Select 'Print' in your browser menu to print this document.

March 1, 2003

11:15-12:00
Anti-circumvention regulations in the US and elsewhere
Mark Lemley, Boalt Hall (moderator)
Graeme Dinwoodie, Chicago Kent LS (paper on European
implementation of anti-circumvention rules)
Bernt Hugenholtz, Univ. of Amsterdam, Information Law Institute
Tony Reese, Univ. of Texas Law School (paper on influence of anti-
circumvention rules on DRM choices)

11:15-12:00
Anti-circumvention Regulations in the US and elsewhere
Mark Lemley, Boalt Hall (moderator)
Graeme Dinwoodie, Chicago Kent Law School (paper on European implementation of anti-circumvention rules)
Bernt Hugenholtz, Univ. of Amsterdam, Information Law Institute
Tony Reese, U Texas Law School (paper on influence of anti-circumvention rules on DRM choices)

Mark Lemley

Our final panel of the day continues what appears to be today's theme of anti-circumvention regulations and their relationship to Digital Rights Management. However, we are doing something shocking and novel for a conference based in the United States which is that we will broaden the focus beyond the United States and actually talk about regulations and potential regulations in other countries. To lead us off, Professor Tony Reese of the University of Texas School of Law - and now counsel to the law firm of Morrison & Foerster - will talk about anti-circumvention regulation in the United States and their relationship to Digital Rights Management. He will be followed by Professor Graeme Dinwoodie of Chicago Kent School of Law and Professor Bernt Hugenholtz of the Institute for Information Law at the University of Amsterdam, both of whom will talk about anti-circumvention regulations in European countries.

Tony Reese

Thank you, Mark. My time is brief, but I would be remiss if I didn't take part of it to thank Pam Samuelson for inviting me, and to thank Pam and her colleagues and her students and the other co-sponsors for putting on this truly terrific and illuminated event.

I am going to be talking about a paper that I wrote that is available on the conference site, so since I am squeezing it into twelve minutes, feel free to look later for more details on the paper. The paper is raising the question of what incentives of legal protections for DRM systems that are in copyright law today offer copyright owners in choosing which DRM systems to deploy. Now, I realize that legal considerations may be a fairly minor consideration in a choice for a copyright owner of which technological protection measure to use. They are also going to be looking at how efficient and effective and costly those systems are, but I'm a lawyer, so I look at the leading regime and I think that given the fact that copyright owners fought very aggressively for the anti-circumvention provisions, and as we saw earlier this morning, speak very strongly about them. It is reasonable to think that they will play some role in the choice of which DRM systems get deployed.

So, to remind you briefly of what Pam told us all at the tutorial on Thursday, the law in the United States protects two different kinds of technical control measures: access controls, measures that control access to a work, and what I call rights controls, measures that protect the rights of the copyright owner, the right to reproduce, distribute, publicly perform, publicly display, and so on. But these two different kinds of controls get different legal treatment under the statute, and access controls seem to get more protection than rights controls. Now, we're going to talk about how that is. The clearest way that that is true, is that access controls are protected against acts of circumvention, and rights controls are not.
Let me give you an example. The paradigm example of an access control might be something like the Divx system, where you buy a disc and the system allows you to access it only on a certain machine and only for twenty-four hours, unless you contact the copyright owner and pay an additional fee. If you hacked around that, if Divx were still in existence, and you hacked around the protection in order to watch the film after your twenty-four hours had expired, that would be a paradigm instance, it looks like, of circumventing an access control. And that act is illegal under Chapter 12 of Copyright law.

Rights controls, on the other hand, are not protected against acts of circumvention and let me give you an example of how that might work. Imagine that I have a DVD player in my laptop computer, which allows me to play the DVD, see the film on the screen attached to my laptop, but doesn't allow me to send the output signal of the DVD to an LCD projector so that I can show it on a screen. That looks like a rights control. It's not controlling my access to the film, but it's limiting my ability to exercise one of the rights of a copyright owner, the right to publicly perform the film. If I were to circumvent that control, assuming that I had any technological ability, which I don't, in order to send the signal to the LCD projector to show the film on a screen, I would be committing an act of circumvention of a rights control, but not violating the law in any way with respect to the anti-circumvention provision, because acts of circumventing rights controls are not illegal. And if I did that in order to send the signal to a screen in front of my class in Copyright Law at the University of Texas, I will also not be committing any act of copyright infringement, because the Copyright Act expressly allows teachers to publicly perform movies in their classrooms. So, that kind of act of circumvention would not be illegal as a circumventing act and there would not be any liability for it as copyright infringement. So, the Act ban applies only to access controls, they give protect against acts, rights controls don't. That offers some incremental protection to copyright owners if they use access controls, that's not available if they use rights controls.

The paper also makes the point that even with respect to the ban on circumvention devices, which applies equally to access controls and rights controls, it is illegal to traffic in circumvention devices when they circumvent either access controls or rights controls. The paper suggests that there might be some uncertainty about the breadth of protection for rights controls against circumvention devices. The argument is rather complicated and I don't have time to go into it here, but I am happy to talk about with any questions and it is in the paper.
But for those two reasons, access controls seem to get stronger legal protection in Chapter 12 of the Copyright Law, and that suggests that a copyright owner who wants to maximize the legal protection for a Digital Rights Management system would be well advised to adopt access controls to get this additional protection.

Now, it turns out, if we look at least at the case law, cases that have been brought under the DMCA anti-circumvention rules, copyright owners may not have to choose between these two types of legal protection, the higher protection for access control and less protection for rights control. And they may not need to choose, because they may be able to merge access and rights controls into a single system. They may be able to protect against unauthorized exercise of the copyright owner's rights by using an access control. One example for this is something we are all familiar with, and that's CSS, the Content Scramble System used to protect DVDs. So, the Content Scramble System allows a DVD to only be played on a compliant player. You have to have a DVD player that have those standards and have the keys in order to play, in order to access, the film on DVD. That's an access control. But why did the industry want you to play your DVD only on a compliant DVD player? Because compliant DVD players allow you to watch the movie, but not to copy the movie. That's essentially a rights control. You can see the movie, but you can't copy it. So the system, and the statements in the cases, indicates that the copyright owners' real interest was in preventing copying, controlling rights, but they did it by controlling access. And that is likely to be true, given with what we heard Thursday and Yesterday, for lots of DRM systems, that is, trusted systems that rely on only allowing a trusted device to interact with the data or the content. And indeed we spent a lot of time on Thursday, with Brian LaMacchia talking to us about how we are going to determine whether a particular piece of software or a particular device is trusted or compliant.

What's the legal effect of merging access controls and rights controls? Well, so far, the legal effect has been that courts will treat a merged system as both an access control and a rights control, and therefore, entitled to both protections, that is, the stronger protection for access controls against even acts of circumvention. And indeed, in the DCSS case, the decision of the district court was premised mostly on holding DCSS illegal as an access control circumvention device. So, by deploying a merged control, the copyright owner gets the benefit of the stronger protection of the access control.

So what? Why do we care if copyright owners undermine this distinction that the statute draws between access controls and rights controls? Knowing whether we care depends on why we treat these things differently. Why do rights controls get less protection and access controls get more? And the legislative history of the DMCA to the extent it is penetrable at all, it's actually replete with statements, including some of the registers of copyright here in the audience. We weren't going to ban the act of circumventing a rights control, because that would allow the public to make non-infringing uses of protected works. We wouldn't make it illegal to circumvent a rights control. If you circumvented a rights control, what you did would be illegal only if it violated Copyright Law, if it was infringing on its own. So, you could circumvent the rights control, you wouldn't face any liability under the circumvention rules, and if what you did subsequently was not infringing, you wouldn't face any liability under Copyright Law. And so Congress said, we're treating rights control differently, giving them less protection in order to give breathing space for non-infringing uses of protected works.

Merging access controls and rights controls, and treating them as entitled to the protections of both, undermines this distinction. Users will have trouble circumventing the rights control, a perfectly legitimate act, without also circumventing the access control, and circumventing the access control, is of course, illegal. So the effect of treating these merged systems as entitled to both sets of protection is in effect to suck all of the oxygen out of this breathing space that Congress tried to allow for non-infringing uses.

So, in the couple minutes I have left, how might we respond to this problem? Well, one response would be to do nothing. We could simply acknowledge that although Congress decided to treat these systems differently, it wasn't really a meaningful distinction. Why might it not be a meaningful distinction? Well, it could be that all the action is really in the device bans, and indeed Fritz Attaway said this morning, what really matters to us in the DMCA is stopping trafficking of devices. We realize that we cannot police these acts of circumvention. So, maybe Congress' intent to leave some breathing space for non-infringing uses by allowing the circumvention of rights controls was just lip-service to get the thing passed, but all of the action is really in the device bans. I think that it is a little too early to give up on this.

And I am going to credit Congressman Lofgren's suggestion that Congress was in fact not just paying lip-service to non-infringing uses. So, the suggestion is that instead of simply ignoring this, what we might do is say that acts of circumvention of an access control measure are not prohibited under the DMCA, as long as the person who circumvents the access control measure doesn't commit copyright infringement, essentially the same thing we have done with rights controls. You are free to circumvent them, and any liability that you face would be liability for copyright infringement, not for circumvention. Now, this wouldn't deal with some concerns that people have about the DMCA, which is that the device bans are really what matter, and those of us like me who can't engage in technological savvy circumvention, won't be able to do this even though it's legal, but it's at least a place to start to try and preserve the breathing space that Congress said it wanted to allow for circumvention of non-infringing purposes. The paper suggests that the Library of Congress might be able to do this as part of its rule making proceeding, so that might be one way to do this, although the paper suggests there are some difficulties given the standards that the Library established last time around. So, it might have to come in the form of congressional action. Indeed, it would certainly be encompassed within both the Boucher and the Lofgren bills although those exceptions are broader, they would encompass this, but they would be broader, so this might be a more narrow exemption that might be able to find some common ground.

My time is up. I will be happy to talk about this in questions or afterwards. Thanks.

Graeme Dinwoodie

Thank you Mark. When they organized this conference and put together a programming session on the implementation of the EU copyright directive, it would seem likely to be very timely, because the deadline for implementation of the directive was December 22 of last year. And you would think that therefore we would have 15 or 25 national laws that Bernt and I can actually talk about, the fact is that 2 months after that deadline, only Denmark and Greece have actually enacted laws implementing the directive. While this isn't an unusually tardy schedule, because the European Union member states don't always enact the directives on time, it does make what we are talking about a little bit more contingent, but a good part about that is that to the extent you don't like the shape that you see the proposals currently take, there is still an opportunity to influence the content in the member states in the way that they implement their laws.

The downside, if that's what they're after, of course is that the member states are under significant constraints, only European Union legislation itself can incorporate a directive about the choices that they can make about DRM regulations, but the directive made a lot of those choices at the European Union level and took away the autonomy from the member states. That hasn't prevented, at least in the proposals that are out there at the moment, some member states from trying some creative things, such as defining technological protection measures in ways that might exclude certain measures like region cloning and things like that, or adding alternative mechanisms, such as the German proposal does, that would require any product that has a technological protection measure on it to be labeled in a way to allow consumers to know about the content of those measures. And therefore, incurring the market mechanisms that have been raised by a variety of persons over the course of the last three days to determine essentially whether DRM is actually something that will work as a business model. The directive does however constrain a lot of the choices, and Pam mentioned some of these in the tutorial on Thursday, and I don't want to spend a lot of time on them. Although I should say that like the DMCA, the European Copyright Directive is aimed both at acts and devices. Unlike the DMCA, it does not distinguish in the way that Tony talked about between access control and rights control measures, both are treated exactly the same. In fact, based upon the recognition that in fact these were likely to at some point to work in combination. One important difference from the DMCA is that there are no exemptions in the directive itself to the anti-circumvention prohibitions. That is not to say that the legislators were unaware of the potential therefore that a broad prohibition might restrict the ability of the beneficiaries of exemptions under the Copyright Laws that have actually exercised those exemptions. And so they enacted a provision, Article 6.4 of the directive that really gives voice to that concern, or at least attempts to give voice to that concern.

Let me read the first paragraph, there are several paragraphs in this Article, and Bernt at some step might talk about the second paragraph, but the first one reads as follows: Not withstanding the prohibitions against acts of circumvention and circumvention devices, in the absence of voluntary measures taken by right holders, including agreements between right holders and other parties concerned, member states shall take appropriate measures to ensure that right holders make available to the beneficiary of an exception or limitation provided for a national law in accordance with - a whole bunch of Articles that permit exceptions - the means of benefiting from the exception or limitation. My first reaction to reading that was that it must read better in French. Partly, this would have an awful lot more guidance or clarity in any language of the European Union. What I'm going to try and do is focus on that provision which really is the vehicle for the infusion of balance into the directive, and suggest some of the ways in which that might operate, and the member states are beginning to consider operating. Before I do so, let me just mention however, it has at times a really potentially very very broad vehicle, but in fact there are some limits built into it at the European Union level, in particular, it's only going to allow beneficiaries of certain listed exemptions to be able to take advantage of the potential 6.4 procedure. And most exemptions must exist in national law. And the national laws of the European Union are harmonized on exemptions, because the European Union decided to make a list of exemptions optional in the European Union Copyright Directive. So the significant limits on the ability of Article 4 is really to provide that balance, but notwithstanding those limits it continues to operate.

Well, the first question I guess is what triggers the obligation on member states to take appropriate measures of those provisions? It appears to suggest that the need to take appropriate measures is brought into play by the inability of the beneficiaries of certain of those copyright exemptions to take advantage of the exemptions because of technological protection measures. While some European commentators suggest that this trigger is unlikely to start for awhile because most works are also going to be available in unprotected formats, so e-groups will also be available, published on paper. End users will also take advantage of the exemptions traditionally granted to them by Copyright Law. And clearly the scope of the rule was similar to the DMCA rule making authorization adverse effects on a class of works. This argument I think, might be stronger. But Article 6.4 isn't limited in those terms. In particular, it would appear to a lot of us to get down into analysis of whether particular uses of particular works in particular settings actually are in some ways being impacted, because if you look at the exceptions that are in Article 5 that are incorporated by reference into 6.4, they refer to uses and purposes, not simply classes of work that are involved.

Now part perhaps in origin there should be some limit to this. That is to say perceived unavailability of a single work for a single use might be insufficiently substantial a cost to warrant construction of an entire apparatus or mechanism contemplated by 6.4. Which I guess points to some extent the idea that was made this morning, but this depends on what's contemplated by the member states as an appropriate measure in a response to the inability to exercise an exception granted by copyright law. If appropriate measures simply means grant an exemption then there may not be any real cost to acting at a pre-rule triggered by a rule threshold. If appropriate measures means, and this actually is plausible and quite clearly contemplated by the directive, mandating particular technological measures to be used by content providers or establishing a complex quasi-judicial system for determining when such a paramount ability for exercising such an exception exists, then perhaps, then in a prior degree of apparantment should be warranted before I think the answer actually is that it's a little bit of both. That is to say, members should consider the combination of these different devices of ways of assuring that copyright….beneficiaries of copyright redemption can actually take advantage of goods exemptions and that a combined system using exemptions and using a quasi-judicial system not unlike rulemaking might actually be a way of doing it. This, I think, you might say, brings us then to the ultimate question of what are these appropriate measures that states are obliged to take if in fact copyright beneficiaries ..copyright exemptions aren't able to take advantage of. However, there is still a courtroom here…it's a little most complicated. And that is that the obligation of member states to take these appropriate measures only arises in the abstinence of voluntary measures taken by right holders. And the recital emphasized that some reasonable time is to be given to right holders to come up with the voluntary measures. Well...what are the voluntary measures and what is the timing issue there that would implicate member state intervention.

Let me tell you about a voluntary measure that expressly referenced is an agreement that rights holders and other parties concerned. But reaching such an agreement is going to be a very difficult task because the range of stakeholders with interests implicated by copyright law are very bad. And to the extent that the exemptions are not linked to particular categories of persons and purposes of use is difficult to know. Who all was the different persons who potentially should be involved in negotiations and actually are. The likelihood is therefore a complex web of agreements would have to be in place to give full effect to the different copyright exemptions. Now, of course, agreements are not listed as the exclusive voluntary measure, so right holders might consider modification of technological measures in ways that allow the beneficiaries of the exemptions to exercise their exemptions. And I think one of the more interesting things I heard on Thursday, though I think there's a lot of debate within the technological community, is the extent to which technology could in fact be developed in a way that reflects some of the nuances. My sense was that to reflect the nuances of fair use is close to impossible, but someone made the point that in fact to write down in legal language what fair use applies in particular settings is very very hard as well, but to the extent that particular exemptions that are actually much more like European clauses of exemptions, technology might be one of the appropriate measures that would be mandated by member states. It's difficult abstinence for voluntary arrangement also implicate the question of claim so that one pauses on the obligation of member states to promote the voluntary measures we've being talking about but suggest they have a reasonable time to develop those measures.

Well, what is a reasonable time? Some propagators in Europe have suggested that such a reasonable period of time should only start to count, only start the clock going, the moment where technological protection measures are sufficiently widespread and have a negative impact on the beneficiaries of exemptions. On one hand it may not be clear what affect the future effect the measure of life that it had and acting too quickly may be both useless and perhaps premature. On the other hand, there is a danger in building exemptions that will apply respectively simply on past practices and those past practices may change. So I think the important thing for member states to recognize is that there are two different acts involved here that affect claim.

One is the obligation to promote voluntary measures and the second one is member state intervention by way of appropriate measures to remedy any imbalance. The first obligation is immediate. As soon as the directive is effective that obligation exists. Only the second one is actually delayed. Some member states in their proposals have understood that and have set up provisions that they think will help promote the development of voluntary measures. In particular, what several states have done is to establish a procedure that doesn't look unlike the rule-making procedure. In particular, the British proposal suggests that beneficiaries of exemptions should believe that they are not able to take advantage of them can apply to secondary state for a direction to the content providers regarding ways in which the content provider should make the exemption available. Now that direction may simply be that they have to obtain attention technology agreements with other interested parties. Let me just mention a couple of other possibilities in terms of appropriate measures that are suggested by recital and some of the commission officials. Recital 51 mentions the possibility of modification of the implemented technological measures actually being the form of appropriate measures that the state could direct.

This actually interestingly throws the no mandate debate sort of invert in that concept providers might someday be very interested in the notion of a no technological mandate philosophy. The other thing that I did noticed that one of the top commission officials suggested in a workshop this last year was, and I quote, because I'm not quite sure what it means, "handing out the lock and keys." And that's in a translation, so quite what he means I'm not really sure. But certainly the language of right holders making available to the beneficiary suggests affirmative conduct on the part of the beneficiary, more than simply excepting the availability of his statutory exemption, and the extent to which member states are willing to push that affirmative measure as one of the appropriate measures, I think will determine the overall effectiveness of article 6.4 ensuring the beneficiaries are able to take advantage of their exemptions.

Bernt Hugenholtz

I am under specific instruction from the Kingdom of Netherlands, not to make any jokes or other derogatory remarks about your President, so I will be brief. Having said that, I would like to talk about something distinctly un-American, a typical product of old Europe, which are levies. Levies for private copying and the way the levy system in Europe is being affected or will be affected by the DRM. Interestingly, while levies appear to be on their way out in Europe, I will demonstrate that later on, they are becoming sort of hip here in the United States, at least in some academic circles. DRM and levies are from two different worlds. Reconciling them is something like trying to square a circle. In fact, this is what the European Copyright Directive attempts to do. I have tried to illustrate these two worlds on this slide, where DRM are based on individual rights management, and I think the ultimate form thereof, levies are a very crude form of collective rights management.

DRM aims at keeping content exclusive, whereas levies operate under a statutory license. Users of DRM protected content paid for what they get or not get, whereas levies are something of a tax on equipment media, very much unrelated to the value of the actual content, distant to that in fact. Consumers pay users of DRM systems, the media distributors, whereas levies are paid by equipment manufacturers or manufacturers of blank media, these costs then are passed on to the consumers. So, these are very, very different systems. Let me tell you a thing or two about private copying levies as we have them in Europe today. They are not the same everywhere. In fact, in New York, in UK and Ireland, they don't have them at all. There's levies on equipment in a couple of countries in continental Europe, but they have been there for ages on copying machines, faxes, audio recording devices, and VCRs, and more recently, more controversially, there have been levies imposed on scanners and MP3 recordables, CD writers, and possibly even on PCs and hard disks, at least levies have been claimed by the big German collecting society, and that will require some further arbitration as to whether that is really going to happen. On blank media too, you have tapes, tape levies, mini-disc levies, levies on CD recordables, rewritables, and even on recordable DVDs. So there's lots of levies in Europe as we speak today.

It is interesting to have a look at the history of the levy system, a very quick look. Levies were invented in Germany, to be precise, by the German Supreme Court, in the 50s and 60s. About five Supreme Court decisions eventually led to the following, according to the German Supreme Court, taping in those days by tape recorder was considered copyright infringement, a big difference than in the United States. Equipment manufacturers were held liable for contributory infringement, so in principle they could sell the machines, however, a request to identify, a request by the German collecting society, enforce an equipment manufacturer to identify the purchases of the recording equipment was eventually denied by the Supreme Court because that would unnecessarily encroach upon the private sphere of the purchases of the consumers, this is a bit like the Derizor case that we talked about yesterday. In the end, levies came out as a compromise between the exclusive right under copyright and protecting the rights of privacy of the end users. In the end this system invented by the courts was codified, first in Germany, and then in many other countries in the EU.

Now, let's have a look at the EC copyright directive and the way it tries to square this circle. It tries particularly hard with six and seven to promote and protect the deployment of the erring systems, but at the same time it mandates levies for analog and digital equipment in respect media in private copying. That's 5.2(b), one of the many opaque provisions in the directive. In fact, what 5.2(b) says is that in so far that member states allow or permit private copying, they should provide for fair compensation, i.e. levies to the right owners. The amount of the fair compensation however, should take account of the application for and amount of application for technological measures. The idea behind this is that, otherwise you would not take into account the existence of deployment of Digital Rights Managements systems, consumers would end up paying twice, twice to the rights owners. Once, directly under the DRM system, and twice to the indirect scheme of levies they would be paying an extra couple of cents on the equipment for media. So, to avoid such double payment, the directive effectively sets out a track for the gradually phasing out of levies. But what to make of that language in the directive, how to take account of the application or non-application of technological measures. That's one of the many big questions that European lawmakers, actual lawmakers as we speak, are facing today.

Does that mean that national authorities are under the obligation to measure the actual degree of use of Digital Rights Management systems? Does it mean they will have to measure what percentage of content is being protected by technological measures? And if so, what is the base line there? Should it be measured against a world where a hundred percent of content is protected under a DRM system? And how should we measure that? If you think about it, and at our institute we have been thinking about it, we are actually drawing up a report trying to make sense of this provision. You come to the rather easy conclusion that this is not doable. This doesn't work. So, you have to look at other alternatives to give at least some meaning to this idea of phasing out levies as Digital Rights Management systems are gradually being deployed. What we have come up with on the basis of the recitals of the directive is that we should not look at actual degrees of usage, because that is an immeasurable and hopeless task. You should look at the actual availability of technological measures of Digital Rights Management systems in the marketplace. Availability, not simply as a function of technology, not just looking at whether or not the technology actually works and certainly not hoping for foolproof, one-hundred percent uncrackable Digital Rights Management systems, because they will never exist, and they need not exist. But also, looking at economic realities, what are costs for content distributors, what are the costs for consumers, and particularly to what extent will consumers actually accept these technological measures, these systems. Here, I think very important and interesting discussion of yesterday afternoon played the role of what kind of expectancies can consumers have of Digital Rights Management systems.

And finally, last but not least, one should have a look at the legal side of the coin. Are there technological measures available in the marketplace that complies with the law? Do they sufficiently respect privacy in such a way that the original rational of introducing the levy system, which was to protect the user's privacy, can be overridden, can disappear? What this means is a rather complex but interesting process of technology assessment in a very broad sense, another rule making procedure, possibly to be conducted by the European Commission or national authorities or possibly copyright tribunals. And what it also shows is that you could use the impending phase-out of the levy system, as another incentive to introduce Digital Rights Management systems that are socially acceptable. Until that happens however, that may take a couple of years or decades or ages, levies and DRM will exist side by side for the years to come, as will, we all hope, Europe and the United States. Thank you very much.

Mark Lemley

We now have some time for some audience questions, there is a microphone and Eddan is standing in the center aisle with it, so if you have a question, come forward and grab the microphone.

Questioner 1 (Fritz Attaway, MPAA)

A question for Professor Hugenholtz. Jack Valenti is frequently demonized for having suggesting back in the 1980s that levies be placed on VCRs, you may have heard that this morning. And my question is, if this is such an absurd idea, why have so many of the world's great democracies in Europe adopted this ridiculous scheme?

Bernt Hugenholtz

To defend Valenti, this is a trick. I wasn't there when all of this happened. But I don't think if the choice is between prohibition on such equipment and a levy, I don't think the choice is that difficult to make, if the choice is between monitoring end users to a degree, as was said yesterday during the privacy session by Professor Cohen, between conveying the user's privacy and levies, I don't think it's a difficult choice. But as I have demonstrated, and as the Supreme Court decisions in Germany demonstrated, very interesting reading material by the way, it all starts with imposing contributory liability. If that had not happened, we wouldn't have left Europe.

Mark Lemley

Let me just add one thing to that. You can look at a levy system in one of two ways right. You might look at a levy system as a replacement for a copyright infringement system, that is, you can be sued for copyright infringement, but if instead you pay a tax that justifies the cost of the copying, then you can no longer be sued for copyright infringement, then your use is justified. Alternatively, you can look at a levy as a supplement to an existing system of copyright infringement in which you say you will pay a tax and you will still be prohibited from making copies that violate the copyright law under the tax, and it seems to me that how one feels about levies might differ greatly depending on what system a particular individual has in mind.

Question 2 (Don Whiteside, Intel Corp.)

This is for Bernt. I am curious what your view is of the efficiency of the levies infrastructure, the objective clearly is to provide compensation, and fair compensation back to the rights holder. And I am curious what your perspective is of how the money is collected and whether the rates are done through a formal process and how much of those funds actually make it back to the rights holders.

Bernt Hugenholtz

The kind of question that actually masks the comment. There's very little transparency in the way that things operate in most of Europe. The tariffs are more or less haphazardly set, often in negotiations between equipment manufacturers and sometimes dictated increasingly by government authorities, sometimes even in form of law increase. But the transparency becomes less and less the more you look at the output side of the system. The way levies are being redistributed, repartitioned as we say, by the large collective societies is murky to say the least, it is a very very in transparent process. In the research that we have conducted, which will soon be published on our website, probably in two weeks, we have also compare existing levy tariffs from one country to the next, and the differences are spectacular. On the order of five-hundred to a thousand percent sometimes for equipment. In some countries you pay only a few euros, those are large dollar cents these days, and in others you pay fifty cents. The difference is spectacular.

Question 3 (Mike Marion, patent attorney with Phillips Electronics)

In discussing levies, we do have a levy system here with the Audio Home Recording Act. And I am one of those people that think that a rather large loophole was left open in that act, simply because it targeted digital audio at the time, and at that point a fairly narrow range of devices that would handle digital audio. I don't think computers were given any thought at all. But in any case, there are levies that apply to digital audio recording devices, and to specific CDR, for example, blanks, designed specifically to work with those devices. Both of these are an outgross of Phillips technology, and Phillips has supported those products from the very beginning. I am curious whether anybody, either on the panel or in the audience, has any feedback as to how well that particular levy scheme has worked in the United Stated. I realize that it has been circumvented quite a bit because of the fact that computers don't require it.

Tony Reese

We actually have some representatives from the Copyright Office who administer that scheme. You might want to talk to them afterwards to see how they view it. But my understanding is that they do publish regular notices of distributions of those funds, so some funds are being paid. I do think it's not quite correct to say that computers were not thought about in the Audio Home Recording Act, they were thought about very carefully, and they were very carefully exempted. Now, it is quite true that the perception may have been that computers were not going to be digital music copying devices, at least not to that magnitude they turned out to be, so the balance was struck that computers ought to be exempt, but that was not an unintentional loophole. The computer manufacturers were there and quite carefully got themselves exempted. So, there are funds that are being paid and distributed. There is not a lot of information out there about how much and who gets it.

Mark Lemley

I think that it is probably fair to say that it is the internet and the distribution of music that was what people did not think of in 1992 when the act was passed.
Alright, please join me in thanking the panel.

Pam Samuelson

In closing, I want to thank you so much as an audience, you've been great. It has been a real thrill for me in having thought about this program and wanting to have a set of speakers that I thought hadn't interacted together before. I think that we have all learned a lot. I have learned a lot and I hope that that is true for you too.

Before closing, I want to thank especially Patrick Riley and Eddan Katz who have done unbelievable service. In addition, the Berkeley Technology Law Journal, especially the Symposium Editor, who has been great. I think we have all worked together. I want to again thank the sponsors, the School of Information Management & Systems, University of Toronto's Innovation Law Center, and BTLJ, and the Berkeley Center folks other than Patrick and me, and of course Mark here. Anyway, we are really very pleased to have had a chance to offer you this content and we are actually planning on putting some digital content up on the web, more than actually is there already. So, in the next week, I invite you, if you are interested to go back to the website, some of the slides that were presented will be on the website. We are trying to do a webcast of some or all of the conference, and so there may be many things available there that will be of use to you and also that you might want to tell other people about, since we think that we have really done a good job in educating people and bringing more light and heat to the subject. Again, thanks very much.