• Selected
• Entire List

Abstract:
This paper examines the Federal Trade Commission's recent settlement with Intel. Evaluating the FTC's case against Intel turns on tricky issues regarding the dynamics of cross-licensing. These are essentially barter transactions, and we have only a weak understanding of when firms will turn to barter. Interfering with cross-licensing will make it more difficult for these transactions to take place, and to understand the importance of that, we need to have a better handle on the relative importance for a licensor of cash returns versus the in-kind returns that are obtained from cross-licenses.

We can say with more confidence that the FTC's case appears to give very little weight to the benefits that arise from royalty-free cross-licenses. These licenses eliminate the double monopoly problem that can arise when two patent holders hold essential patents. Royalty-free cross-licensing eliminates through a contract an externality between the patent holders that would otherwise push up prices, to the detriment of the patent holders and their customers. The settlement may very well make it more difficult for Intel to negotiate royalty-free cross-licenses and may harm society in doing so.

The licensing regime that emerges from the settlement may have the benefit of making it possible for prospective PC makers who might not deal with Intel to do so -- though this point does not appear to have figured in the FTC's calculus. These prospective PC makers will have less reason to fear that Intel will later pressure them into a cross-license. This will increase the pool of PC makers, though the benefits of this are quite speculative. Equally speculative is whether the settlement will foster R&D on microprocessors -- the chief focus of the FTC's complaint -- though there is little public evidence to suggest that outcome.

Abstract:
Since the passage of the Interstate Commerce Act (1897) and the Sherman Act (1890), regulation and antitrust have operated as competing mechanisms to control competition. Regulation produced cross-subsidies and favors to special interests, but specified prices and rules of mandatory dealing. Antitrust promoted competition without favoring special interests, but couldn`t formulate rules for particular industries. The deregulation movement reflected the relative competencies of antitrust and regulation. Antitrust and regulation can also be viewed as complements in which regulation and antitrust assign control of competition to courts and regulatory agencies based on their relative strengths. Antitrust also can act as a constraint on what regulators can do. This paper uses the game-theoretic framework of political bargaining and the historical record of antitrust and regulation to establish and illustrate these points.

Abstract:
More than a century ago, the federal government started controlling competition, first railroads through the Interstate Commerce Act and then the general economy under the Sherman Act. The Commerce Act assigned primary responsibility to the first great federal agency, the Interstate Commerce Commission, while the Sherman Act relied for its implementation on federal courts of general jurisdiction. Since that time, there has been an ongoing struggle to formulate the appropriate policy for controlling competition and to determine the right balance between antitrust and regulation for implementing that policy.

Regulation and antitrust are two competing mechanisms to control competition. The early history in which special courts were established and then abolished, and in which the FTC was created illustrate this point. The relative advantages and disadvantages of each mechanism became clearer over time. Regulation produced cross-subsidies and favors to special interests, but was able to specify prices and specific rules of how firms should deal with each other. Antitrust, especially when it became economically coherent within the past 30 years or so, showed itself to be reasonably good at promoting competition, avoiding the favoring of special interests, but not good at formulating specific rules for particular industries. The partial and full deregulation movement was a response to the recognition of the relative advantages of regulation and antitrust. This does not mean that no sector will be regulated, but rather that competition, constrained only by antitrust, will be used over more activities, even in regulated industries.

Aside from being viewed as substitutes, antitrust and regulation can also be viewed as complements in which the activities of an industry can be subject to both regulatory and antitrust scrutiny. In this way, the complementary use of regulation and antitrust can assign control of competition to courts and regulatory agencies based on their relative strengths, and in some settings, antitrust can act as a constraint on what regulators can do. The trends in network industries indicate that regulators, not antitrust courts, will bear the responsibility for formulating interconnection policies in partially deregulated industries, but antitrust will remain in the background as a club that firms can use if regulators allow incumbents to acquire market power either through merger or predatory acts. The history shows that at least for the United States, the increased use of the Sherman Act instead of regulation to control competition, and when necessary, the complementary use of the two, has brought benefits to consumers.

Interstate Commerce Act, Sherman Act, federal regulation, restraint of trade, monopoly

Abstract:
This paper considers the proposed settlement agreement between Google and the Authors Guild relating to Google Book Search (GBS). I focus on three issues that raise antitrust and competition policy concerns. First, the agreement calls for Google to act as agent for rightsholders in setting the price of online access to consumers. Google is tasked with developing a pricing algorithm that will maximize revenues for each of those works. Direct competition among rightsholders would push prices towards some measure of costs and would not be designed to maximize revenues. The consumer access pricing provision might very well fail a challenge under Section 1 of the Sherman Act. Second, and much more centrally to the settlement agreement, the opt-out class action will make it possible for Google to include orphan works in its book search service. Orphan works are works as to which the rightsholder cannot be identified or found. The opt-out class action is the vehicle for large-scale collective action by active rightsholders. Active rightsholders have little incentive to compete with themselves by granting multiple licenses of their works or of the orphan works. Plus under the terms of the settlement agreement, active rightsholders benefit directly from the revenues attributable to orphan works used in GBS. We can mitigate the market power that will otherwise arise through the settlement by expanding the number of rights licenses available under the settlement agreement. To do that, we should take the step of unbundling the orphan works deal from the overall settlement agreement and create a separate license to use those works. All of that will undoubtedly add more complexity to what is already a large piece of work, and it may make sense to push out the new licenses to the future. That would mean ensuring now that the court retains jurisdiction to do that and/or giving the new registry created in the settlement the power to do this sort of licensing. Third, there is a risk that approval by the court of the settlement could cause antitrust immunities to attach to the arrangements created by the settlement agreement. As it is highly unlikely that the fairness hearing will undertake a meaningful antitrust analysis of those arrangements, if the district court approves the settlement, the court should include a clause-call this a no Noerr clause-in the order approving the settlement providing that no antitrust immunities attach from the court's approval.

D4, K20, K21, K41, L4, L43, O34

Abstract:
This paper considers the proposed settlement agreement between Google and the Authors Guild relating to Google Book Search. Google boldly launched Google Book Search in pursuing its goal of organizing the world’s information. Even though Google was sensitive to copyright values, the service relied on mass copying and thus Google undertook a substantial legal risk in setting up the service. That risk was realized with the lawsuits by the Authors Guild and the Association of American Publishers. The October, 2008 settlement agreement for those suits will create an important new copyright collective and will legitimate broad-scale online access to United States books registered before early January, 2009.

The settlement agreement is exceeding complex but I have focused on three issues that raise antitrust and competition policy concerns. First, the agreement calls for Google to act as agent for rights holders in setting the price of online access to consumers. Google is tasked with developing a pricing algorithm that will maximize revenues for each of those works. Direct competition among rights holders would push prices towards some measure of costs and would not be designed to maximize revenues. As I think that that level of direct coordination of prices is unlikely to mimic what would result in competition, I have real doubts about whether the consumer access pricing provision would survive a challenge under Section 1 of the Sherman Act.

Second, and much more centrally to the settlement agreement, the opt out class action will make it possible for Google to include orphan works in its book search service. Orphan works are works as to which the rightsholder can’t be identified or found. That means that a firm like Google can’t contract with an orphan holder directly to include his or her work in the service and that would result in large numbers of missing works. The opt out mechanism - which shifts the default from copyright’s usual out to the class action’s in - brings these works into the settlement.

But the settlement agreement also creates market power through this mechanism. Absent the lawsuit and the settlement, active rights holders could contract directly with Google, but it is hard to get large-scale contracting to take place and there is, again, no way to contract with orphan holders. The opt out class action then is the vehicle for large-scale collective action by active rights holders. Active rights holders have little incentive to compete with themselves by granting multiple licenses of their works or of the orphan works. Plus under the terms of the settlement agreement, active rights holders benefit directly from the revenues attributable to orphan works used in GBS.

We can mitigate the market power that will otherwise arise through the settlement by expanding the number of rights licenses available under the settlement agreement. Qualified firms should have the power to embrace the going-forward provisions of the settlement agreement. We typically find it hard to control prices directly and instead look to foster competition to control prices. Non-profits are unlikely to match up well with the overall terms of the settlement agreement, which is a share-the-revenues deal. But we should take the additional step of unbundling the orphan works deal from the overall settlement agreement and create a separate license to use those works. All of that will undoubtedly add more complexity to what is already a large piece of work, and it may make sense to push out the new licenses to the future. That would mean ensuring now that the court retains jurisdiction to do that and/or giving the new Registry created in the settlement the power to do this sort of licensing.

Third, there is a risk that approval by the court of the settlement could cause antitrust immunities to attach to the arrangements created by the settlement agreement. As it is highly unlikely that the fairness hearing will undertake a meaningful antitrust analysis of those arrangements, if the district court approves the settlement, the court should include a clause - call this a no Noerr clause - in the order approving the settlement providing that no antitrust immunities attach from the court’s approval.

Google, Google Book Search, Google Settlement, Authors Guild, antitrust

Abstract:
In this paper, I consider how copyright law influences entry in digital distribution of music and video. The subject encompasses past and current successes in distribution-cable TV and the VCR-current and recent controversies-Napster and the pending cases addressing its successors-as well as possible next steps in distribution, such as web radio, interactive music services and the digital video recorder.

Section I of the paper considers six ways that online distribution matters: 1. as a new medium, online distribution adds to the existing set of versioning opportunities for producers; 2. distributional bottlenecks are weakened and gatekeeping roles minimized; 3. sellers receive direct, detailed information about consumer preferences; 4. bundling and packaging opportunities are greatly expanded; 5. pay-per-view or pay-per-listen is easier to implement; and 6. it is possible to devolve control over distribution through peer-to-peer distribution.

Section II of the paper considers two cases of distribution entry, devices and online radio. Much of the relevant distribution entry policy is set through copyright law. For new devices that facilitate distribution-the VCR, Napster and the DVR-key features of the reigning copyright test are not sufficiently demanding of entrants. The Sony test for contributory copyright infringement-whether the object in question is capable of substantial noninfringing uses-is far too weak and fails to take into account at all the scope of the infringing uses that will result. It is bad third-party copyright policy. Sony may fare better as a matter of independent entry policy and the flexible fair use doctrine of copyright law creates room for courts to operate in setting that policy in an economically sensible fashion. As to online radio, we have severely constrained potential entry in digital radio through recent copyright enactments. Incumbent over-the-air stations are substantially favored and diversity and entry are limited by statutory fiat. The current statutes as written are troubling enough, but the statutory distortions are being exacerbated through the royalty setting process before the Copyright Arbitration Royalty Panel.

antitrust, intellectual property, Internet, cyberspace

Abstract:
With Web 2.0, we have once again changed how we use computers. That change has brought with it new intermediaries who sit at the crossroads of the matching and coordination that define how we use the Internet today. Those intermediaries - Google first and foremost - have access to extraordinarily detailed information about their customers. That information arises naturally from the very services provided. We will see a similar pattern as cloud computing becomes more important, and cloud service providers will also have available to them a rich datastream that arises from their customer's activities.

To date, these intermediaries have faced few limitations in how they use the information that they see. That information can be used to improve their core businesses - adding collective intelligence to search to increase relevance - and to finance - through advertising backed by rich databases that allows ads to be matched to individual customers - virtually any content or service that can be provided through a screen. To focus on Google as the largest player in this space, there is no obvious limit to its scale and an advertising-supported business adds revenue with each additional screen that is viewed.

In the past, we have regulated intermediaries at these transactional bottlenecks - banks, cable companies, phone companies and the like - and limited the ways in which they can use the information that they see. Presumably the same forces that animated those rules - fundamental concerns about customer privacy - need to be assessed for our new information intermediaries. In doing that, we need to be acutely aware of how our choices influence competition. An uneven playing field - allowing one firm to use the information that it sees while blocking others from doing the same thing - creates market power through limiting competition. We rarely want to do that. And privacy rules that limit how information can be used and shared across firms will artificially push towards greater consolidation, something which again usually works against maintaining robust competition.

Web 2.0, cloud computing, privacy, behavioral advertising, Google, antitrust

Abstract:
This paper considers the four antitrust decisions in the Supreme Court's 2006 Term.

It offers brief discussions of Weyerhaeuser and Credit Suisse. Weyerhaeuser is a small, modest decision. The Court isn't likely to see another predatory bidding case soon and the Court chose to minimize doctrinal complexity by bringing predatory bidding analysis in sync with the Court's prior treatment of predatory pricing in Brooke Group. Credit Suisse too is minimally incremental. In concluding that federal securities law implicitly precluded claims asserting antitrust violations in the sale of new securities, the Court followed its prior decision in Gordon as well as the Court's more recent preference for regulatory schemes over antitrust as seen in Trinko. Pushing antitrust authority toward specialized regulators like the Securities and Exchange Commission broadens the trade-offs that can be made between antitrust concerns and other values and almost certainly expands the circumstances under which industry actors can act collectively. That matters, so Credit Suisse covers more of the economic landscape than Weyerhaeuser, but the decision itself is a small step from prior doctrine.

Twombly and Leegin are each, in their own ways, blockbusters. Twombly will appear in case after case, as antitrust defendants try to rely on its new tougher rules for FRCP 12(b)(6) motions. Twombly represents a preference for blunt instruments over sharp edges. The central problem confronted by Twombly is discovery run amok. The Court has the tools in its hands to control that by rewriting the discovery rules and overturning lower court decisions implementing those rules. Twombly suggests that the Court believes that refinement of those rules will fail in controlling discovery and it is willing to pay the price that private plaintiffs will have no good way to get at the best-hidden antitrust conspiracies.

Finally, Leegin brings to a close - for now or forever? - the 100-year saga of contractual minimum resale price maintenance. Since its decision in 1911 in Dr. Miles, the Court has confronted this issue again and again in the slightly-refined versions that make up the art of institutional design. Over time, the Court has chipped away at Dr. Miles, first in not finding a violation of Section 1 of the Sherman Act for the unilateral minimum RPM in Colgate in 1919 and in then broadly subjecting nonprice vertical restraints to rule-of-reason treatment in Sylvania in 1977. Given that, on what basis would Dr. Miles survive?

That is a question of stare decisis and Leegin ends up in an all-out fight over stare decisis in antitrust. That is new: the Court has been overturning old decisions in antitrust for some time and has done so with little stare decisis fanfare. That suggests that the dispute over stare decisis in Leegin is just a convenient forum for the larger dispute over stare decisis that is percolating through a divided Court. I don't have a full-blown theory of stare decisis but I do suggest why the Court has been mistaken to treat stare decisis in statutory cases differently from that in constitutional cases. The Court has made too little of one of its critical tools in shaping statutes, namely, the power to set a default point for subsequent congressional action. Once we treat the Court's decisions as inputs in subsequent lawmaking, there is greater reason to think that the Court should have a uniform approach to stare decisis across the Constitution and statutes.

Supreme Court

Abstract:
Encryption propertizes copyright. Prior to the rise of encryption, intellectual property wasn't really property. Instead, these rights consisted of rights to sue to block use or seek damages for after-the-fact use. Encryption makes possible before-the-fact limits on use of the sort that we associate with physical property. Copyrighted works are becoming real property in a way that will never happen for trademarks or patents.

This essay traces the role of consent and refusal for copyrighted works in the context of mass media entertainment. Early phonograph and movie projector manufacturers sought to limit the use of the equipment to specified music or films, almost certainly in an effort at price discrimination. With the rise of radio, we see similar use restrictions imposed by the record companies, but these restrictions should be seen first as a form of raising rival's costs. The fight over home-taping use of the VCR should be seen in consent terms as well, where entry was clearly facilitated by the absence of any requirement of advance consent from copyright holders.

Now with the current dispute over a possible broadcast flag for digital TV, we may complete the path started by Edison and his contemporaries. In considering the extent to which we should embrace full propertization of copyrighted works through encryption, we should expect transaction forms to vary with transaction costs and technological possibilities and should not somehow privilege a narrow set of institutional arrangements that arose under different constraints.

digital TV, copyright, encryption, mass entertainment

Abstract:
Copyright has emerged as a pliable tool, to be bent and shaped by firms and frequently with an eye towards disadvantaging competitors through the erection of entry barriers. The easy manner in which copyright arises makes it possible for firms to get copyrights and threaten competitors with costly infringement actions. This is the use of copyright as more than just defining property rights, the use of copyright in creating market locks. But we would paint with too broad a brush were we to condemn all of these market locks.

Market locks facilitate product differentiation and that may expand the range of ways that fixed costs can be recovered in a competitive industry. This can be useful and can improve outcomes for consumers. We should think this pattern to arise most plausibly in industries with foremarkets and aftermarkets. These would include original equipment markets and repair parts, printers and toner cartridges and garage door openers and would encompass a series of important cases, including Chamberlain, Kodak, Lexmark and Toro. Market locks in these settings may appropriately limit partial entry, as when an entrant wants only to supply replacement parts. In these settings, contractual product degradation - typically a license limiting permitted uses of the product - will make it possible to offer different products to different consumers. We should not routinely condemn market locks in these situations and should be troubled if we shape copyright law in a way that prevents these market locks from operating.

At the same time, we should be troubled by market locks that create entry barriers for an entrant willing to enter on the same scale as the incumbent, for an entrant willing to undertake full rather than partial entry. These will frequently be situations characterized by high switching costs, where the incumbent's advantage may arise from the simple fact of being first. In these cases, copyright law can make entry barriers concrete and we should be concerned if copyright law works to disadvantage full entry. This pattern matches well with cases such as IMS, Lotus and Southco, though courts have done reasonably well in making possible entry, even if the courts have only glimpsed the full competitive issues at stake.

Finally, the Digital Millennium Copyright Act has emerged to play a central role in these cases. Firms looking to limit use naturally move from weakly-enforceable paper contracts to self-enforcing technological contracts in the form of lock-out chips and the like. But the heart of the DMCA is technological controls for copyrighted works and the DMCA does not implement a pure regime of technological contracts. It is hardly surprising that firms in cases such as Chamberlain and Lexmark have tried to squeeze within the DMCA and no less expected that the courts have refused to expand its scope. That reluctance, though, tells us little about the real merits of technological contracting or about the true copyright protection scheme erected by the DMCA.

copyright, DMCA, after-markets

Abstract:
Next time you turn on your television, actually watch the commercials and you will quickly see how poorly the economic model of TV is working. They put on a commercial for dog food, but you don't have a dog. Many of the commercials are for product categories that you do not purchase, and others are for products, such as cars or computers, that you use constantly but purchase only sporadically.

The digital video recorder (DVR) may change this dramatically. The DVR will allow us to unbundle content and advertising. Content that comes from broadcasters bundled in one form - the TV show itself, the station identifications, the ads selling Budweiser and the promos for a very special Dawson's Creek - can be reshaped and separated before the viewer sees it. The ability to delete commercials puts at risk the basic financing model for free broadcast TV. At least as important, unbundling of ads and content allows personalization of commercials and that in turn may change content itself.

Personalization will change the core role that content plays in intermediating between advertisers and audiences. Traditionally, content plays a dual role: it attracts a particular type of reader, and that in turn determines the type of advertising that can be sold. Creators shape content to best fit the intersection of advertisers and audience. A DVR with ad personalization has the capacity to alter completely this critical matching process played by content creators.

How the DVR technology is organized will almost certainly help to determine whether viewers can commit to not deleting the ads. Decentralized, free-standing DVRs - the current model - will make commitment difficult. Centralizing the DVR technology in the pipes bringing the content into your home - putting the DVR technology in the cable box - may make the ad commitments more credible. The range of possible, supportable economic models turns directly on how the DVR technology is organized and those models matter directly for the kind of content that will be made available.

How the technology is organized also turns out to be quite important for law. Standard legal instruments such as copyright or contract work only so well with widespread, decentralized use of a technology. Centralized technology is easier to control, either through contract; through common law doctrines such as contributory copyright infringement, the tack tried in Sony; or through direct regulation, as we have done with cable. The regulatory path for the DVR probably turns on whether it emerges as a decentralized technology ala the VCR or as part of the cable/DBS system. And, if the DVR technology is centralized, we may see a substantial asymmetry between broadcast and cable. The dispute over "must carry" will morph into a fight over "must store" or "must be smart" as over-the-air broadcasters will seek access to the storage and intelligence that will come to reside in the set-top box.

digital video recorder, DVR, TiVo, copyright, cable, broadcast television

Abstract:
In this paper, I make a number of points about the Microsoft case itself and the next steps that should take place. In particular, I argue that:

- No Liability for Tying. Microsoft should not be found liable under the Sherman Act for tying Internet Explorer to Windows. In the pre-networked world, Windows played the central role in coordinating the sharing of software. Incorporating a browser would have been perfectly consistent with that role.

- The Drop in the Cost of Software Coordination. The rise of the network changes how software should be distributed and changes the role of Windows in software coordination. There is less of a need for mandatory incorporation of software into Windows, as decentralized distribution and coordination is now possible.

- Distorted Distribution Channels. As found by the D.C. Circuit, Microsoft engaged in impermissible monopoly maintenance. In so doing, Microsoft distorted the channels for software distribution and added software to Windows for the purpose of raising the cost of distribution of rival software.

- Distribution Remedies. A proportionate Microsoft remedy should address that distributional distortion and seek to prevent future distortions. These remedies should:

- foster desktop flexibility for distributional intermediaries, so that there are no mandatory icons on the Windows desktop or spots reserved in the Start Menu or its equivalent;

- require Microsoft to engage in mandatory versioning, so that it issues Windows versions with and without any new middleware that it adds to Windows;

- impose a moratorium period of six months to 2 years during which Microsoft would be able to distribute through distributional intermediaries only the baseline Windows without new middleware, while permitting distribution of the full version of Windows via CD or Microsoft's website;

- after the moratorium period, rely on competition among software producers and others for distributional intermediary shelf space to control software distribution abuses; and

- sunset, perhaps after a period of three years.

- Equilibria and Restoring Competition in Distorted Markets. The direct-distribution-only remedy will likely impose interim costs on consumers. We need to assess those costs and understand whether they need to be borne. That turns in part on whether preventing further anti-competitive acts will suffice to create the competitive level that would have existed absent Microsoft's acts, or whether such competition can be restored only through a more direct measure such as the suggested direct-distribution-only remedy.

Microsoft, antitrust

Abstract:
The wonder of the Internet is incredibly capable computers connected with each other under the control of individuals. For all of the reasons that we think that decentralization is a powerful force we have applauded the ability of individual users to set up websites and make their ideas available to others. But there is a dark side as well. Always - on connections, extra computing cycles and gigabytes of storage to burn mean that individual decisions can propagate throughout the network quickly. The small-worlds phenomenon that is the Internet means that my computer is only a handful of clicks away from a malicious computer programmer.

My decisions matter for your computing life. A malicious hacker can turn my computer into a zombie and use my broad-band connection and my computer to shut down websites, to send millions of spam emails, or worse. The network is a sea of computing externalities, many extraordinarily positive but others that can range from everyday bothersome to enormously disruptive. And, in the hands of a cyber-terrorist, the more we embed critical infrastructure into the public network, the more we make it possible for a cyber-terrorist to turns our computing resources against us and thereby harm critical infrastructure, such as the electricity grid or our communications networks.

Addressing cyber security is a mixed question of engineering - computing architecture - and legal rules. The zombie PC problem emerges with the rise of the Internet and decentralized control over PCs. The pricing structure of the Internet world-one-price, all-you-can-eat broadband and lumpy computing power in the form of powerful CPUs kills off many of the natural incentives for an individual to ensure that her computing resources are not being used by others. This can be good, as it creates many opportunities for sharing, but the downside is that there is little reason for the individual computer user to police against zombification.

In this article, I consider two issues in detail. The monoculture argument is one approach to architecting the network. That argument suggests that we should focus on forcing heterogeneity in operating systems to enhance our cyber security. I think that is the wrong emphasis. On its own terms, the argument tells us little about the extent of diversity that would be required to achieve meaningful protection, especially if our concern is the cyber-terrorist. The argument also ignores the more important question of adaptability, meaning how quickly can the current system adapt to new conditions. Instead, I argue in favor of the traditional approach of isolation - autarky - in separating critical infrastructure from the public network.

Second, I consider the way in which liability rules for software might influence the quality of software and software use decisions. Hackers can exploit defects in software to seize control of machines. Fewer defects to exploit and we might reduce the harms of hackers. This turns out to be tricky. Broad liability rules that would protect consumers from the harms of hacking will lead to the standard moral hazard problem that we see in insurance. Consumers who shouldn't be using computers or on the network will jump on once they are protected from hacking losses.

These are standard products liability issues, but software has two particular features that suggest that we should not just apply our standard approaches to products liability. First, we learn about software through use. One piece of software is combined with other software in a way that a Coke bottle is rarely combined with anything else. Second, software can adapt and can be fixed in place after-the-fact. Both of these features should push towards earlier release of software, for buggy software to be fixed later.

cyber-security, monoculture, network externalities, computer software, product liability rules

Abstract:
In this paper, I make four points.

1. The copyright act defines use rights, not access rights. That overstates slightly - especially with the Digital Millennium Copyright Act in the statute - but the core of copyright law addresses how works can be used assuming that legal access has been obtained. Other law addresses the circumstances under which works can be accessed.

2. Nothing in copyright itself suggests that use rights should trump access rights; indeed, our core access principles suggest just the opposite. We frequently speak of a fair use "right." I am doubtful about that on its own terms but even if we find something there, a fair use right isn't an access right. Fair use doesn't equal fair access.

3. The scope of rights given to an initial author will effect the timing and scope of investment she will make in creating a work. For many works, those investments can be made in discrete lumps. As a society, we want investments to be made incrementally rather than as one large lump as doing so allows us to get feedback from the market on the value of a work. We don't want to throw good money after bad, and if we learn that, say, the English version of a work is a failure, we don't want to bother translating it into Mandarin. Plus we will delay the time that works reach the market if we create an incentive to do large, lumpy investments rather than a sequence of investments coupled with market feedback. Authors start with one monopoly: their unique access to the work that they have created. If we do not give authors control over these follow-on works, authors will overinvest upfront in the works, since that is the only way that the can gain a return on their initial monopoly over access to the work. In that situation, we are better off to hand the author a statutory monopoly over the follow-on work rather than see the author invest real resources in creating a property right over that work.

4. Fair use is a form of rights bundling. If we decide that, say, format-shifting is fair use or is otherwise a permitted use - you sell me a music CD and I have a use right to make a personal copy on a cassette or my iPod - we are making a decision about the rights that we are bundling together. The nature of bundles is that everyone gets stuck buying the same set of rights. These bundles can be inefficiently large. Consumers would often be better off if instead we allowed rights to be unbundled, so that consumers could buy just those rights that they wanted rather than being forced to take unwanted rights. Doing that requires a narrow conception of fair use.

copyright, use rights, access rights, Digital Millennium Copyright Act

Abstract:
We are at an unusual moment in telecommunications. We have two very live cases of entry: Internet protocol television (IPTV) and municipal wireless broadband. IPTV will create new competition with cable, satellite and over-the-air broadcast TV, promising lower prices and new services. Muni wireless makes it possible for local communities to add new broadband capabilities to compete with DSL and cable broadband.

Unsurprisingly given the newness of the services, there is substantial uncertainty about whether and how these services should be regulated, and we have seen legislative action at municipal, state and federal levels. To assess that, I set forth a general framework for matching jurisdictions to tasks and consider coordination costs; information aggregation; speed; tailoring; and competence, capture and corruption. I also set forth a typology of legislative approaches: mandatory federal; default federal; uniformity by choice; experimental labs and competitive federalism; and mixed jurisdiction regulation. I also consider specific regulatory issues for telcom entry control, namely, the extent of the natural monopoly and the desired level of cross-subsidization. I consider four prior of telcom entry and regulation: cable TV franchising; control over pole attachments; the local entry preemption provisions of the 1996 Telecommunications Act; and entry into satellite broadcasting.

With that framework, I turn to muni wireless and IPTV. Muni wireless turns on decisions about quintessentially local assets, such as municipal light poles, and provision of the service involves few across-jurisdiction externalities. Plus there are a number of decisions regarding the service - tradeoffs between advertising and price and disagreements over the extent of the digital divide - that suggest we should see natural variation in the services. That makes local control appropriate and we should be critical of contrary state and federal efforts. For IPTV, a critical question is the required scope of entry: to what extent will an entrant be required to build-out services to serve an entire local area? Again that turns on contested conceptions about universal service obligations and would be best addressed locally.

Internet protocol television, cable, satellite, DSL, broadband, regulation, 1996 Telecommunications Act

Abstract:
The powerful shift in copying technology over the last thirty years has destabilized how we produce copies and the economic arrangements associated with prior technologies. These technological changes have created a broad shift in the ability to make copies moving control away from producers towards consumers. As a consequence, these technologies have altered the practical enforceability of the rights that law assigns to copyright owners.

Digital rights management technologies are an effort to make meaningful the legal rights of copyright owners. DRM faces severe obstacles. For preexisting products like the music CD, it has proven to be very difficult to add DRM after the fact. CDs need to work in standard CD players, and the limits DRM. The firestorm over Sony BMG's effort to produce CDs subject to DRM suggests that we are unlikely to see meaningful DRM for music CDs soon.

But we are switching how we deliver content from products to services. Music CDs and eventually DVDs will be replaced by online services such as Apple's iTunes and Google Video. Both of these come with DRM built-in and both rely on identity-based DRM. Identity-based DRM ties identity to content. Content can be shared widely, but absent access to identity, the content is worthless.

This is a substantial step forward for DRM, but may still be a step short of where we need to be. Content purchasers still have no reason to protect purchased content. Identity-based DRM coupled with bounty tags will create an incentives wedge between content purchasers and stripping/p2p software and with peers in a p2p network. We should want a system where content purchasers are as careful with content as they would be with identity and mistrust-based DRM may be that system.

DRM, digital rights management, sony, BMG, rootkit, iTunes, Google Video

Abstract:
This is a comment on Kal Raustiala & Christopher Sprigman, The Piracy Paradox: Innovation and Intellectual Property in Fashion Design, 92 Va. L. Rev. 1687 (2006).

The Piracy Paradox builds on the fun of fashion to undertake a serious exploration of whether we can sustain innovation without property rights. That is an important question, as copyright brings with it a real cost in blocking follow-on uses and a new fashion copyright would limit subsequent copying. We need to ask whether that price is worth it. In this brief response, I emphasize two points. First, the case of the Fashion Originators' Guild of America suggests that we did see a design response to the private property rights regime created by the Guild. More property rights resulted in greater efforts to innovate. Second, copying is likely to be one-sided: low-end firms copy from high-end firms. With a fashion copyright, high-end firms could commit to their customers that they would not face quick matching by low-end copyists. Rapid imitation limits the value that high-end designers can promise to their customers.

Kal Raustiala, Christopher Sprigman, intellectual property, copyright

Abstract:
The emergence of distributed storage, machine intelligence and cheap communications has given rise to the networked product. These are products that can evolve even after versions of the product have been put into the hands of consumers. The most interesting consumer products of the day are networked products. This includes the natural successor to the VCR - whether the plain digital video recorder or the TiVo favored by the digerati - and the ubiquitous iPod and its less chic cousin MP3 players. This category also includes peer-to-peer software in its various forms, whether as Napster, Aimster or Grokster.

More than twenty years have passed since the Supreme Court confronted the VCR in the Sony case. The substantial noninfringing use test has both virtues and vices. It has provided a safe harbor for product innovation. It makes it possible for a creator to toss a product onto the waters to see what happens, having only a vague sense of what will happen next. But Sony also provides no reason for a creator to design products to eliminate infringing uses.

The core fight over Sony turns precisely on the uncertainty of what happens next: what is the next use of the product not seen today? But Sony is framed in the context of episodic design with an installed-base constraint and no real possibility of feedback between actual use of the product and design. We are at a very different point now. Networked products evolve and we are now going to frame what ongoing design obligations should exist with regard to these networked products.

Once we combine software with communication to create networked products we then have products that can evolve in real-time (and do). Smart products phone home and update themselves. Phoning home - and the control that results from that - is a choice and one that designers of networked products make every day. Design ceases to be a one-time event and instead becomes a continuous process. And that is true not only for the next product sold, but also for the entire installed base. The dead hand of the past and the constraints of backwards compatibility are lifted.

We need to update the Sony test to reflect these possibilities. If the producer chooses to let go of a networked product so that the producer cannot exercise control going forward and therefore cannot evolve the product in response to actual use, the producer should face a hard use test, perhaps one tied to whether the primary use of the product is noninfringing. If instead the producer ensures that the product can phone home so that updates can be promulgated throughout the system for the networked product, the producer should face a substantial non-infringing use test, coupled with the duty to evolve the product to eliminate infringing uses.

Napster, Aimster, Grokster, Supreme Court, Current Supreme Court cases,

Abstract:
This paper offers legal and economic analysis of two recent Supreme Court decisions, AT&T Corporation v. Iowa Utilities Board and Verizon Communications v. FCC. The paper is written with two audiences in mind. For those unfamiliar with the cases, we offer what we hope is an accessible yet detailed account of the underlying policy issues raised by a legal regime that requires incumbent local telephone carriers to lease parts of their telephone networks to would-be rivals. To that end, we discuss the main reasons why sharing rules are sometimes imposed in markets like the market for local telephone service, and we then link those issues to the specific legal questions at issue in these cases. For those already well versed in those issues, by contrast, we have woven into our account a variety of new ideas about both the relevant legal analysis and the underlying economics. We explain, for example, how low access prices might encourage incumbents to invest in new infrastructure despite the intuitive argument to the contrary, and how the Commission's seemingly nonsensical pick-and-choose rule can actually accomplish important policy goals, working in essence as a statutory most-favored-nation clause. In the end, then, we hope this paper will have value both for those relatively well steeped in telecommunications policy and for those just beginning to learn these issues.

telecommunications, Verizon, Iowa Utilities, natural monopoly

Abstract:
This essay was written for a volume celebrating Judge Richard Posner's 25 years on the bench. The article considers his opinion in Bank of America v. Moglia, which addresses the status of rabbi trusts in bankruptcy. The rabbi trust is first and foremost a tax device, a way to ensure a contigent delayed payment and yet do so without incurring current tax income to the beneficiary of the trust. But the key condition that delays the receipt of income - namely that the assets of the trust remain available to the general creditors of the entity creating the trust - means that the trust operates as an asset partitioning device and a way to protect assets for unsecured creditors from potential invasion by secured creditors. Asset partitioning is ubiquitous, and we see it in many forms, such as limited liability corporations and securitization. It is rare to see a device that works to the benefit of a changing group of general creditors. The rabbi trust, as seen in Moglia, emerges as superior to the negative pledge, which may be its closest cousin.

Richard A. Posner, Judge Posner

Abstract:
In this essay prepared for the American Bar Association's 56th Antitrust Law Spring Meeting, I consider two issues that pertain to the overall question of what antitrust doctrines are up for retirement. First, we can't consider that without understanding how the Supreme Court approaches stare decisis in antitrust. The Court's 5-4 decision in Leegin identified some of the fault lines on this issue. The Court has suggested that it should approach stare decisis differently in statutory areas from the way it approaches it when it reconsiders constitutional decisions. I think that that is wrong and that the Court should apply its approach to stare decisis in constitutional cases to cases based on statutes, such as the Sherman Act. Second, I focus on the evil of evils: horizontal price-fixing. I don't think that the Court is likely to retire the per se rule against horizontal price-fixing, certainly not directly. We might only realize that it had been overturned after the fact, after the Court had so chipped away at the doctrine that nothing remained. That said, as again Leegin itself suggested, we can't be fully confident that horizontal price-fixing is always pernicious, especially when it might be implemented as part of a larger vertical arrangement.

Abstract:
Scope-of-permission goods are goods of arbitrary scope, where consumption of the good is non-rivalrous, where users can be excluded from consuming the good - through market organization, technology or law - and where increments to the good can be added to the good, once they are created, at zero marginal cost. Scope-of-permission goods naturally include pay TV, computer software, copyrighted works and licenses from collective right collectives such as ASCAP and BMI.

These goods have been at the heart of some of our most difficult cases in antitrust law and competition policy. This includes the extended antitrust litigation over the blanket licenses for the use of copyrighted works issued by ASCAP and BMI. It also includes the Windows operating system, especially as it has grown over time with the addition of Internet Explorer and the Windows Media Player.

In the ASCAP cases and in the U.S. and EU antitrust actions against Microsoft, the core question is to what extent do we want to re-scope a scope-of-permission could so as to foster entry. In the recent revision of the 40-year-old consent decrees in ASCAP, we have once again pushed ASCAP to offer meaningfully smaller licenses - a required subtraction of scope - with the hope that we would create entry in collective rights organizations.

The U.S. and EU have taken different paths in their actions against Microsoft. Both focus on the scope of the rights given to end-users in Windows. The U.S. has chosen to limit the visibility of the Windows Media Player by allowing computer sellers to hide visible means of access to WMP. WMP remains present to rise up if invoked by a savvy consumer or by a third-party. In contrast, the European Commission has required Microsoft to engage in mandatory versioning, requiring Microsoft to offer computer sellers versions of Windows with and without WMP.

The U.S. remedy intrudes less directly into product design, the EU remedy does a better job of preserving competition in media players by limiting the reach advantage that Microsoft has by being able to tie and distribute WMP with Windows. But we had a better alternative available, one that was rejected by both the U.S. and the EU. Imposing a must-carry obligation - requiring Microsoft to distribute other media players if it chose to distribute WMP with Windows - would have neutralized Microsoft's ability to tie WMP to Windows, while avoiding concerns about fragmenting the programming infrastructure available to third parties. This would have created the possibility of strong competition, akin to the facilities-based competition we have sought to create in U.S. telecommunications.

At least for software, we should think that there are strong asymmetries between subtraction and addition remedies. Subtracting disrupts the natural flow of product development and leaves the software producer with the difficult task of unscrambling the software code. It also creates the risk of fragmenting the programming base available to third parties. Subtraction may be sensible when the underlying goods are more distinct - when we can separate Bach from the Beatles - but in the Microsoft cases, instead of subtracting scope, as we did in ASCAP and the U.S. and EU have done in Microsoft, we should have expanded the scope of Windows by imposing a must-carry remedy.

software, pay-TV, copyright, licensing, Windows

Abstract:
For individuals, the basic architecture of computing is changing. That is obviously about the device itself, with the desktop or laptop computer now being supplemented with other computing devices such as the smartphone and the netbook. That switch, coupled with ubiquitous wireless access, means that many people have access to computing power whenever and wherever.

The way in which we use these devices has changed. We have switched from the freestanding world of the desktop computer and the next stage of surfing the Internet net to consume provided content to a world in which users interact with each other. This is the world of Web 2.0, the world of Google, Facebook and Twitter. This is not just a change in use, but also a change in the organization of computing power and storage, cloud-computing in phrase.

This is also a world of identity, often direct actual real me, on Facebook and Twitter; an authenticated identity to access my data stored in the cloud when I use Google Reader or Gmail or another cloud-based mail service; and a browser-identity when I use a search service. And this is also a world of advertising. Web 2.0 and cloud-computing services are often free to individuals, but they have to be paid for somehow, at that is usually through advertising. Advertising is also increasingly important in a world in which the integrity of the copy itself has weakened and the copy may no longer serve as a reliable means of organizing payment for content.

This combination of identity and advertising means that this will be targeted advertising, that is, advertising directed to some version of me, perhaps actual me as Facebook sees me or browser-me as Google sees me. Regulators are now confronting this intersection of commerce and identity. Individuals have a real interest in seeing targeted advertising work. That advertising supports the free services and content that we have all come to expect on the Internet. But individuals also have a strong interest in controlling their identities.

Regulators, especially in the European Union, are moving towards what they regard as "privacy friendly" default settings for information tracking by Web 2.0 providers. To date, default settings have usually put the burden on individuals to opt out of information tracking. An EU privacy-friendly approach would seem to reject that. But Web 2.0 providers and cloud-providers have strong tools for inducing opt in and, indeed, their ability to provide different levels of their services for different individuals should make it possible for them to assess quite carefully what it takes to get individuals to opt in to targeted advertising. So long as those service providers are not blocked from providing different levels of service to individuals who have not elected to receive targeted advertising, moving towards the EU's privacy-friendly defaults may have the virtue of pushing us away from an often not-so-meaningful default opt in towards more meaningfully calibrated opt ins exchanged for higher quality services, such as seeing fewer, better matched ads.

Google, Facebook, Twitter, privacy, identity, targeted advertising, behavioral advertising

Abstract:
Three recent appellate decisions - Goldwasser, Trinko and Covad - have addressed the interplay of the 1996 Telecommunications Act and the antitrust laws. This area raises questions of both substantive law and standing. This essay focuses on standing and in particular the question of how the antitrust doctrine in Illinois Brick should apply to situations in which there is an alleged breach of an access duty owed by an incumbent local exchange carrier. That access duty might arise under the 1996 Act itself or under applicable antitrust doctrines, such as the essential facilities doctrine or the duty to deal with competitors seen in Aspen Skiing. The essay sets forth a model of access duties leading to entry and Cournot duopoly and evaluates outcomes when that access duty is breached. The essay discusses various approaches to allocating suit rights depending on the purpose of enforcing the duty. I argue that the Illinois Brick doctrine which bars suits by consumers as indirect purchasers should have little application to the breach of access situation as the de facto compensation rationale of Illinois Brick won't operate when the entrant has been denied the mandated access.

Goldwasser, Trinko, Covad, Illinois Brick, Aspen Skiing, Cournot duopoly

Abstract:
This is a review of Herbert Hovenkamp's book 'The Antitrust Enterprise: Principle and Execution.'

Herbert Hovenkamp, The Antitrust Enterprise

Abstract:
Text in hand, we have read books by candlelight, oil lamp and Edison's incandescent bulb, maybe even the occasional CFL. But even as light itself has changed, the book has remained constant. Until now. With the rise of Google Book Search and ebook readers like Amazon’s Kindle, we have entered the era of the mediated book. We will still browse and read books, but we will do so through a screen.

This is more than just a change in medium. Digital texts are inherently on-demand works, that is, works that can be produced at the instant that a consumer wishes to interact with the text. Physical books historically have been printed in batched runs in advance of demand. This fact of production matters relatively little for the texts themselves, as we typically want books to be fixed, reliable artifacts.

This changes matters for how we finance books. On-demand texts can be financed through advertising. Printing in advance means that embedded advertising has little chance of being relevant at the point of reading. Mediated texts can be updated instantly with new, continuously timely advertising. That advertising also can be personalized for individual readers as the interaction between the mediating device and the reader will create a rich information stream to enhance the relevance of this advertising. That process of course will raise standard privacy issues.

The short history of 20th Century advertising expenditures in the United States is characterized by two facts. First, overall expenditures as a percentage of GDP are relatively constant over time, bouncing around over the last sixty years between 1.5% and 2.5%. The emergence of new advertising platforms - say radio in 1927; broadcast TV in 1949; cable TV in 1980; and the Internet in 1997 - hasn’t altered that essential fact. The emergence of another new platform - advertising-supported books - isn't likely to expand overall advertising expenditures much if at all. Second, print’s advertising market share has declined steadily, from roughly 55% of advertising dollars in 1935 to a little under 21% in 2007.

Mediated content accounts for a large chunk of that decline. Now books and of course print more generally will be mediated too. And we will get a nice test. Does the decline in the role of print as seen in advertising dollars reflect the decline of words relative to images and sounds? Or is this a story not of content but of technology, in which a mediated platform is a better advertising platform? The rise of the new mediated books will change how we finance books and will change our understanding of the relative roles of content and technology in driving advertising.

Amazon, Kindle, Google Book Search, advertising

Abstract:
Google takes products out of beta status slowly, even while it is making substantial improvements in the product. Objectors will see the amended settlement agreement as a mixed bag, with some finding almost nothing in the changes (privacy advocates and those who fear high prices for institutional subscriptions), while others will find their concerns addressed (foreign governments acting, one hopes, with the correct sense of the interests of foreign authors).

The amended settlement agreement clearly responds to the concerns raised by the Department of Justice. In waiving the benefits of possible doctrines of antitrust immunity, the ASA solves a timing problem for DOJ. DOJ faced an all-or-nothing quandary: challenge the agreement now or risk the possibility losing the right to challenge it later. The Noerr waiver solves that problem. There may be real benefits to seeing how the pricing provisions play out in actual operating conditions. Don’t shadow box now but fight later if necessary. I could easily see DOJ or Judge Chin reaching that conclusion and choosing to defer consideration of the pricing issues to another day.

The orphan works licensing is differently situated. The revised agreement creates a new unclaimed works fiduciary but does so in incomplete fashion. The UWF takes over some of the registry’s responsibilities rather than acting as a true fiduciary for orphan works holders. Such a fiduciary would be situated to license the orphan works to third parties on a going forward basis. That would have been an elegant solution to the competitive issues raised by the current plan to grant a license to the orphan works to Google and only to Google. The revised settlement makes real progress on these issues only to stop short of a visible and attainable real solution.

Google, Google Book Search, Google Settlement, Authors Guild, antitrust

Abstract:
The Google Book Search Settlement has received a great deal of attention. In response to opposition, the original settlement has been delayed and will now be resubmitted. In this brief paper, I address three points. First, I do a quick status update on competition issues in the case. Second, I turn to a key issue that has emerged in the commentary on the competition issues, namely, what is the right way to frame the competition policy baseline for assessing whether a new arrangement such as GBS is procompetitive? That question is of general interest to the intersection of antitrust and innovation policy and given the importance of both to the health of the economy, it is critical that we get the baseline question right.

We will be misled if we simply track expansions in output. Clever cartelists will want to cartelize new industries in their infancy, as they know that a new product innovation will inevitably raise output, even if it does so by much less than we would see in the face of full competition. And innovators will want to bundle anticompetitive features with competitive ones if they know that they are simply being judged against the pre-innovation baseline.

Third, as applied to the Google Book Search settlement itself, antitrust enforcers need to disentangle the genuine benefits of the project from anticompetitive features. Obviously, that is a conventional problem in antitrust but it means here that product innovation can’t be used as a general shield against standard antitrust analysis. A single infrastructure such as the digitized book scans can be used to offer many products simultaneously and competitive benefits from one product cannot insulate anticompetitive steps in a second product using that same infrastructure.

Google, Google Book Search, Google Settlement, Authors Guild, antitrust

Abstract:
In this essay prepared in celebration of Judge Frank Easterbrook’s 25th year on the bench, I focus on what copyright students learn from him. Three of his dozen or so copyright opinions turn up repeatedly in copyright casebooks: Nash v. CBS, Inc.; Lee v. A.R.T. Co.; and ProCD, Inc. v. Zeidenberg. This is a surprising success rate for a judge from the copyright-starved 7th Circuit. Judge Easterbrook has an eye for fundamental questions, writes opinions that are brief while treating issues fully and has a distinctively lively Easterbrookian style, one that he preserves by refusing to outsource his opinions to his clerks.

Nash poses a key conceptual question: if only one person believes something to be a fact, is it a copyright fact? We confront the idiosyncratic fact, that is a claim of fact that may be believed by only one person and by no one else. Nash is casebook-worthy alone because of the factual situation it encompasses, as it is the law-school hypo come alive. The opinion nails down a key conceptual boundary question for copyright: copyright facts and actual facts may have little to do with each other.

Lee answers the age-old question: what does glue do? Annie Lee created postcards of her original art. A.R.T. Co. glued postcards to tiles and sold them. In doing so, does A.R.T. violate Lee’s exclusive right to make derivative works as set forth in Section 106(2)? Lee is a refreshingly brief opinion, little more than five columns in F3d, yet, like Nash, it poses in simple fashion a basic question about the operation of copyright. Boundary cases are particularly important because legal analysis frequently builds off of what is taken as given: if x is right, then y must follow. Lee does exactly that for derivative works, an area of increasing importance for copyright.

Finally. ProCD is one of Easterbrook’s best-known decisions, studied by contract students and copyright students alike. ProCD is the opinion that the copyright casebooks love to hate. Easterbrook validates the contract that limits the subsequent use of the ProCD database and wrestles with the tricky question of the interaction between copyright and contract.

Student learn to pay close attention to the text of the copyright statute and to appreciate how that text operates in critical boundary settings. The opinions are written with a distinctive élan, with a little bit of law and economics thrown in, though less than you might expect given Frank’s deep academic roots. Students should understand that the business of deciding cases is a different one than of engaging in an abstract academic inquiry. Easterbrook on copyright is somehow a work of interest, fun and yet discipline all at the same time.

copyright, derivative works, price discrimination

Abstract:
This is a review of the book Invisible Engines: How Software Platforms Drive Innovation and Transform Industries by Evans, Hagiu & Schmalensee.

What makes the PlayStation 3 tick? The Apple iPod? Your BlackBerry? Software, or, more precisely - and much more interestingly - a software platform makes the hardware sing and sits in the middle of a business ecosystem of users, hardware makers and software developers. An invisible engine.

The book centers on software platforms, one example of a two-sided market. These are old markets - newspapers, for example - but many new markets are organized around these software platforms. The core question in two-sided markets is open or closed? That turns importantly (but not necessarily decisively) on the pricing approach of the platform. If the platform is sold at a substantial loss, money has to be made somewhere. It is hard to sell the platform at a loss and open it fully for third-parties. The platform either needs to be bundled with something else - cell phones sold below cost bundled with service plans - or the platform needs to be locked and participants need to be charged for unlocking it.

Multi-sided markets are rich places, and we need to master new rules of the road to navigate there. Our simple one-market understandings will not map easily to this new, richer space. Invisible Engines sets all of this is out in a comprehensive and interesting way. If you are ready to jump in to better understand these markets, Invisible Engines is a very good place to start.