There are many illustrations of this failure to adjust terms to reflect mandatory law in a standard form. We can see it in standard form disclaimers of warranty in local transactions in states that ban broad disclaimers in the relevant transactions. It also occurs in drafts of guaranty and indemnity provisions, which are seldom updated.
But the most important context in which this problem occurs today involves adjusting forms to reflect modern (and ubiquitous) provisions of privacy and e-commerce laws. The issues here are addressed in an important article by Holly Towle, a partner at KL & Gates and a leading expert in both privacy and e-commercial law.
Her article and the general issue of keeping forms updated to reflect modern law and terminology need to be addressed.
]]>
An owner of a genuine and unmodified item purchased at an authorized sale may resell the item, describing it appropriately, even if doing so requires use of the trademark. Thus, if I own a Dell Laptop computer, I have the right to sell it describing it as a “Dell” computer. The right to accurately describe the resold property comes from first sale doctrine and from concepts of nominative fair use.
Issues about first sale or fair use almost never arise where the reseller sells only one item. They arise when the reseller resells multiple items. In this setting, where the items resold are unused, the reseller may become the direct competitor of the mark owner and its authorized dealers. The question is under what conditions trademark law permits this. Viewed from the mark owner’s perspective, this involves its ability to control marketing channels for its product, including the online marketplace, and to protect its trademark.
Not all resale activity is allowed even if the reseller owns the item.
Trademark owners’ protected interests that continue to apply relate to maintaining the trademark as identifying the source of the goods or services involved, which is the bedrock of trademark law. The balance then is between allowing the reseller to deal with its property while preventing confusion as to its immediate source or denigrating of the value of the mark.
The most commonly litigated limitation centers on whether the reseller creates the appearance of a relationship between it and the mark owner. Basically, while the reseller can use the mark to describe the product it resells, the reseller cannot create the impression that it acts with the sponsorship or authorization of the mark owner (unless, of course, that is true). Thus, while a used car seller can sell and advertise that it sells Mercedes cars, it cannot falsely convey the impression that it is a Mercedes authorized dealership.
Some settings make clear that no sponsorship is present. For example, in Tiffany (NJ), Inc. v. Ebay, Inc., 576 F.Supp.2d 463 (SD NY 2008) the court held that Ebay and its auction resellers of genuine Tiffany products were protected by first sale concepts even though the sellers referred to their products as Tiffany and Ebay advertised the availability of Tiffany products on its auction website. In context, there was no reasonable inference that Tiffany sponsored, authorized, or was in control of the sales of Tiffany products in this environment.
Lines are drawn based on language and context. Thus, in another online case, text on a site reselling diet supplements from various manufacturers originally commented as to one manufacturer that “we” are engaged in various steps. This would have established a claim that affiliation was being suggested, but the text was removed.
Small words often make big differences.
The cases also present issues associated with the mark owner’s obligation to monitor the quality of goods or services offered under its mark. This translates into the rule that the first sale doctrine does not apply to resale where material changes have occurred in the product. An unlicensed difference between two products bearing the same trademark is material if it “confuses consumers and impinges on the trademark holder's goodwill.”
The material changes that vitiate the first sale defense vary. They often relate directly to changes in the product (e.g., reprogramming an aspect of software), but can apply to other elements related to it, such as warranty terms. The mark owner has a continuing interest in controlling changes that on resale diminish or, even, simply alter the product in a manner relevant to the customer, while still using the trademark. Changes, when not clearly made known to the customer or the public as not coming from the mark owner or with its endorsement, threaten the goodwill built up by the manufacturer and diminish the ability of the mark to identify the source of goods and the predictable quality.
The limited right to use a mark without implying affiliation or sponsorship by the mark owner extends into all marketing systems, online and otherwise. Thus, keyword advertisements or metatags may fall within the protective scope of the first sale and related fair use protection, unless circumstances create the appearance of an affiliation between the reseller and the mark owner or what is being sold is a materially altered product. Needless to say, however, this does not give the online vendor the right to a site that is named Dell Computers and sells Dell along with Toshiba etc.
So the message for online and other vendors is the same here as in patent and copyright law. Merely “owning” a product or a copy does not give unlimited rights in it when intellectual property interests are involved. The IP rights remain dominant.
]]>Merely posting a work online does not relinquish all rights. As in other environments, merely placing property in public does not release property rights. The Internet context, however, may indicate that some actions with respect to the work are implicitly permitted.
]]>The proper answer to what is permitted with reference to a work posted by its author in any case depends in part on what harm the third party conduct causes for the copyright owner and on the circumstances of the posting. A copyright owner that posts a work but says – “don’t make permanent copies” and “don’t send it to others” has a right to do so.
This is a circumstance that in law that juxtaposes the copyright owner’s property interests and supposed social interests. In this case, the relevant social interest lies in reasonable expectations in the media. But the copyright interest should not be sacrificed to accommodate competing commercial systems in which another desires to exploit the person’s copyright without permission. Simply put, placing a work into a digital environment may be a reason in law for implying limited permission (license) to use that work, but only in narrow, non-commercial ways.
Consider an author who places her copyrighted poem online on the first page of a site that requires neither password, nor express permission to view that page. A user who comes across the site, views the poem (thus making a brief copy of it) and then leaves. This user did not infringe the copyright. Where and how the work was placed implies permission to do what the user did. Does the posting imply permission to make a permanent copy of the poem in the user’s computer – no. Does it imply a license to make and distribute multiple copies to third parties – no. Does it imply the right to commercially exploit the work – no.
Implied licenses in law arise only in narrow circumstances and, then, only within narrow limitations. When created, they are typically subject to being revoked at will. Having posted works online does not preclude the copyright owner from rescinding that permission, in which case, subsequent use is infringing.
Especially online, there is often a close connection between an implied license analysis and an analysis grounded in the express terms of an online license. See Raymond T. Nimmer & Jeff Dodd, Modern Licensing Law ch. 10 (2009). Implied licenses (permissions) cannot supersede express terms. If I say one thing, a court should not permit you to act as if I meant a different thing.
The policy tension here entails a conflict between rights owners who desire to retain control of their works, while still enjoying the benefits of online systems, and those who desire an Internet environment in which information is free, interrupted only in clearly stated rights-based limitations. Inevitably, law and practice will reach some balance between these. But some courts push too far favoring an environment in which property rights are not effectively maintained.
The court in Parker v. Yahoo!, Inc., 88 USPQ2d 1779 (ED Pa. 2008) approached this point in considering the copyright implications of the ordinary practice by search engines in making and displaying cached copies of online pages or works. Parker argued that, in the absence of an express license, this was infringement, but the court held that there was an “implied license”. The source of this implied license apparently was in the mere act of placing copyrighted material online and not taking technological or other steps to prevent the search engines from making the copies. The court’s explanation:
Parker acknowledges in his complaint that the defendants honor “electronic protocols” that would prevent the search engines from displaying a “cached” copy of his works … He also acknowledges that the defendants remove offending content upon request. … Parker contends that he has provided constructive notice to the defendants that he has not granted a license because he registered his works and included a copyright notice on his website. The Court is persuaded that Parker's complaint conclusively establishes the affirmative defense of implied license. At the very least, [it] suggests that Parker knew that as a result of his failure to abide by the search engines' procedures, the search engines would display a copy of his works. From Parker's silence and lack of earlier objection, the defendants could properly infer that Parker knew of and encouraged the search engines' activity, and, as did the defendants in Field, they could reasonably interpret Parker's conduct to be a grant of a license for that use.
Flip this into an affirmative statement and you will see the problem. This ruling requires that a copyright owner affirmatively give notice or use technological controls to prevent copying or lose the right to preclude at least this type of copying. But the source of the obligation to take affirmative steps lies in the court’s view that search engines are a structural part of the Internet and that rights owner’s must modify their conduct to accommodate the commercially profitable business that runs the search engines. That is wrong.
It is a tradition in U.S. contract law that mere silence does not generally constitute acceptance. So also in copyright law. But silence and inaction here was found to establish a quasi-contractual license. Parker inverts the meaning of property rights. It reduces the need for the search engines to negotiate permission from rights owners, but at the cost of control that otherwise would be vested in the property rights owner.
This should not be the law, nor is it likely to be. In any event, not all copies can be treated as being within the scope of an arguable, implied license, since this license must have some limits to it. Defining those limits should place close attention on protecting the interests of the copyright owners.
]]>The Google Book Settlement (GBS) would give Google unprecedented power over hundreds of thousands of copyright owners and control of an asset that may become essential to 1) book publishers, 2) book authors, and 3) any entity that desires to operate a search or archive function regarding published books. The Settlement should not be approved unless it ensures all competitors and others access and fairly protects the copyright owners who were not represented at the “settlement” discussions and whose property is being stolen.
]]>Over-arching any support for the Google “Settlement” is the view that this deal may create a cool and valuable “public” resource and that, without the settlement, the resource may be impossible to create. But the impossibility issue relates to the fact that the asset cannot economically be created without trampling on the property rights of hundreds of thousands of copyright owners. How to negotiate a license with the hundreds of thousands of copyright owners that exist and the many that would say no? The GBS solves this by answering – “no need to do so, Google is doing a good (and profitable) thing, and the property owners should be happy.” The fact that in the end there may be a valuable asset available to some of the public should not distract us from the reality that this asset will be under the control of a private, large and profitable corporation. And no one else will be realistically able to replicate it. Ever.
Google is not altruistic, nor should it be. Making a profit is a good thing, for Google. It is doing what it proposes to do in the “Settlement” for profit.
What source of profit does Google expect? That is simple. If the settlement is approved and the project completed, Google will have a perpetual lock on the digitized services involved in reference to the books, to archiving books in digital form, and to provide search services with respect to that archive. The advertising and subscription income from this monopoly will be immense. Monopolies are not illegal. But monopolies created by joint arrangements imposed on other parties should be. As much as the cost of creating the copies is huge, the profit from having created them and the market power it will give Google are far greater. And the market control will not end. All new authors and new copyright owners, though not expressly covered by the “Settlement” are likely to find irresistible the need to allow their works to be posted in the only game in town. If your book is not there, where is it and how do I find it?
In a recent decision, the Court of Appeals for the Federal Circuit observed that a patent pool, created by agreement in order to implement a jointly developed standard, might be engaged in patent misuse to the extent that the pooling arrangement contained provisions that limited use of affected patented technology to develop a potential competing technology. I am no fan of intellectual property misuse doctrines. But this use of the “pooled” copyrights (voluntarily or forcibly pooled) to prevent development of competing technologies is an inherent feature of the Google Settlement.
Is the GBS fair to copyright owners? Certainly not:
The GBS should be rejected.
If it is not rejected, the possessor of an essential asset that dominates a market and that was created by joint conduct (the agreement) should be required to provide low cost, non-discriminatory access to all others and its profit from this anti-competitive agreement and the asset it created should be closely regulated.
If it is not rejected, has the court not created a private entity that wields eminent domain powers with respect to copyright? Google would possess an asset dominating a market, created by joint conduct (the agreement), and vested with the public power to take the copyright interests by judicial order. In such circumstances, why should it not be treated like a utility, common carrier, or essential facility that should be required to provide low cost, non-discriminatory access to all others and its profit from this anti-competitive agreement and the asset created with the power and sanction of the court should be closely regulated.
]]>This is not a settlement of a judicial dispute, it is an attempt to reorganize and concentrate an industry into the hands of one company and its co-venture partners (former plaintiffs) without the participation of most of those affected by it. It is also legislation conducted in a courtroom.
The Google Settlement is a bad deal for everyone except for Google. For Google, it eliminates the cumbersome need to contact copyright owners for permission to copy their works, and the need to worry about their suing if Google acts without permission. For all companies who want to compete with Google, those needs remain. Google obtains a huge market advantage. It may be cumbersome to deal with property rights, but that is no cause to trample on them. If carried out, the deal will create a repository of digital books but do so by wildly restricting the rights of copyright owners to control their own property, and by creating a new monopoly – Google.
The settlement terms are over one hundred pages long. Indeed, very few people have actually read the “settlement”. No legislature has approved or drafted it, and none will be asked to do so.
It is drafted as a complex commercial transaction. That, of course, is what it is. But those who are copyright owners: did you authorize Google and the narrow organizations that sued it to represent you in a commercial deal transforming your rights? For most, the answer is no.
There are so many, obvious public policy objections that it is almost not necessary to mention them, but let me start with two concerns.
First, the “settlement” purports to bind copyright owners who, by the parties’ own admission cannot be found or identified. This is not an issue of how many of them there are, but supposedly about the difficulty of establishing proper provenance through analysis of transfers, inheritance, and the like. But if Google cannot find them as part of a billion dollar project, how are they expected to identify their interests, hire a lawyer, and do so before the date set for withdrawing from the “settlement”?
Second, the “settlement” purports to resolve claims that do not yet exist. This is not a case where bad acts have occurred, but damages are not yet manifest, such as in some mass tort litigation. This settlement covers future conduct that, if it occurs, would infringe a property right. Even Google (a huge company with massive assets) has not yet copied more than a small fraction of all books protected by copyright law. If the settlement were approved, it would cover all copyright owners in all books registered in the U.S. Copyright Office, including those that have not yet been copied by Google. This reaches too far. In my opinion, these uninjured owners could not even file a lawsuit against Google if they wanted to do so since no infringement has occurred in reference to their works. But they purportedly are bound by the result of a lawsuit they never authorized. Do not let advocates of the settlement claim that this is like class action litigation in mass torts – in those cases the class only covered people that had been affected by wrongful acts at or before the time of the lawsuit.
Consider this: You own a house in a large subdivision. Big Company (Google) plans to build towers on each property in the subdivision to create a commercial product for its use. There is no government action or regulation involved. Big Company is sued by three landowners on the north side of the subdivision whose property has already been wrongfully invaded (you are on the south side and live in pristine ignorance of the plan). The plaintiffs and the Big Company transform the case into a class action covering all owners in the subdivision and “settle” it, giving Big Company the right to invade every lot in the subdivision, if it pays a small standard fee and a royalty for some of its future profits One year later, the crew shows up on your property and builds a tower authorized by the “settlement.”
Do you have a right to prevent them from digging up your yard and building their tower? Do you have a right to complain that the damage to your property value was far greater than to others because your lot was expensively landscaped (e.g., more valuable)?
Not under the Google Settlement. The settlement purports to authorize future and past infringing acts, rather than simply to settle claims arising out of prior infringement. It gives Google the right to “go forth and infringe.” This not only perverts the idea of class action, it is unfair to the millions of copyright owners who were not consulted and cannot truly participate. The unfairness lies in taking my property without consulting me and paying for it without distinguishing my loss from the loss suffered by others
Google was prepared to defend the lawsuit by claiming that copying books was fair use because it was not selling the copies, but only access to them as a commercial indexing tool to support a lucrative advertising and search engine business. There might be reasonable debate on this in law, but the settlement gives Google that right without the contest and effectively excludes any ability of copyright owners to contest the issue because of the economics involved. It also gives Google the right to copy and distribute some books - clearly not a fair use. I think it is unfair (and against the law) to take my property and to place me in a position where realistically I cannot effectively complain.
But Google might argue – the settlement gives you the right to withdraw the work if you choose. The settlement does not actually do so, but this is a form of an argument Google has made for years in all contexts: “we have the right to do what we want with your property unless or until you complain about our conduct.” This argument turns property rights upside down. Clearly, property rights grant a right to exclude others, but recognition of a property right in law also assumes the willingness of those in society to respect and avoid trampling on that right. The Google Settlement takes that away: my rights become your privilege.
Consider this: I see your car parked on a public street and decide to drive it. I enter the car, start it, and drive off, leaving you a note that, if you find me and object to my use of your car, I will return it. You find me and I return the car. Have I done anything wrong? Of course.
The Google Settlement adopts the view that property rights are merely a right to object and not a right to expect that people should respect them. This is a terrible change of perspective that is unfair to the millions of property rights owners involved.
It also effectively strips away the property right. Why? Consider what you as a copyright owner will do as Google marches on with its system. If you “withdraw” and sue (assuming that you can), what is the effect of being removed from Google’s all encompassing index system? One commentator observed that an “orphan work” removed from Google’s massive system is truly gone. Would you elect to do that for your work? Most will not. The settlement gives Google effective control over other persons' property.
There is more to say, but that is for another blog.
]]>
It is fair to say that, while the user generated content machines (such as YouTube) thrive, many commercial content producers are struggling. Clearly, the user-generated content sites document that if one gives content or the ability to distribute it away free, there will be takers and the site may become immensely profitable. But – what if a content owner desires to charge for content? Are current laws and business practices adequate to support that choice?
Should we care? Yes.
While valuable works are produced by individuals not seeking to commercialize their work, other important information projects only occur where substantial financial support or commercial benefits exist or can be realistically sought. Intellectual property law must support both and the range of options in between, but it is too often easy to ignore one of these models in policy discussions.
Some of the economic stress of the content industries is attributable to the inevitable adjustments caused by new technology and social expectations. Thus, a business model that assumes that a content producer or publisher can charge high prices for copies faces the likelihood of greater seepage (piracy) in a digital world than in a plastic world. But while many in the public are comfortable copying commercial content created by others, most do not do so without authorization. One role of law and its enforcement, of course, is to shape and influence expectations about complying with the norms set out in law. The business model may remain viable if relatively effective technology controls become acceptable in modern markets or if social expectations change to reduce or cabin in piracy at a level that is sustainable and whose costs are covered by sale or license fees generally. But the greater digital seepage means that the economics and scale have changed; this heightens, rather than reduces, the need for and the character of supportive laws for the content producer.
Unless properly adapted, modern copyright and related law may not support content industries even remotely of the scope and depth of those that historically have existed.
It may seem odd to be discussing a risk of failure in content industries at a time when the Internet created a huge upsurge in the amount of information available. But the reason lies in large part in the provenance and character of information. A difference exists between content generated and published without review by individuals, and content generated by concerted research or creative effort of financed groups of professionals. The motion picture “Wizard of Oz” may not be perfect, but the quality of that product differs from the video on YouTube by Joe Smith who filmed himself eating pizza. Many studio products, of course, are not well done, while many individual products are excellent, but one does not have to denigrate the work by Joe Smith to recognize that society is the loser if his content becomes the sole entertainment content available. If these types of content reside on a continuum, however, the problems of law sustaining the content development reside closer to the “Wizard of Oz”, while the explosion of information on the Internet lies closer to the Joe Smith video. Law should provide a basis that promises to sustain both the commercialization model and a model in which the content developer gives her product freely to the public. It is with reference to professional or commercial content that the incentive structure of traditional intellectual property law has its greatest bite, and it is here that the Internet and other digital systems have their most potentially disruptive impact.
The floundering newspaper industry is an urgent and important example of the commercial problem. Judge Posner correctly suggests that a risk of the demise of the newspaper industry creates a significant social policy issue since it would reduce the amount of professionally investigated and vetted information available to the public. It would also reduce the extent to which third parties monitor and report on public and business conduct. Of course, not all newspapers do significant research or plausible analysis, but for those that do the research and editorial staffs supported by news organizations add value to the information. They require financial support and one source is by selling content or access to it. Posner suggests that law might be expanded to cover a right to prevent appropriation of factual content from the news sources. The suggestion drew predictable skepticism and questions about how one distinguishes among newspapers and content providers such as Joe Smith. But it underscores the potential relationship between law and the sustainability of at least some information content models.
We are where we are today in terms of intellectual property policy and its impact on the content industries in part because of an on-going conflict between groups that, respectively, seek enhancement or protection of property rights, and those who would prefer a diminishing role for intellectual property (especially for copyright) in digital environments. One theme frequently cited by the latter group argues that “information wants to be free” and that to the extent that commercial incentives or profit online are desired, they should be sought in ancillary ways, such as through advertising revenue, while the information itself is made freely available. The difficulty is that this business model has largely been a failure and those who have been able to follow it tend to be content “aggregators” or providers of other services, not content creators or publishers. They survive and thrive, rightly or wrongly, by collating the work of others (often obtained for free). But in a world of limited or no profit for newspapers or other content-providers, who will be left to create that content?
As Tom Rubin commented regarding newspaper media:
Starting back in the early 1990s, some leading Internet pundits espoused the motto “information wants to be free” and implored content owners to simply give away their content and monetize it through secondary means – such as concerts and tee-shirts for musicians and, in the case of media, … by adopting a business model consisting of free and liberal distribution plus online advertising. And that’s exactly what most newspapers did…. [But the] evidence is in, and I think we can safely say that the “information wants to be free” approach not only does not work, actually it has been a disaster for almost all newspapers. Yet even today, despite being confronted with mountains of evidence of failure, some Internet leaders continue to propose the very same prescription for the future of newspapers…. But for the media at least, the verdict is in and the time has come to reject these claims once and for all.
There are, of course, many other issues regarding the impact of digital technologies on content creators and providers. But the basic issue in reference to law and legal policy involves a need to develop a balance in intellectual property law that preserves what we have and what, for the United States, is one of our few thriving industries – industries that produce and commercially distribute informational content – and that supports realistic business models for those industries. To do so requires a shoring up of property law and judicial interpretations of existing law that support, rather than a continued erosion of the position of content developers and providers. In doing this, the interests of individuals and companies who create content they do not desire to commercialize also need to be protected, but this does not support widespread, substantial copying of the work of others or relying on that work to develop a commercial enterprise profitable for the aggregator or other commercial entity.
Some anti-rights commentators criticize the notice and take down process of Section 512 as creating an environment in which posted material will be taken down even if it does not actually infringe a copyright since, from the service provider's perspective, there is an advantage to responding in that manner and coming within the safe harbor rules of Section 512. But, of course, the statutory intent was to provide an easy, efficient remedy for alleged infringement, and the Section 512 counter-notice provisions give a wrongly accused person the ability to have the material re-posted.
In addition, Section 512 requires that the notice contain a “statement that the complaining party has a good faith belief that use of the material in the manner complained of is not authorized by the copyright owner, its agent, or the law.” Section 512(f) further states: “Any person who knowingly materially misrepresents under this section … that material or activity is infringing … shall be liable for any damages … incurred by the alleged infringer … as the result of the service provider relying upon such misrepresentation in removing or disabling access to the material or activity claimed to be infringing ….”
In Online Policy Group v. Diebold, Inc., 337 F.Supp.2d 1195 (ND Cal. 2004) an anti-DMCA District Court held that a voting machine manufacturer was liable for issuing infringement notices to ISP's whose systems posted e-mails collected from within the company and criticizing the quality of its voting machine products. The court concluded that, in light of the public interest in information regarding voting machine quality, no reasonable copyright owner could have concluded that copying of these e-mails was anything other than fair use. In reaching that conclusion, it suggested that: “A party is liable if it "knowingly" … misrepresents that copyright infringement has occurred. "Knowingly" means that a party actually knew, should have known if it acted with reasonable care or diligence, or would have had no substantial doubt had it been acting in good faith, that it was making misrepresentations.”
This standard wrongly imports an objective “reason to know” that is not implicit in ordinary usage of the word “knowledge.”
Shortly after Diebold, the Ninth Circuit considered a similar issue and announced the better standard. In Rossi v. Motion Picture Ass'n of Am., 391 F.3d 1000 (9th Cir. 2004), the Motion Picture Association (MPAA) was sued for tortuous interference with a contractual relationship when it issued a notification to the ISP that held the plaintiff's Website. Without having done any investigation beyond viewing the site, MPAA claimed that the site made infringing copies of motion pictures available to users and notified the ISP of this. The ISP shut down the site. The issue was whether the notice was privileged under DMCA. The court held that it was, even if MPAA acted unreasonably in not investigating beyond a mere view of the site.
The court held that the proper standard is the presence of a subjective belief in the accuracy of the notice (subjective good faith). There is no requirement of reasonable care or investigation -- that is, a reason to know standard does not apply. It held that: “A copyright owner cannot be liable simply because an unknowing mistake is made, even if the copyright owner acted unreasonably in making the mistake. Rather, there must be a demonstration of some actual knowledge of misrepresentation on the part of the copyright owner…. [A] lesser "objective reasonableness" standard would be inconsistent with Congress's apparent intent …”
Unfortunately, the Rossi standard may not insulate all notices from a claim of bad faith. In Lenz v. Universal Music Corp., 572 F.Supp.2d 1150 (ND Cal. 2008) the District Court held that the notice-giver must consider probable claims of fair use and that a failure to do so may constitute bad faith in clear cases. It commented: “Universal suggests that copyright owners may lose the ability to respond rapidly to potential infringements if they are required to evaluate fair use prior to issuing takedown notices [and] that the question of whether a particular use … constitutes fair use is … difficult for copyright owners to predict … However, while these concerns are understandable, their actual impact likely is overstated. Although there may be cases in which such considerations will arise, there are likely to be few in which a copyright owner's determination that a particular use is not fair use will meet the requisite standard of subjective bad faith …”
So the standard may still be in flux, but the balance seems to be being drawn in favor of allowing honest copyright owners to use the take-down procedures to protect their works without risking liability. This is an important approach to following in making the notice, take-down, counter-notice system work.
]]>
Although written by two respected academics, the Principles are fundamentally flawed in part because they were drafted with virtually no input from commercial software producers and drafted under the influence of a group dominated by people and companies who failed to achieve their goals in UCITA or in drafting revisions of Article 2. They thus shifted over to a friendly forum they could dominate. I will have more to say about this, but let’s start with an illustration – a misguided, non-disclaimable warranty that no “hidden” “defects” exist in software.
]]> “Principles” § 3.05 suggests that: “A transferor that receives money or a right to payment of a monetary obligation in exchange for the software warrants to any party in the normal chain of distribution that the software contains no material hidden defects of which the transferor was aware at the time of the transfer. This warranty may not be excluded.”
I put aside the facts that the form of this language (and followed throughout the “Principles”) reads as if it were a statute even though it was not vetted through any legislative process and will never be. The fundamental flaw is that this alleged warranty is not supported in any body of contract law. Any body of contract law! It is not present in the Restatement of Contracts. It does not exist in UCITA. It does not exist in UCC Article 2 or Article 2A.
The Reporters’ Notes cite several cases allegedly in support of this purported, non-disclaimable warranty. But one case, cited twice, granted summary judgment against liability for a non-disclosed defect (because of a damages limitation clause). And all of the cited cases that supported liability did so under the law of fraud, not contract law. Further, there are many cases in the law of fraud that hold that there is no duty to disclose in arm’s length deals. In fact, at least two of the cited cases did not involve an implied obligation, but liability for express lies.
There is no support in law for the purported warranty under contract law. The law of fraud contains numerous limits on liability that are not present under a non-disclaimable warranty. Indeed, since fraud liability requires reliance, it can be disclaimed by statements that disavow any assurances in reference to the allegedly hidden problem. Many cases in the area of fraud law enforce specific disclaimers as defeating a claim of fraud. The “Principles” reverse that rule as a matter of contract law. If I say to you: “I am not giving any assurances about there being no defects in this software”, why should that statement not be effective when it clearly would be under the law of fraud?
So, why, you might ask, is this “principle” a bad rule? One answer is simply that, if the “principle” were followed, the software industry would be subject to a rule that does not apply to any other industry. Why discriminate against one of our few burgeoning industries?
But there are several other points:
First: The scope of the obligation is obscure. What does the word “hidden” mean? Is a “defect” “hidden” unless it is affirmatively disclosed? Or, is it hidden only if the producer or other licensor actively conceals it? Is it hidden if the defect, once discovered, is discussed on the Internet, but the particular licensee is not aware of that discussion? Does the idea of a “hidden” defect require intent to deceive such as is ordinarily required under the law of fraud?
Second: The coverage of the “principle” is obscure. What does the word “material” mean? It is not defined. Remember that the “principle” is not limited to the immediate person to whom the software is transferred, but to all others in the “normal” chain of distribution. If a defect is not material to the person with whom I deal, why am I still liable to a later person for whom it is material?
By the way, whatever happened to the idea of contractual privity?
By the second way, what constitutes a “defect”? There is a body of research that suggests that it is impossible to create software with no flaws. Is any flaw a “defect” that creates liability if it is not disclosed and it happens to have a material impact on some party down the distribution chain? If so, the “principle” extends the notion of contract liability beyond reason. The car that I drive has a plastic cover over the track on which the driver’s seat rests. The “flaw” is that the plastic breaks if you step in the wrong spot when entering the car. Is that a “defect” that is a breach of contract if it was “hidden”, e.g., not affirmatively disclosed? Is that the type of breach that should be non-disclaimable?
Third: The idea that the supposed warranty is non-disclaimable and runs to all persons in a distribution chain creates a veritable circus of potential liability and resulting costs. The people who will pay are the purchasers subjected to higher prices for ordinary software. The beneficiaries are ---------?
Fourth: The concept, such as it is, of this “principle” is to require disclosure of what is often proprietary (trade secret) information. In a risk adverse company, how does one decide what “flaw” is a “defect” that may be “material” to unknown users?
Enough said: if it were adopted, this is a bad rule applied on a discriminatory basis to one of our few vibrant industries. It should be rejected or, simply, ignored.
Oh….. One last thing. The “principle” is limited to entities that charge “money” for the software. This is intended to exclude “open source” participants and providers, including FSF. So, in “principle”, they do not need to disclose, but Adobe does. I sure hope the discrimination in this rule is not based on the idea that all users can read software code and find and cure defects. I certainly cannot do so. Recall what I said above: the Principles were drafted in a political context dominated by people who had lost their pet positions in UCITA and Article 2, but now dominated a friendly forum.
]]>
The fundamental policy issues require that we ask how much law should be given over to protect non-confidential, personal information and whether that law should be in a form of liability suits or non-litigation guidelines. Even if protection of non-confidential personal information is vital, laws grounded in rules not susceptible to high cost litigation and damage claims can better establish social expectations without causing a massive shift of value, largely to plaintiff’s lawyers.
There are two liability issues.
The first is whether the holder of the data owes a duty to the person about whom the data relate in the absence of an express assumption of such duty. This “duty” issue can arise in tort or under implied warranty rules in contract law.
Either way, no implied obligation should exist. Most courts so hold. The traditional rule is that a person who properly obtains non-confidential data has a right to use it. The fact that I know your home address does not create a duty to keep that information secure. Indeed, such information is known by many people.
Some information is delivered under confidentiality restraints or is sufficiently sensitive that an implied duty can be inferred. But the presumption should be that data is free from legal constraints unless there are over-riding reasons to restrict its use, or impose liability for its disclosure. No general obligation of maintaining security should exist. If it were created, we would face an unwarranted restriction on ordinary discourse and information sharing, socially and commercially. While there are some benefits in reference to a sense of data security, these benefits do not over-ride the benefits of being able to use and deploy the information one knows without fear of a lawsuit.
The second issue is the “damages” issue.
Even if a duty were created, no cause of action should exist if there are no proven, foreseeable damages cognizable under the particular cause of action chosen. The mere compromise of a database involving personally identifiable information does not necessarily lead to legally cognizable damages in the absence of a foreseeable and provable connection to actual harm to the data subject.
The damages most frequently asserted in security breach settings entails the risk that a wrong-doer may use the data for identity theft. But, while there have been numerous security breaches of identity theft incidents associated with those breaches has been very low. Thus, the litigation issue has been that, even if no identity theft occurred, is the distress and preventive actions caused by the risk of identity theft compensable. Most courts correctly hold that they are not.
In Forbes v. Wells Fargo Bank,420 F. Supp. 2d 1018 (DMN), the court held that a bank was entitled to summary judgment on claims of negligence and breach of contract because the plaintiffs had no damages. There were no unauthorized transactions and plaintiffs could not recover damages for a risk of harm unless that risk resulted from a present injury, that is, “the threat of future harm, not yet realized, will not satisfy the damage requirement.”
Similarly, in Pisciotta v. Old Nat. Bancorp, 2007 WL 2389770 (7th Cir. 2007), the court was asked to decide whether Indiana law would allow individuals receiving notice of a security incident to recover their costs for credit monitoring or emotional distress. The Seventh Circuit said no. An Indiana statute imposed an obligation to provide notice in the event of a security breach but not liability:
Had the Indiana legislature intended that a cause of action should be available against a database owner for failing to protect adequately personal information, we believe that it would have made some more definite statement of that intent. … The narrowness of the defined duties imposed, combined with state enforced penalties as the exclusive remedy, strongly suggest that Indiana law would not recognize the costs of credit monitoring that the plaintiffs seek to recover in this case as compensable damages.
The Seventh Circuit explained that plaintiffs had “not suffered a harm that the law is prepared to remedy. …”
Although, personal data security has become a burgeoning field in law, courts properly have shown a reluctance to impose an implied obligation to maintain the security of data of a non-confidential kind, regarding another party. A person rightfully in possession of such information has a right to use and disclose it – rights co-equal to the data subject. There is no actionable legal obligation to the other person, except for confidential or highly dangerous information.
]]>A recurring issue today in court is who control or has an advantage in the newly emerging digital and elated markets. Are technologists like Google in control with a right to use any content any time without permission, or do we still value the content creators? This is not a philosophical issue and the combatants are not professors or mavericks. They are large companies with billions of dollars at stake. This is the era of the information wars.
In the Cartoons case, Cablevision implemented a remote DVR system, allowing customers to select, store, and later play cable broadcasts. In the system, when cable programs are received by Cablevision, they are routed through a buffer (buffer 1) and copied briefly while software checks if any customer had requested copying of the program for later replay. If there was a request, the program would be copied into a server (buffer 2) and held for later viewing by the customer. When the customer later desired to watch the program, the DVR system delivered a performance to the customer’s home video. Variations of this system are widespread in the cable market and are a robust competitor to home recording systems.
Cablevision did not ask for licenses from the content (program) providers for copying their programs or publicly performing them at times other than the original transmission. It simply implemented the system and charged customers who desired to use it.
The Second Circuit concluded that this was fine – a major corporation (a cable company) could reuse another company’s copyrighted product without permission or payment. This was a complicated decision, but fundamentally, a choice by the panel to turn potentially billions of dollars away from the creative parts of the industry to those who copy and retransmit. It was a wrong decision.
First, the panel held that copying was not copying if the copied image lasted for only a second or two. So, the entire line of cases started in the Ninth Circuit’s MAI decision remains intact, but now seems to become a question of ….. I do not know what.
Second, the full content was copied into buffer 2 by Cablevision’s system. But, no, cablevision who charged for this service, did not make the copies – the customers did – at least if you believe the panel of he court!
Strike 2.
Strike 3 is when the court held that, when the customer who caused the copy to be made, eventually asked for it to be performed in the customer’s home – this was not a “public” performance.
This may be among the worst appellate court decisions in copyright law history
]]>The Supreme Court in Acuff-Rose described a “transformative use” as follows:
The enquiry here may be guided by the examples given in the preamble to § 107, looking to whether the use is for criticism, or comment, or news reporting, and the like. The central purpose … is to see … whether the new work merely “supersede[s] the … original creation, or instead adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message; it asks, in other words, whether and to what extent the new work is “transformative.”
“Criticism, comment, news reporting, parody, and the like” – using parts of the original and adding something new that makes for a new work as compared to merely superseding (using) the original work; these are the hallmarks of a transformative work. A transformative use in this form is not per se legal, but it weighs in favor of fair use. This is because the copying of the original author’s expression contributes to a new work and achieving the “goal of copyright, to promote science and the arts…”
“Transformative copying” (use) may involve making a derivative work or a partial copy of the original in a new work. This may be held to be non-infringing for reasons that lie in a goal of permitting creative adaptation, criticism and the like. The reasons are not grounded in the notion that a third party can supersede a commercial market simply because the copyright owner has not yet and may never decide to enter.
Assume that I write a book for lawyers on the topic of contract law. A third party copies the book and sells copies for use in teaching her high school class. Is that a “transformative use”? Of course not. Even if I never plan to enter the high school market, the use of the book in that market by the high school teacher supersedes my work in that market. It appropriates a commercial market for the free-riding teacher – the high school class market. Most copyright owners would grant a fee-based license even if we had never thought about the value of that market in the first place. But a finding of fair use makes the work free for the taking. Even if the owner refused to grant a license, a decision to withhold a work from a market is just as fully within the scope of a copyright as is a decision to enter it.
Is a motion picture derived from a best-selling novel a transformative fair use because the book author does not intend to sell motion picture rights? Again, of course not.
But compare this to a few recent decisions.
In Perfect 10 v. Amazon.com, 487 F.3d 701, 82 U.S.P.Q.2d 1609 (9th Cir. 2007), the Ninth Circuit held that comprehensive copying of images from the internet into thumbnail format and using them in an Internet visual search engine was “highly” transformative fair use. Google did not add to the images, did not critique them and created no new work, but merely reduced their size and resolution and used them for its own purposes. Perfect 10 in fact had begun marketing thumbnails for cell phones. Yet, the court concluded Google’s unauthorized copying did not affect the market for the images. But of course it did. A market existed for use in search engines. The court’s ruling closes that market – why license and pay for what you can get for free?
Consider also the AV Paradigms case (2008 WL 728389 (ED Va. 2008)). There, students were required to submit their school papers to the iTurnitin service, which checked for plagiarism. The students did so, but objected to iTurnitin copying the work into its database. In lengthy dicta, the court held that this copying (of the entire work) was a “transformative use.” But what was the transformation? It resided in how the defendant used the copy it made – as part of a searchable database, rather than as a school paper.
These two courts mistakenly view transformative use as merely using a work in a way different from that which the copyright owner currently does. The true meaning is that transformative use is a use that transforms portions of the original into an entirely new work that does not simply supersede the original in a market or use to which the original may not yet have been applied. Consider again what the Supreme Court said about why a parody using small parts of the original was a transformative commercial use:
Indeed, as to parody pure and simple, it is more likely that the new work will not affect the market for the original … by acting as a substitute for it … The market for potential derivative uses includes only those that creators of original works would in general develop or license others to develop. Yet the unlikelihood that creators of imaginative works will license critical reviews or lampoons of their own productions removes such uses from the very notion of a potential licensing market. “People ask ... for criticism, but they only want praise.”
]]>
FOSS licensors applaud the Federal Circuit decision in Jacobson, but the real winners are the vastly more numerous software producers who reject the “free” software model. Why………?
]]>The reason is simple.
The decision expressly recognizes that downloaded software or other copyrighted content can come with conditions on its use, apparently without requiring that those conditions be part of a contractual arrangement. That result seems correct, and it supports not only FOSS software, but more importantly the wide-ranging commercial software regime under which rights-owners conditionally distribute their software, limiting the permissions and conditions under which use is permitted. “Free” and “proprietary” software distributors engage in the same process. They are distribution models, differing only in the details of the terms.
So, let’s look at the issue.
The FOSS argument is that licenses are (at least in some cases) non-contractual permissions – a rights-owner gives the other party permission to use the owner’s rights, but can restrict that permission in whatever way the rights owner chooses. This does not require agreement.
There are obvious limits on this concept.
One is the idea of “fair use” in copyright law. If a transaction is non-contractual, fair use is a defense to the copyright infringement claim. If the transaction were contractual, however, a contract claim is not affected.
Another is “first sale” or “exhaustion” doctrine under patent or copyright law. Non-contractual notices do not prevent exhaustion (or first sale) rights from transferring to the purchaser under intellectual property law. Quanta Computer, Inc. v. LG Electronics, Inc., 2008 WL 2329719, 86 USPQ2d 1673 (US 2008).
Neither of these issues were raised in Jacobsen v. Katzer, 2008 WL 3395772 (Fed. Cir. 2008). The defendant downloaded a copy of the “free” software and used it in its own products, but did not comply with the provisions in the free “Artistic license.” It was sued for copyright infringement. Without addressing whether there was a contract or whether the downloader agreed to terms, the Court of Appeals focused on whether the provisions the defendant failed to comply with were promises or “conditions” on the permission granted under the license. That is, was permission to copy and distribute the software conditioned on compliance with the full terms of the license, or was that permission granted in return for a promise by the licensee to comply with the provisions of the license. The terms provided, in relevant part, that the licensee had the right to copy, modify, and distribute the software:
provided that [the user] insert a prominent notice in each changed file stating how and when [the user] changed that file, and provided that [the user] do at least ONE of the following: [stating several steps required with respect to the copies]
The court focused on “provided that”, concluding that, under California contract law, this is treated as a language of condition. As a result, when the licensee did not comply with these terms, it acted outside the scope of its license and, thus, could be sued for copyright infringement.
The court’s opinion did not explicitly address whether this license was a contract or not, but did engage in the following analysis, which implies that it was looking to establish that a contract did exist and that the terms were important even though no money exchanged hands. Whether they talk about being free, many FOSS providers receive commercial benefits from their licenses.
So the net result is a court of appeals ignoring whether there was assent to terms or an opportunity to review them and focusing on the fact that “Copyright holders … have the right to control the modification and distribution of copyrighted material.” If I were a commercial (or an open source software provider), I would make certain that my licenses conformed the terms of condition recognized by this court.
]]>Must the proof be that the defendant actually pushed the copy into the hands of third parties? The answer must be no.
Some argue that to distribute a copy the actor must initiate a direct exchange giving the copy to a another person. But taking all the steps to place a copy in a known distribution system from which others take copies surely suffices. Courts so hold. One court, in a criminal law case, analogized placing a copy in a peer-to-peer system to operating a self-service gas station:
[Defendant] may not have actively pushed pornography on Kazaa users, but he freely allowed them access to his … stash of images and videos and openly invited them to take, or download, those items. It is something akin to the owner of a self-serve gas station. The owner may not be present at the station, and there may be no attendant present at all. And neither the owner nor his or her agents may ever pump gas. But the owner has a roadside sign letting all passersby know that … they can stop and fill their cars for themselves, paying at the pump by credit card. Just because the operation is selfserve, or … we do not doubt for a moment that the gas station owner is in the business of ‘‘distributing’’… gasoline”
But must the complaining party show that particular people accepted the invitation and obtained copies? No. The wrongful act is not delimited by a particular transaction, but by distribution in general.
But must the complaining party show that at least someone accepted the invitation? Here we must distinguish between questions about what is sufficient proof and what is the legal standard being proven. Proof that a copy entered an active downloading environment may well create an inference that the invitation to “take” a copy was accepted by someone.
But can distribution occur even if no copies exchange hands? Some courts, equating this possibility with a “make available” right under copyright law have said no. But the answer in the statute is yes, at least in some circumstances.
Copyright law does not define “distribution”, but does define one type of distribution – called “publication” – which includes an offer to distribute. Here is the definition:
“Publication” is the distribution of copies or phonorecords of a work to the public by sale or other transfer of ownership, or by rental, lease, or lending. The offering to distribute copies or phonorecords to a group of persons for purposes of further distribution, public performance, or public display, constitutes publication.
The Electra case correctly held that this definition provides a measure of what the term “distribute” means in copyright law - an offer that constitutes “publication” is a distribution under copyright law.
But this leaves the final question of whether other offers qualify as a distribution. Electra held that they did not and that there must be an allegation that the offer be for purposes of further distribution etc. But nothing in the statute suggests that this language restricts the meaning of distribution. Instead, the definition illustrates of one type of distribution and indicates that distribution can occur by making an offer, rather than only by actually delivering copies (e.g., an offer to distribute can be a distribution for purposes of copyright law). There are those who disagree, but it seems to me that this is the clear meaning of the statute.
But what of the “make available” right?
The issue is open. This concept comes from language in an international treaty. The scope of this proposed right seems similar to that of an offer to distribute, but the latter, grounded in the statute, suggests an active role by the defendant intending actual use by others. Clearly, however, placing a copy in an active downloading and transfer system online suffices to establish an offer to transfer a copy or allow it to be transferred, and an act making a copy available to others.
]]>Quanta involved an arrangement in which LG licensed Intel to sell products that largely encompassed LS method and other patents. A “Master Agreement” required Intel to give notice to buyers that this sale did not grant a license to use the items with products of third party manufacturers. The issue was whether the method patent claims embodied in this license agreement were exhausted by Intel’s sale of product to third parties. The Court held that exhaustion occurred.
There were two primary issues. First, does the doctrine of exhaustion apply to “method patents.” The Court said yes, if the product was sold in an authorized sale. That is a correct ruling. To hold otherwise, would create artificial barriers and distinction. According to the court, the doctrine now applies whenever the item sold embodies a sufficient amount of the claims method patent. On what is sufficient, the Court reverted to earlier authority commenting that exhaustion is triggered by a sale if the sold items’ only reasonable and intended use is to practice the patent and they “embodie[d] essential features of [the] patented invention.” This was met in Quanta.
That was the easy issue.
The more at risk issue was whether the sales by Intel pursuant to its license triggered the exhaustion doctrine. The Court could have gone off the deep end here and eviscerated distribution methods used in software and many other industries. But it did not. Instead, the Court emphasized that exhaustion (first sale) occurs only if there was an authorized and unconditional sale of a product. This is the right rule – first sale is not a right of the purchaser or an absolute mandate in law, but simply a consequence of a marketing decision by the rights owner. It arises only if the rights owner authorizes unconditional sales of copies.
Having gotten that part right, the Court then held that the Intel-LG arrangement authorized unconditional sales. The key to this conclusion was that the license to Intel authorized sales and that the Master Agreement merely required Intel to give notice to buyers that the sale did not create an implied license to use the product with non-licensed other products. This did not condition the authority to sell or the terms of the sale. Bad drafting of the arrangement perhaps.
The sales of product were unconditional. But what of the notice? It merely blocked any implied license argument, but did not become a limiting part of the contract with the buyer, or a condition on Intel’s right to make an authorized sale. But, the buyer’s defense did not lie in an implied license. It was based on exhaustion of the patent by an unconditional sale.
The distinction here lies in the nature of the sale authorized by the license. If the transaction between Intel and LG had restricted Intel’s right to sell products to cases in which the sale was conditional on the buyer agreeing to not use the product in a third party configuration, then exhaustion would not apply. Here, the agreements, as viewed by the Court, authorized unconditional sales and merely required a notice relating to the absence of any implied license.
So, what is the problem? Rambus is an innovator. It held valid patents on important technology essential to use the standard promulgated by JEDEC. JEDEC promulgated the particular standard because it believed that the standard reflected an optimal technology for the particular application. This enhanced the market value of the Rambus technology. Rambus deserved to be compensated for use of its property by other companies.
The problem, according to the FTC, was that Rambus' failure to disclose enabled it (1) to acquire a monopoly through standardization of its technology (rather than alternatives), or (2) to avoid limits on its licensing fees that the SSO might have imposed had it known of the proprietary rights. Notice the word “or”, the court did.
But the first of these was not proven and the second is not an antitrust violation.
Antitrust law only precludes actions that adversely affect the competitive process – not conduct that might harm particular competitors or increase the value of particular products or technologies. If there had been actual fraud or breach of contract, then that is how the claims should have been brought; but these are not claims over which FTC typically has jurisdiction.
So, we had FTC believing it important to require participants in an SSO to disclose and continue to disclose proprietary rights developments that might relate to a standard that an SSO might adopt. Certainly, an SSO could adopt such a rule by contract or as part of its operating rules. The FTC was willing to use antitrust law to transform this into mandatory law. Indeed, one court later did so relying on the now-abrogated FTC decision. The Rambus court, however, properly slapped the agency for over-reaching. Non-disclosure is not an antitrust issue unless there is an anti-competitive impact. The fact that it may harm some competitors or increase the price of a product does not qualify.
The Rambus decision is part of a broader debate about IP in technical standards setting. One could say that the FTC believed (as a matter of antitrust law) that the existence of IP rights or potential IP places a burden on the owner to share or to avoid having those rights capture the standard. The appellate court rejected that. Clearly, the mere fact that IP held by one company relates to a proposed standard is not a reason to penalize the owner or for the SSO to promulgate standards that avoid the IP but are less than optimal otherwise. As I have written elsewhere, an SSO should not be used to replace competition, even if that competition will be between proprietary and non-proprietary technology.
Certainly, the potential presence of IP rights is relevant to an SSO process and its legality, but not always in the way some anti-rights adherents believe. A rights owner might manipulate the process to its advantage and use the market power or monopoly it creates (if it does) to wrongfully exclude competition. But that package of wrongful action, when it occurs, goes well beyond simply getting a standard that requires the IP for its use. That is essentially what the Rambus court held.
On the other hand, SSO’s involve group action, often controlled directly or indirectly by competitive companies or their employees, and permitted under antitrust law only so long as anti-competitive effects are not generated by the group. Action by a group to exclude another competitor because it hold relevant IP rights may be an antitrust violation.
So the issues cut in both directions.
Internationally, SSO’s take many different approaches to the relationship between IP rights of participants and standards development – typically involving either a limited disclosure obligation or an obligation to license the right on reasonable and non-discriminatory bases (
Rambus is part of a debate about what some call “open standards.” This term has numerous different meanings, but fundamentally, some argue that it signifies standards not “infected” by proprietary rights or as to which such rights are waived. This is just another manifestation of the rights – anti-rights debate that pervades most of modern IP law. The better view is that a standard is “open” if it is developed in an open process without discrimination against any technology, including any technology covered by proprietary rights.