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.

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 (7^th 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”    

U.S. v. Shaffer, 472 F3d 1219, 1224 (10^th Cir. 2007).

            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 (RAND terms).  But why would an absolute and continuing disclosure rule be a problem in an SSO?  If we think about one person with one patent and no others, disclosure as a legal norm might not be harmful.  But consider large companies with development work going on continuously in numerous venues continuously.  The cost of monitoring would be immense.  The liability risk (especially under an antitrust regime) would be ridiculous.  So, the response would be to not participate in at least some SSO projects.  That would deny the SSO the involvement of many of the best resources potentially available for standards development.

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.

            The lower court in C.B.C. Distribution and Marketing, Inc. v. Major League Baseball Advanced Media, L.P., 2007 WL 2990366 (8^th Cir. 2007) got the analysis terribly wrong, but luckily was abrogated by the Court of Appeals.  The lower court argued that contract restraints on use of non-secret data without underlying IP rights violated fundamental public policy.  What policy?  Why would that policy over-ride contract law?  The opinion gave little insight into this. 

The case involved a license from Major League Baseball to C.B.C. of the right to use player names and statistics in interactive media fantasy baseball systems.  The contract broke down, but CBC continued to use the data in violation of the license which specifically prohibited such use.  The District Court held that enforcement of this clause was precluded because a “strong federal policy favoring the full and free use of ideas in the public domain … prevails over [those] contractual provisions.”  The District Court was apparently referring to a general First Amendment right, but that right can be and frequently is contractually waived.  It also referred to the ancient Supreme Court decision in Lear v. Adkins, but that case dealt with patent law and has not been expansively applied by courts since originally announced during an era of judicial hostility to intellectual property rights.

The appellate court in a 2-1 decision avoided the lower court’s error by reading the license to create a warranty of title in a way that will give drafters of such licenses pause.  The “warranty” was breached, thus precluding enforcement of the non-use clause because the subject matter of the license was not covered by intellectual property rights.  The warranty language? “[The] Players Association “is the sole and exclusive holder of all right, title and interest” in and to the names and playing statistics of … major league baseball players.” But the interpretation of all right and title is, at best uncertain in a setting where both parties must have understood that the data (as compared to the names) involved was not covered by intellectual property rights.   

            The dissent came closer to getting it right.  It concluded that there was no warranty.  Instead, the contract said “that the parties “agree” that MLPBA is the sole and exclusive holder of all right, title and interest in and to the Rights. CBC surely can “agree,” as a matter of good business judgment, to bargain away any uncertain First Amendment rights that it may have in exchange for the certainty of what it considers to be an advantageous contractual arrangement.”  As to the public policy issue, the dissent observed that there was no federal statute that addresses state-law contract obligations on this issue and, thus, “no indication that Congress sought to abrogate contracts in this area that are otherwise enforceable under state law.”

            So, the lower court’s wrong analysis is gone as a matter of precedent, but we have a precedent that reads a common contract clause in a way that makes severe trouble for any license drafting regarding data. 

            The far better result would have been to have both courts recognize that nothing is strange about contractually licensing information that is not within the scope of intellectual property law.  That is and has long been the law.  See Ross, Brovins & Oehmke, P.C. v. Lexis Nexis Group, a Div. of Reed Elsevier Group, PLC, 463 F.3d 478, 80 U.S.P.Q.2d 1518 (6th Cir. 2006).

But what is the appropriate remedy if the court holds that a permanent injunction is not warranted, but that infringement has occurred and may occur in the future? 

One response is simply to assess damages as to past infringement, leaving any future use of the patent for a voluntary agreement of the parties (a license) or a subsequent infringement suit for the subsequent infringements.  That is clearly the preferable option, although it does raise limited issues of judicial economy.

A second alternative is to permit subsequent use by the defendant subject to the payment of a reasonable royalty imposed by the court.  This is a form of compulsory licensing that rewards the wrongdoer, unless the remedy has been requested by the patent owner.  Nevertheless, a panel of the Federal Circuit indicated that such a remedy may be appropriate.  One wonders why.

            In Paice LLC v. Toyota Motor Corp., 2007 WL 3024994 (Fed. Cir. 2007), after a finding of infringement and damages of $25 per use of infringing drive trains for hybrid motors, the District Court denied the patent holder’s request for a permanent injunction, and instead sua sponte issued an order permitting Toyota to continue to produce the infringing items subject to payment of a $25 per motor royalty.  The refusal to grant an injunction was not appealed, but the Federal Circuit noted that the court’s decision was based on at least the following factors:

·         In context, the absence of an injunction would not adversely affect the patent holder’s ability to license the patent to others. 

·        The patent holder did not actually manufacture any goods, and there was no threat that it would lose name recognition or market share without an injunction.

·        Monetary damages were adequate as suggested by the fact that the patent owner had offered to license Toyota during the post-trial period.

·        The balance of hardships favored Toyota since issuing an injunction would disrupt not only its business, but that of suppliers, dealers and others.

This gives you some idea of the type of issues likely to adversely influence a patent owner’s ability to obtain injunctive relief.  They depict a plaintiff company whose only interest in control of the patent is to obtain licenses or judgments of infringement, rather than to build or protect an existing or potential market.

            Should a denial of permanent injunctive relief automatically require judicially forced licensing?  The answer is no.

The Federal Circuit suggested that courts have long had the ability to order ongoing royalties as a remedy in appropriate cases, but carefully said that the District Court may do so, not that this result is ever required.  Even then, the cases it cited – an antitrust remedies case and an infringement case in which the ongoing royalty was not contested – do not support the conclusion that broad judicial power exists to make an award that effectively takes away the winning party’s property without being requested by the prevailing party.  Indeed, the Patent Act talks about granting injunctions to prevent violation of the property rights, while this remedy permits use of the invention.

But even if a court has power to make such an award, what is an appropriate case?  The one “appropriate” case the Federal Circuit mentioned was where, after the judgment of infringement, the parties attempt but fail to agree on a license for future use.  The court commented that that this may be a case for granting ongoing royalties as a remedy, not that this was required.  The court commented:

But, awarding an ongoing royalty where “necessary” to effectuate a remedy … does not justify … such relief … whenever a permanent injunction is not imposed. In most cases, where … a permanent injunction is not warranted, the district court may wish to allow the parties to negotiate a license amongst themselves … before imposing an ongoing royalty. Should the parties fail to come to an agreement, the district court could step in to assess a reasonable royalty in light of the ongoing infringement.

But why is a failed bargain as to future conduct grounds for the court to step in and impose a bargain of its own forcing the injured party to allow continued use of its property, even though the owner did not request that remedy?  The court did not explain this strange reversal or comment on whether a rights owner who does not want the court to control its future income would be better off not negotiating an attempted license at all.

            In Paice, the court remanded because the lower court’s order contained nothing to show why the selected royalty was appropriate.  But there is a more fundamental issue.  Judge Radar, writing a concurrence, directly approached that issue.  To him, giving the parties an opportunity to negotiate a license should be a precondition to the court having any power to impose an ongoing royalty.  He commented:

[This] court should require the district court to remand this issue to the parties, or to obtain the permission of both parties before setting the ongoing royalty rate itself. District courts have considerable discretion in crafting equitable remedies, and in a limited number of cases, as here, imposition of an ongoing royalty may be appropriate. Nonetheless, calling a compulsory license an “ongoing royalty” does not make it any less a compulsory license.

The preconditions should be both an opportunity to negotiate a license and, failing a bargain, a request by both parties for the court to impose a royalty as part of the remedy for infringement.  A patent creates a right to exclude and, where the patent owner prefers to exercise that right, it should not be forced into a licensing arrangement resulting from a case in which it prevailed on the infringement claim.  There may be some cases in which vital public policy interests justify this result, but those cannot be grounded simply in the fact that the court denied a permanent injunction or the parties have not agreed to license terms.  A remedy should not penalize the person to whom the remedy is awarded.

For example, if I examine four hundred technology stocks and publish my estimates in numerical form of the price the stocks will be one month from today, can you copy those estimates and publish them in competition with me?  Contracts solve some issues.  But we are talking about copyright. In copyright, the issue turns on whether the numbers are protected expression and, if so, whether the manner of the defendant’s taking of them constitutes infringement or (for example, fair use).  They are not protected expression if they constitute facts or if the number is merged with the idea it expresses.

So, what is a fact and what is expression? 

This question has tantalized generations of copyright law scholars.  But in our context, the answer seems simple.  Information in “print” cannot be a “fact” unless it describes or purports to describe something that exists in the real world.  “Facts” and “fiction” (or estimates) are opposites.

The Second Circuit has held that predictions of market value for automobiles were expression when based on the estimators’ judgment and not summarizing actual sale prices.

This is the right approach.  A numerical database is like to a textual work containing various expressive words and phrases.  This leaves ordinary doctrines such as “fair use” or “substantial similarity” the task of determining whether a specific appropriation of the numerical text is infringement.  It thus protects against substantial copying for competitive purposes while not precluding references to any aspect of the numerical text.

But, in New York Mercantile Exchange v. Intercontinental Exchange, 497 F.3d 109 (2nd Cir. 2007), the Second Circuit muddied the waters in what was admittedly unnecessary dicta.   The case involved estimated “closing” prices for commodities future contracts on the Exchange.  The estimates did not rely on the price of the last recorded trade.  Indeed, for many futures contracts on many days, there were no trades.  Yet, the court suggested that it was a close question of whether the estimates were “facts.”  This was supposedly true for estimates on days where many trades had occurred and, thus, the estimated price was grounded in substantial data.  But of course that is nonsense.  Regardless of how much data were used, an estimated price is just that – an estimate, not a fact.   The court also suggested that on heavily traded days for particular contracts, the might be only one “correct” estimate.  But, again, that is nonsense.

At worse, New York Mercantile undermines prior law.  At best, its dicta suggest a line difficult if not impossible to draw in practice. 

The better rule is that expression of estimates in numerical form are protected expression and that whether infringement occurred when some or all estimates are copied hinges on issues of fair use, which in this case were unlikely to protect the defendant.

After rambling about facts and expression, the court decided the case on a different basis, holding that the numerical expression merged with the idea being expressed; thus, it could not be protected under copyright law.  If expression is so “merged” with the idea that protecting the expression effectively precludes others from expressing the same idea, then the copyright statute precludes protection of the expression.

But what was the “idea” expressed?  The court never defined it.  Let me try.

·        Perhaps the idea is that: “an estimated closing price can be stated in numerical terms.”   

·        Alternatively, the idea is: “providing guidance for persons who will engage in commercial transactions regarding futures contracts.”

If either of these or many other “ideas” apply, they can be expressed in thousands of ways in numerical form.  The expression does not merge with the idea.

But the court revealed its true purpose – to deny protection of the numerical estimates.  Why?  Because the court did not think that protection was appropriate since the database producer would continue to produce the data regardless of copyright protection.

Bad law for a bad purpose.  Copyright law seeks to create incentives for creative authors, but it does so in general terms.  The analysis is any particular case is not whether the particular author will continue to produce its works without copyright, but whether denying copyright to an entire set of works will reduce the incentive and the production generally for works of the type.

            The LG appeal challenges that concept as it applies to patents. 

LG licensed Intel to use a portfolio of patents and to sell products covered by several of the LG patents.  The license provided that Intel sales did not cover buyers’ use of the product in combinations covered by other LG patents.  Intel was required to notify buyers from it that the sale to them did not give them a license to use the products in combinations that controlled by LG patents.  Those notices were part of the Intel-buyer contracts.

So it seems simple.

            Property owner gave a limited license to Intel.  Intel produced products under that limited license and conveyed them to third parties pursuant to the license and under a contract limiting the scope of the sale transaction.  Transferees went outside that scope.  While transferees are shielded from some infringement claims by exhaustion doctrine, that doctrine applies to authorized unconditional sales; these transactions were not that because there were contractually effective restrictions on the rights transferred.

            The Supreme Court has accepted cert on the following issue:

Whether the Federal Circuit erred by holding … that respondent's patent rights were not exhausted by its license agreement with Intel Corporation, and Intel's subsequent sale of product under the license to petitioners. 

The license here did not permit sales by Intel that would enable the purchaser to use the product in combinations that would infringe the licensor’s other patents.

            Of course such transactions did not “exhaust” the patent.  They were a contractual method of commercializing patents under license agreements that commercially apportioned rights or permissions.   If I rent my condo in Taos for a specified, limited use, have I exhausted my rights of ownership as to the lessee for any uses it desires to make of the premises?  Of course not.  The Court needs to make this clear.

            So what was the argument by the losing party below?  The following is quoted from their brief:

Under the patent exhaustion doctrine that this Court has applied for more than 90 years …. an authorized first sale of a patented article exhausts the patent owner's rights in that article, and nullifies any “conditions” that the patent owner has tried to attach to its use or resale. Beginning with its decision in Mallinkrodt v. Medipart, Inc., 976 F.3d 700 (Fed. Cir. 1992), however, the Federal Circuit has steadily eroded the exhaustion doctrine. In this decision the Federal Circuit held that exhaustion is entirely optional, and easily nullified by a “notice” announcing that the patent owner would prefer that it not apply. That is an unprecedented and extremely dangerous expansion of the patent monopoly, in direct conflict with numerous decisions of this Court.

Frankly, this misrepresents the Federal Circuit’s actual ruling, prior law, and what the case involves.

The basic rule is and has long been that ideas of exhaustion and first sale only apply where there has been an authorized unconditional sale of a product or a copy or an unconditional and unrestricted license.  As the Federal Circuit commented:

The theory behind this rule is that in [an unconditional sale], the patentee has bargained for, and received, an amount equal to the full value of the goods.  This exhaustion doctrine, however, does not apply to an expressly conditional sale or license. In such a transaction, it is more reasonable to infer that the parties negotiated a price that reflects only the value of the 'use' rights conferred by the patentee."

The LG?Intel license expressly excluded granting a license to Intel’s purchasers to combine Intel's parts with other components in ways that would infringe LG’s other patents.  This conditional agreement required Intel to notify customers of the limited scope of what they were purchasing, which it did.  Although Intel was free to sell the products, those sales were required to be conditional, and Intel's customers were expressly prohibited from infringing LG's combination patents.

            This is the proper rule. If rights owner chooses to commercialize its invention by apportioning rights or permissions, rather than simply selling product, that choice should be respected.  It is not an effort to circumvent ideas of exhaustion or first sale, but simply a decision to not engage in transactions to which those theories apply.

            In Nagrampa v. Mailcoups, Inc., 469 F.3d 1257 (9^th Cir. 2006) the Ninth Circuit held unconscionable as to a California franchisee,  a mandatory forum and arbitration clause picking Massachusetts as the venue for disputes in a franchise agreement . The alleged procedural problems were the unbalanced bargaining positions and a lack of actual bargaining. But contracts are seldom bargained as to matters other than price and quantity or scope. 

             The alleged substantive problems? 

(1) The clause lacked mutuality because the franchisor could use judicial process in some cases (e.g., to protect its trademark rights), while the franchisee could not. But contract law has never required that each clause have mutuality, only that the entire contract reflect a mutuality of obligation, which this one did.

(2) “[The] contract would require a one-woman franchisee … to fly across the country to arbitrate a contract signed and performed in California…. She may not be able to maintain her claim … The forum selection provision has "no justification other than as a means of maximizing an advantage over [franchisees]."” But the Supreme Court has held that mere cost is not enough to invalidate such clauses. Also, there was a justification for the clause – without it, the franchisor would be forced to litigate in all of the states in which it granted franchises and, thus, to charge all franchisees a premium to enable it to do so.

            Nagrampa is bad enough, but the Ninth Circuit later continued the exegesis of its view of law in Douglas v. U.S. District Court, 2007 WL 2069542 (9^th Cir. 2007), a mandamus action. There, after holding that an amendment of an online agreement lacked assent, the Ninth Circuit gratuitously commented that even had there been assent, the terms for class action waiver and mandatory arbitration were unconscionable. The procedural flaw: the clause was a non-bargained standard form. The court said: “The district court held that the arbitration clause [was] not procedurally unconscionable … because Douglas had meaningful alternative choices for … service. Under New York law [this] forecloses any procedural unconscionability claim. However, [in Nagrampa] we noted that California “has rejected the notion that the availability ... of substitute ... services alone can defeat a claim of procedural unconscionability.” [A] contract can be procedurally unconscionable if a service provider has overwhelming bargaining power and presents a “take-it-or-leave-it” contract to a customer - even if the customer has a meaningful choice as to service providers.”

            But– why is that true? If the other party has a valid option that would allow it to choose another provider and it does not exercise that option, why is the result procedurally unconscionable?  The court did not provide any answer to this.

            Then in Davis v. O'Melveny & Myers, 485 F.3d 1066 (9^th Cir. 2007), the Ninth Circuit held that an arbitration clause proposed by a law firm and assented to by an employee was unconscionable. It was procedurally unconscionable not because of any lack of actual notice, use of misleading language or terms, or lack of options for the employee. It was procedurally unconscionable because the law firm did not offer the employee the option of rejecting the clause and continuing as an employee with an unmodified contract. The law firm “merely” offered the employee a three month period of employment in which to decide and accept the term or find a new job.

            This collection of cases is bad law that transfers making and amendment of contracts to courts, rather than parties acting in a marketplace. Procedural unconscionability is properly associated with lack of notice, deception, advantage-taking, and quasi fraud. As the doctrine itself states: “The principle is one of prevention of oppression and unfair surprise and not of disturbance of allocation of risks because of superior bargaining power.”

             Under the Ninth Circuit view, however, procedural unconscionability morphs into mere use of non-negotiated standard forms and sets the stage for a court hostile to the use of arbitration and class action waivers (or other terms) in consumer and employee cases to invalidate otherwise valid agreements. 

• First, SSOs must recognize the inevitability of competition issues whenever a proposed standard would resolve current competition in the market place.
• Second, consensus standards-setting should in fact reflect a true consensus and, particularly, agreement among all major competing firms in the affected market before a mandatory or quasi-mandatory standard is promulgated. This gives veto power to all competitors and runs counter to the goal of those who seek expansive scope for standards drafted on the basis of pure technology, but it recognizes the inevitable competition impact that such mandatory standards may have.
• Third, where standards are proposed in environments where two or more approaches are in active competition, any standard adopted should be permissive by focusing on only characteristics shared among the competitors, are unlikely to adversely affect one or another of the competitors, or are formulated solely for one of the approaches and not mandated for the others.