (2000) [hereinafter Braucher, Delayed Disclosure].

NCCUSL revised UCITA in 2002, see supra note 9. The Virginia and Maryland enactments, supra note 9, were based on the 2000 version, with some nonuniform amendments, some of which were the basis for the 2002 amendments made by NCCUSL to the Official Text.

See IOWA CODE ANN. § 554D.104

See supra note 11.

See UCITA, supra note 9, §§ 208, 209, 112-114.

See id. § 307.

See id. § 503(2) (generally permitting limitations on transfer of contractual rights, but with a few limited exceptions, including for a gift transfer with a computer to a public elementary or secondary school, to a public library, or from a consumer to a consumer, but not, for example, gifts without a computer or to a private school or to a public or private college or a second-hand sale by a consumer to a consumer). This 2002 NCCUSL amendment to UCITA is based on a provision enacted in Virginia’s version of UCITA, supra note 9, at § 59.1-505.3(2)(C).

See Jean Braucher, When Your Refrigerator Orders Groceries Online and Your Car Dials 911 After an Accident: Do We Really Need New Law for Smart Goods? 8 WASH. U.J.L. & POL’Y 241, 252-58 (2002) [hereinafter Braucher, Smart Goods] (defending price differentiation as in the consumer interest, and thus defending the need for licensing that enables this practice, but drawing a line at licensing that limits fair use).

See Arthur Allen Leff, Unconscionability and the Crowd—Consumers and the Common Law Tradition, 31 U. PITT. L. REV. 349, 357-58 (1970).

See UCITA, supra note 9, § 105(b) (providing that a court may refuse to enforce a contract term that is against a “fundamental public policy” … “to the extent the interest in enforcement is clearly outweighed by a public policy against enforcement”…. In contrast, the Restatement (Second) of Contracts, Section 178, does not include the “fundamental” limitation on public policy as a ground for non-enforcement of a contract term. See also Braucher, Failed Promise, supra note 13, at 415-416.

See id. at 413-415 (noting that the protection of public comment in UCITA Section 105(c) is only for comment on products made “generally available” and “in final form,” and extends only to comment by end-users, so that distributors and retailers are not protected, and magazines and developers who acquire products to test them and disseminate information might not be covered

See Virginia UCITA, supra note 9, at § 59.1-501.5(c) (provision with limited protection for public comment). The reverse engineering provision added to UCITA in 2002 as Section 118, see supra notes 9 and 43, has not been enacted in either Maryland or Virginia.

See Cem Kaner’s Blog, Software Customer Bill of Rights (Aug. 27, 2003), http://blackbox.cs.fit.edu/blog/kaner/archives/000124.html (last visited Jan. 21, 2006). See also Cem Kaner’s home page, http://www.kaner.com/ (last visited Jan. 21, 2006).

See http://www.fairterms.org/, supra note 6 (in the technical version of the principles project, giving a list of the full drafting committee membership).

My presentation at the Shidler Center on the AFFECT principles had the same title as this chapter.

See Alan Schwartz & Louis L. Wilde, Imperfect Information in Markets for Contract Terms: The Examples of Warranties and Security Interests, 69 VA. L. REV. 1387, 1450 (1983).

See William C. Whitford, The Functions of Disclosure Regulation, 1973 WIS. L. REV. 400, 426, 448-49.

See Clayton P. Gillette, Rolling Contracts As An Agency Problem, 2004 WIS. L. REV. 679, 692-93.

See Ed Foster, The Gripe Log, http://www.gripe2ed.com/scoop/ (last visited Jan. 21, 2006) and Newitz, supra note 2. Currently, those digital product producers who post their terms subject themselves to this scrutiny, while competitors who do not post terms are more likely to escape negative publicity.

See James J. White, Contracting Under Amended 2-207, 2004 WIS. L. REV. 723, 748.

See Braucher, Delayed Disclosure, supra note 32, at 1807, 1853.

See http://www.gateway.com/about/legal/warranty.shtml (and go to “Gateway Standard Terms of Sale”) (setting forth Gateway standard terms that require return of computer and software together and include a 15 percent restocking fee) (last visited Jan. 21, 2006)

http: //www.bestbuy. com/site/olspage. j sp?type=page&entryURLType=&entryURLID= &categoryId=cat10004&id=cat12097 (and go to “What is your return policy at Bestbuy.com?”) (providing for 15% restocking fee for return of a computer)

See supra note 54 for examples of terms that impose charges for returns.

See ProCD, Inc. v. Zeidenberg, 86 F. 3d 1447 (7th Cir. 1996)

But see UCITA, supra note 9, at § 304 (authorizing this form of “continuing contractual terms”).

See Gillette, supra note 50, at 692-97 (discussing weaknesses in market policing).

See David Bank, Companies Seek to Hold Software Makers Liable for Flaws, WALL STREET JOURNAL, Feb. 24, 2005, at B 1 (discussing how major customer companies are fed up with having to spend millions of dollars to fix problems caused by software). A

particularly dramatic example of a flawed product that caused substantial problems is the Sony BMG rootkit technology. See http://www. eff.org/IP/DRM/S ony-BMG/

See Tom Zeller Jr., Data Security Laws Seem Likely, So Consumers and Businesses Vie to Shape Them, N.Y. TIMES, Nov. 1, 2005, at p. C3 (noting 80 data breaches, affecting 50 million people, from February to November 2005), and Tom Zeller Jr., Identity Crises, N.Y. TIMES, Oct. 1, 2005, at p. C1 (reporting FTC data that there are 10 million identity theft victims a year and also noting that there are low tech as well as high tech causes).

Both the common law of contract, see RESTATEMENT (SECOND) OF CONTRACTS, § 351 (1981), and current Article 2, supra note 1, at §§ 2-712, 2-713, 2-714, and 2-715(2), provide for seller liability for foreseeable consequential damages as the default rule. AFFECT proposes to change this rule and instead to have refund of the price as a minimum remedy, to make it more feasible for producers of digital products to use default terms supplied by law, while providing a usable basic remedy to customers.

See U.S. Copyright Act, 17 U.S.C. §§ 501-505 (providing civil remedies for infringement, including injunctive relief, statutory damages and attorneys’ fees).

See Lemley, supra note 8, and Nimmer, supra note 8.

See Samuelson & Scotchmer, supra note 43, at 1580, 1614-15.

See id.

See http://www.switchboard.com (providing free access to nationwide white page and yellow page listings with names, telephone numbers and addresses of individuals and businesses

See Lemley, supra note 8, and Nimmer, supra note 8.

See http://www.opensource.org/licenses/apl1.0.php (last visited Jan. 21, 2006). See also MARC SMITH & PETER KOLLOCK, EDS., COMMUNITIES IN CYBERSPACE (1999). A public revision process for the general public license began in late 2005. See www.gplv3.fsf.org (last visited Jan. 21, 2006).

KARL N. LLEWELLYN, THE COMMON LAW TRADITION 370 (1962).

See Braucher, Delayed Disclosure, supra note 32, at 1846, 1850-51.

See Jon Hart & Steve Blumenthal, Software is Often Sold, Not Licensed, Despite What License Agreements Say, WALL STREET JOURNAL ONLINE, June 27, 2002 (setting out satirical conversation between a store clerk and a customer, in which the customer notes that he is acquiring a “perpetual license” to use the software and not buying the copy).

See current and proposed Amended Article 2, supra note 1, § 2-106(1).

See Braucher, Smart Goods, supra note 40, at 252-58.

See RESTATEMENT (SECOND) OF CONTRACTS, § 211, cmt. f (1981).

See KARL N. LLEWELLYN, THE CASE LAW SYSTEM IN AMERICA ix-x (1989) (concerning the original German lecturers of 1928-29 and the publication of the original book in German in 1933) and 67 (giving quote in the text).

See id. at 67-68.

See Braucher, Delayed Disclosure, supra note 32, at 1818-19. Misleading analogies are often made to manufacturers’ warranties, which can only add to a deal, or to terms required to be disclosed by federal law, such as those for airline tickets, or regulated by state officials, such as insurance policies. See id. at 1823-1825.

For example, current Article 2, supra note 1, at § 2-207(2) does not provide for delayed material alterations to become part of a contract even between merchants. In addition, comment 3 to that section contemplates that “express agreement” is needed to bring in material alterations. Also, unconscionability analysis includes a procedural dimension, and delayed disclosure contributes to “unfair surprise.” See current Section 2-302 and comment 1, stating “The principle is one of prevention of oppression and unfair surprise….” Furthermore, current Article 2 regulates disclosure of disclaimers of implied warranties. See Section 2-316(2) and (3).

See Federal Trade Commission Policy Statement on Deception, in Letter to John D. Dingell (Chairman, Subcommittee on Oversight and Investigations, Committee on Energy and Commerce) Oct. 14, 1983, reprinted as appendix to In re Cliffdale Assocs., 103 F.T.C. 110, 175 (1984)

See Schwartz & Wilde, supra note 48.

See Gillette, supra note 50, at 692-693.

See William J. Woodward Jr., Neoformalism in a Real World of Forms, 2001 WIS. L. REV. 971, 989-90.

See Whitford, supra note 49, at 403-427.

See id. at 429-430.

See Braucher, Delayed Disclosure, supra note 32, at 1812-1813.

See John Hanson & David Yosifon, The Situation: An Introduction to the Situational Character, Critical Realism, Power Economics, and Deep Capture, 152 U. PA. L. REV. 129 (2003) (concerning the evidence that situations affect human behavior more than preexisting dispositions).

See Leff, supra note 41, at 357.

See William C. Whitford, Structuring Consumer Protection Legislation to Maximize Efectiveness, 1981 WIS. L. REV. 1018.

See Federal Trade Commission Credit Practices Rule, 16 C.F.R. § 444.2 (making it an unfair practice to put certain terms in credit contracts).

See Ed Foster, Signing Up for The FEULA, The Gripe Log, March 25, 2003, http://www.gripe2ed.com/scoop/story/2005/3/25/870/70193 (concerning interest in his draft Fair End User License Agreement) (last visited Jan. 21, 2006).

See ProCD v. Zeidenberg in Context, 2004 WIS. L. REV. 821, 827-829, 833 (providing transcript of interviews with Matthew Zeidenberg and with his lawyer, indicating that the lawyer was sworn into the Wisconsin bar on the day that Zeidenberg contacted him about providing representation

Circuit and that his lack of knowledge about this appellate court allowed the software industry to get precedent that has been an impediment to achieving balanced law governing digital product transactions.