ARCHIVED — Ian Goldberg

COPYRIGHT REFORM PROCESS

SUBMISSIONS RECEIVED REGARDING THE CONSULTATION PAPERS

---

Documents received have been posted in the official language in which they were submitted. All are posted as received by the departments, however all address information has been removed.

---

Submission from Ian Goldberg received on September 14, 2001 via e-mail

Subject: Comments for the Canadian Copyright Reform Process

Comments for the Canadian Copyright Reform Process

Introduction

Copyright in the Information Age is a tricky business. The very properties which allow information so easily to spread to users and consumers, also facilitate the copying of creative works that happen to have been fixed in digital form. It is vital that input on this reform process be solicited from all of the stakeholders, which include not only the members of the copyright industries which produce the creative works we enjoy, but also the members of the Canadian public, who consume them. The open nature of the current process is inclusive, and commendable.

This note is intended to provide some comments on various portions of three documents:

• A Framework for Copyright Reform
• Consultation Paper on the Application of the Copyright Act's Compulsory Retransmission Licence to the Internet
• Consultation Paper on Digital Copyright Issues

Quotations from these documents will appear as follows:

The departments acknowledge the concerns of these copyright stakeholders, but must consider these concerns within the framework of Canadian copyright law, where certain uses of works and limitations on copyright protection are recognized as serving legitimate and important public policy objectives. Such limitations are evidenced by the finite term of copyright protection, the fair dealing provisions and the exception provisions. These elements of our copyright law have been the outcome of extensive debate, consultation, jurisprudence and legal obligation, both domestically and internationally. Any attempt to affect that balance may require a reconsideration of the current extent of the exceptions provisions.

Comments on A Framework for Copyright Reform

This document outlines some issues that may be addressed in the current and upcoming copyright reform process. I will only comment on two of them, other than the ones discussed in more detail in the two Consultation Papers.

Folklore

Most folklore, as is mentioned in the Framework, entered into the public domain long ago. To use a phrase from Computer Science, this isn't a bug, but rather a feature. The public domain consists largely of creative materials whose time-limited monopoly on certain rights has expired, and it forms a rich cultural heritage on top of which we build new works, and tell new stories, which will enrich the lives of all Canadians, and which will, presumably, eventually themselves enter into the public domain.

The Government is committed to ensuring that copyright law promotes both the creation and the dissemination of works.

The public domain is exactly where works should end up, in order to achieve the widest dissemination possible, and to reach, and touch, the greatest number of people. One of the primary purposes of copyright is to grant economic incentives to artists, authors, and others, in order that those works actually get created in the first place.

For works such as folklore that already exist, and are already in the public domain, I would say that we have already achieved the desired goal, and no further action need be taken. The creators of folklore likely did not get to take advantage of the economic incentives that today would be afforded them under the Copyright Act, but as the Framework states, ``The creators of folklore see their work as gifts that no longer belong to their creators once they are shared with the community.'' The creators gave their work as a gift to the community (as, must be said, do many creators today, who explicitly place their works into the public domain), and we are all now enriched by the broad cultural heritage of the peoples that make up Canada.

Term of Protection

Copyright lasts only for a defined period of time. In Canada the term of copyright, in most cases, is set at the lifetime of the author plus fifty years after the author's death. The issue is whether or not the term of protection ought to be extended to life plus seventy years, as has been done in the United States and the European Union. If an amendment of this type is considered, it would also be appropriate to consider whether it would be granted retrospectively or only to authors who died after the amendment came into force.

The objectives of the Copyright Act, as stated in the Framework, are twofold: the Economic Policy Objective aims to ``reward the creation and dissemination of knowledge and cultural content, and facilitate access to this knowledge and content,'' while the Cultural Policy Objective is to ``ensure appropriate access for all Canadians to works that enhance the cultural experience and enrich the Canadian social fabric.''

As mentioned above, the Cultural Policy Objective can be most easily met by providing that such cultural works enter the public domain as soon as possible, while at the same time not discouraging their initial creation. The Economic Policy Objective at the same time provides direct financial rewards for the copyright holder, only for as long as the copyright is valid.

I will first address the Economic Policy Objective. I would argue (though, admittedly, with no statistics at hand), that any work which produces a significant income to the copyright holder's estate in the two-decade period between fifty and seventy years after his death, must also have been amply rewarding for the half-century prior, if not during the creator's own lifetime. Many works, on the other hand, will have stopped receiving copyright-related income well before fifty years after the death of the creator. I do not believe that in either case the extra income provided to the creator's estate justifies the twenty extra years the work is kept out of the public domain.

Addressing the Cultural Policy Objective is simpler. The issue is whether authors or artists are more likely to create works if they know that they will be protected by copyright for seventy years after their death, as opposed to fifty. I honestly don't believe that that distinction is a pressing one on their minds. No more works will be created because of the extended term of protection, and all works will have their best form of dissemination, being in the public domain, delayed by twenty years.

As to whether the term extension should be granted retrospectively, it seems unnecessary to add to the reward of an estate, rather than an author or artist, and also somewhat unfair to the Canadian public to change the rules of the game part way through; the creators of works knew what the copyright deal was at the time they created their works, and they agreed to play by those rules. The public was also promised that that work would enter the public domain, in order to be built upon, or just freely enjoyed, at a certain time, and that time should not be changed once the work is made. In addition, of course, it borders on nonsensical to incentivise to create new works authors and artists who died between fifty and seventy years ago. Whatever the term of copyright is when a work is created, should be its immutable duration. An author can always voluntarily truncate the term of copyright, of course, by releasing it to the public domain, but even in the event that the Copyright Act is changed to shorten the duration of copyright, the new term should only apply to works created after the amendment takes effect (and not even to all works whose authors are still alive at that time).

I believe that a copyright term of fifty years, let alone life plus fifty years, is more than a sufficiently long time for authors and artists to get rewarded for their efforts, as well as being more than enough time that authors and artists will not be dissuaded from creating new works. We should not be arguing how much longer to make copyright protection, but rather, how much shorter.

In addition, the whole issue of copyright term depending on the life of the author is troubling. As we see advances in life extension and other biotechnologies, which will surely only increase, this will unexpectedly bring along with it an automatic (and retroactive) corresponding increase to the copyright terms of all works. This doesn't seem like it's in the public interest. When this issue is brought up for consultation, I would strongly recommend removing the dependency on the life of the author from the equation. The public should know, at the time of creation of the work, when they can expect the work to enter the public domain.

Comments on Consultation Paper on the Application of the Copyright Act's Compulsory Retransmission Licence to the Internet

At issue is whether retransmission of broadcast television and radio content over the Internet should be allowed under the Compulsory Licence, and if so, under what conditions. Let me start off by saying that, as a consumer, I really enjoyed iCraveTV while it was online. Having grown up in Toronto, but now living in Montreal, iCraveTV allowed me to watch local Toronto television stations; in particular, CityTV. At the time it made no sense to me that a broadcaster would not want to reach more people with their brand and advertising, without it costing them a cent.

Section 2.8.1. of this Consultation Paper reveals that, in fact, broadcasters really don't want to reach more people, since broadcast rights are location-based, and a movie might appear on television in Canada before it would in Europe, for example. This leads me to ask the same question as in section 1 of the Consultation Paper on Digital Copyright Issues:

Do the challenges to copyright truly represent challenges to core copyright principles or are they primarily challenges to existing business and distribution models?

It is unclear to me that the objectives of the Copyright Act would be troubled by refusing to allow rights holders to be able to enforce discriminatory pricing based on location. The copyright holder certainly may control when his work gets disseminated to the public, but the retransmission exception in section 31 of the Act should then allow that publicly-available signal to reach anyone who wants it.

So I am in principle opposed even to restrictions based on receiver location, but am well aware that the opposing interests here are strong, and will be unwilling to change their business and distribution models to accommodate the globality of the Internet. I certainly agree with the departments' view in section 2.8.4 that if the Licence were to not apply to the Internet, it would be unlikely that an Internet-based retransmitter would be able to negotiate all of the required individual rights, and this would ``risk inappropriately limiting the ability of the Canadian broadcasting distribution system to ensure through significant ongoing investment that it continues to employ the most effective technologies available in a time of rapid technological change.''

Allowing a technology-neutral licence, subject to territorial restrictions if necessary, is certainly a better solution, more in line with the principles stated in section 3.1, than an outright ban on Licences for Internet-based retransmitters. In particular, the first point, including ``helping to ensure that all Canadians, no matter where they live, continue to have appropriate, shared access to diverse cultural content provided by Canada's broadcasting system,'' is very well-attuned to the nature of Internet retransmitters.

As mentioned above, I understand that the practicalities of the situation support the departments' view of section 3.2 that ``if Internet-based retransmission were to have the benefit of the Licence it likely must, at a minimum, be subject to an appropriate territorial restriction.'' Therefore, we turn our attention to section 5 of the Consultation Paper.

Scope of the Licence

I support the notion that the Licence be technology-neutral, and that any simple retransmission, subject to the territorial restrictions, be covered by it. The principle of Technological Neutrality from section 3.1 suggests that that should be the case. (This is all assuming that the more general provisions of the section 31 exception to the Act would no longer apply.)

As to the question of whether over-the-air broadcasters should be excluded from the Licence, and how to distinguish them from the retransmitters in rural and remote Canada, I would suggest that the Compulsory Retransmission Licence only apply to retransmitters that simply extend the range of content by retransmission, but do not have the opportunity to select the content to retransmit. In this scenario, a retransmitter that enabled the reception of CityTV in remote Canada would qualify, but one which picked the ``best of'' whatever was on in Toronto at the time would not.

A Territorial Restriction

In section 5.2, the departments seek comment on which classes of retransmitters should fall under the territorial restriction. I agree with their analysis that differentiating between classes is problematic. The issue of DTH satellite systems is not really difficult, as those systems already make best efforts to deliver content only to specified persons, and not even to the public at large. Occasional illicit reception outside of Canada should legally be treated the same way, regardless of the technology used. As to border-area multipoint wireless systems, the amount of spill-over is likely to be minimal, and some leeway may be allowed in the territorial restriction, though again, the same leeway should apply regardless of the technology used.

Note that this leeway is distinct from the threshold considered in part 3 of the proposed restriction; the above leeway concerns the amount of spill-over that is part of the system as designed, whereas the proposed restriction is concerned with unanticipated, and largely unreported and unknowable, circumventions of the technological measures.

As to the latter threshold, it makes sense to use a court-ajudicated ``commercial scale'' standard rather than a numerical one, since, as the departments mention, quantitative monitoring of illicit access is likely to be difficult, if not impossible, to measure accurately. While this more vague threshold adds uncertainly, factors such as whether circumvention devices are in fact being advertised and sold commercially could be used to remove some of the question.

As to the remedies, I am in agreement with the view ``that non-compliance with a territorial restriction should not render the entire retransmission an infringement of copyright, subject to the full range of existing remedies for direct infringement, but rather available remedies must be related specifically to non-compliant foreign reception.'' If most of the people receiving the retransmission are in Canada, a few foreigners somehow managing to enjoy the content should not cause the majority of the users to lose the retransmission service.

Banner Advertisements

The issue at hand is as follows:

Whether the Licence should be amended to expressly provide that (1) except as otherwise required or permitted by law, a signal must be retransmitted without additions to the content thereof, and (2) without the consent of the affected broadcaster, no information of a commercial nature transmitted by the retransmitter, or a person acting in association with the retransmitter, is to be rendered perceivable by a recipient's broadcasting receiving apparatus at the same time as is the content of the signal, unless a result of the recipient's own actions which are not necessary to render the content of the signal itself perceivable.

I disagree with (1) in its obvious interpretation, primarily because it needlessly legislates away otherwise valid and useful behaviours. For example, the retransmitted signal might be accompanied by other original content, such as commentary, or even interactive systems such as chat rooms. As a consumer, I agree with the sentiment of (2); i.e. that if I so choose, I must be able to see just the retransmitted signal, but I can also optionally access the additional content, which might be value-added services that are bundled with advertising, for example. In fact, I would go farther than (2), and say something more along the lines of, ``No information additional to the signal is to be rendered perceivable by a recipient's broadcasting receiving apparatus at the same time as is the content of the signal, unless a result of the recipient's own actions which are not necessary to render the content of the signal itself perceivable.'' Note that I don't think the original broadcaster ought to be able to force the user to view material added by the retransmitter. If the original broadcaster wants the user to be forced to view certain material, he can add it to the original signal.

Unauthorized Retransmission by Authorized Users

Retransmitters must never be held responsible for the actions of the people who receive their transmissions. There are many reasons for this. The retransmitter has no way of knowing whether the retransmission by the authorized user is in fact unauthorized, and even if so, is contrary to the Copyright Act. For example, I would opine that if I go on vacation outside of Canada, and set up some device at my Canadian residence that would send a signal received from a retransmitter (be it a rural retransmitter, the Internet, or even ordinary BDUs) to my foreign location, that ``space-shifting'' would be as legal as making a videotape for the purposes of ``time-shifting''; i.e. personal use. Alternately, the user may enjoy one of the exceptions listed in the Act.

In the event the authorized user was retransmitting the signal to the public, then that user is quite simply violating the Copyright Act, and if done commercially, would be criminally liable under section 42 of the Act.

We certainly do not want it to be the case that my local CBC (over-the-air) retransmitter is liable if my neighbour streams its signal to his friends, or even to the public at large, in the US.

Subsidiary Signals

It seems to me that the determination of whether the subsidiary signal must be retransmitted is related to whether the content of that signal is part of the ``literary, dramatic, musical or artistic work'' (Copyright Act, 31.(1)) which is contained in the primary signal. If so, then the secondary signal should be considered part of the same signal, and must be retransmitted, according to 31.(2)(c). If it is not part of the same work, then it could be considered a separate signal, and not mandated for retransmission. It seems unclear that this is a matter for the CRTC to decide.

Simultaneity and Entirety

Defining ``simultaneity'' to allow for delays due to the retransmission process, and defining ``entirety'' to allow for some loss of information due to the retransmission method, is obviously reasonable. No retransmitter today truly transmits simultaneously, or retransmits the exact signal originally broadcast; small delays and analog or digital errors are certain to occur. Such a stipulation may as well be written into the Licence.

Comments on Consultation Paper on Digital Copyright Issues

This Consultation Paper addresses a number of vital issues that it is important that Canada gets right. Similar issues were decided badly in the United States, and we are only now starting to see the repercussions, including chilled academic research, and jailed programmers. Of the four issues brought up in this document, I will comment on three: legal protection for technological measures, legal protection of rights management information, and liability of network intermediaries.

Legal Protection for Technological Measures

The scenario is this: a content provider uses a technological measure to prevent a user from being able to copy, reproduce, or in fact use, a work in a way other than that desired by the provider. If the user circumvents that measure, whether or not the resulting copy or use is contrary to the Copyright Act, should that very act of circumvention be considered in itself unlawful?

I don't think I can be more clear on this point. No, no, no. The departments seem to realize this, in writing:

The prohibition on circumvention devices and services could have the effect of overriding the traditional contours of copyright protection that emphasize a balance between the rights of creators and the interests of users. More particularly, the prohibition could have the effect of potentially blocking all types of access and use, whether or not they constitute an infringement of copyright.

The thoughts of the departments, in fact, seem to be clear in this area; the entire Proposals section of 4.2 shows a good understanding of the dangers of anticircumvention measures. I am going to make two suggestions that would show Canada to strongly support the rights of Canadians to enjoy, and make use of, their rich cultural heritage.

Rather than criminalizing the circumvention of technological measures, explicitly allow such circumvention in the Act where the purpose is to create a copy or otherwise use the work in a manner not proscribed by the Act. Make it written into law that content providers and rights holders cannot unilaterally change the balance that is copyright. As a concrete example of such a change, the Adobe eBook reader has an option that content providers select, as to whether a particular eBook is prevented from being fed into a voice synthesizer. This option, if selected, would prevent a blind Canadian from exercising his ability under section 32.(1) of the Act to make an audio version of that work.

Going even further in that direction: suppose a content provider does come up with a technology that prevents certain uses of a work, even uses explicitly allowed by the exceptions in the Act. Has not the balance then been swayed? I propose (though I do not claim credit for inventing this proposal) that if a technological measure used to protect a work has the effect of making it impossible through ordinary means to exercise at least the abilities to copy and use the work as described in the Exceptions in the Copyright Act, then that work should not be eligible for legal protection under the Copyright Act at all. This is not such a strange argument; elsewhere in Intellectual Property law, we see a similar choice that has to be made by idea holders: one can either patent an idea, and gain legal protection for it, or keep it a trade secret by technological means. If one chooses to protect the idea as a secret, then if that protection is flawed, the secret is simply revealed, and there is no recourse. On the other hand, if legal protection is sought, then the idea becomes publicly known, and can be used by the public in limited ways outside of the granted patent monopoly, and will eventually rise into the public domain.

On an individual note, I have personally been involved in the mess that is the US DMCA; some of my own work as a cryptographic researcher, as well as that of my colleagues, has come under question as to whether merely publishing an academic paper is a violation of its anticircumvention provisions. Canada has developed a strong cryptographic industry, partially as a result of a more restrictive US legal regime in this area, and this industry, as well as our high quality of research and education, would be directly threatened if DMCA-like provisions were introduced here. I will not live or work in a country that imposes such restrictions on scientific inquiry. We must not allow academic speech to be chilled, stifled, and censored by any person, group of people, or industry.

Legal Protection for Rights Management Information

There are, broadly, two kinds of rights management information: simple data fields that are attached to a work and which indicate things like first author, the holders of certain rights, etc. (or, as the departments correctly point out, the holders of certain rights at some unknown point in the past, as the identities of those holders could change), and data integrated into technological measures, such as watermarking, that will prevent certain uses of the work unless some conditions set by the content provider are met.

As for the first kind of rights management information, I agree with the departments in asking ``whether the integrity of certain information ought to be protected, given that, over time, the information may cease to be accurate.'' There seems to be no sense in giving special legal protection to that kind of data. Even if modification of such data were disallowed, enforcement would amount to seeking legal recourse against the individual information-modifier. But this is no easier than just seeking legal recourse for individual copying! The new capability would seem to gain nothing.

As for rights management information integrated into technological measures, the same arguments apply here as in the section above. If tampering with the rights management information is necessary in order to circumvent the technological measure preventing one from copying or using the work in a way allowed by law, then by all means that rights management information must be allowed to be tampered with.

Liability of Network Intermediaries, such as Internet Service Providers, in Relation to Copyright

The departments' proposal consists of three elements:

• a limitation on liability for copyright infringement;
• a provision for notice and take-down;
• a limitation on the liability of the ISP for complying in good faith with a notice and take-down regime.

Limitation on Liability for Copyright Infringement

I would say that this part of the proposal does not quite go far enough. In general, the role of an ISP is to store, transmit, and forward, bits of information. The general rule that should apply to an ISP is ``bits are bits''; that is, while the ISP is going about its normal roles, shuffling bits around, it should not have to care one whit just what it is those bits might represent.

ISPs who store data on behalf of users should not be forced to police their clients, looking for copies of works of which they must somehow ascertain the legality. If they routinely make backup copies of users' storage spaces, these copies should not cause the ISP to infringe the rights of every copyright holder whose work was contained therein. (It was not clear to me whether such backups, or other day-to-day operations of an ISP with respect to data stored by (as opposed to transmitted by) users, was covered under this element.)

A Provision for Notice and Take-Down

This is certainly the most troublesome element in terms of working out the details. We see a similar regime under the US DMCA, and we have already seen abuses of it. For example, at (http://www.salon.com/tech/log/2001/08/31/dmca_animals/index.html) we read about a British medical research firm using the notice and take-down provisions of the DMCA to force the blocking of the website of an animal-rights group critical of its activities. Under the DMCA, any claim to an ISP that a certain site allegedly infringes copyright, forces the ISP to remove the site from the Internet for at least 10 business days, whether or not the ISP believes the claim is credible.

I would suggest some higher standard for take-down than simply someone's say-so. This could be implemented by having the ISP's limitation on liability, as described above, not disappear simply upon notice. Rather, something more akin to an injunction from a Canadian court (which could likely be obtained under section 34.(1) of the current Copyright Act) would be required to force the ISP to interfere with its customer's bits. That being said, there's nothing to stop the rights holder from contacting the ISP and/or the user prior to that time, simply to make them aware of the situation.

In any event, the mandatory 10 business day take-down minimum period of the DMCA should not be emulated. Once the ISP receives the required notice, and takes down the allegedly infringing data, it must inform the affected user as expeditiously as possible. The user would then have the ability to claim that the notice was in error. At that point, the material should be reinstated immediately, and the rights holder and the user have their lawyers at it, in a straightforward copyright dispute in which the ISP is not involved. If the notice was in fact an injunction, then of course the ISP would not reinstate the material; in fact, both options could be offered to rights holders: if they just use a ``say-so'' notice, then the user might object, and have the material reinstated at once (and risk damages under the Copyright Act; additional damages might be specified for this case); if they use an injunction, the material will stay down until and if a court reverses it.

A Limitation on the Liability of the ISP for Complying in Good Faith with a Notice and Take-down Regime

I agree with this limitation, whether we are talking about a ``say-so'' regime or an injunctive one (though for the latter it's really a non-issue), unless in a ``say-so'' regime the ISP fails in its duty to notify with reasonable promptness the user whose data it interfered with of the notice and take-down (and so the user was denied his opportunity to respond). In that instance, the user, if he prevails against the alleged rights holder, may have grounds for action against the ISP. This is just the other side of the ``good faith'' compliance with the notice. An ISP who takes down a user's data for insufficient reason, or who fails to notify the user of a notice and take-down, would not have his liability limited by this element.

Conclusion

Copyright is a careful balance. The fact that works are easier to copy (both legally and illegally) in the online world does not mean that we should automatically assume everyone an infringer. Rights holders should be looking less towards how to prevent users from making copies that may or may not be infringing, and more towards how to adjust their models in order to thrive in a world where physical distribution is expensive, digital distribution is cheap, and information flows freely between all points on the globe.

Ian Goldberg
(e-mail address removed)
14 September, 2001

---

About the Author: Ian Goldberg is Chief Scientist at a privacy technology company in Montreal. He holds a PhD in Computer Science, and his fields of research include cryptography, security, and privacy. He is, as they say, proud to be a Canadian.

---

Copyright notice: It seems fitting that some mention should be made of the copyright status of these very comments. The original portions are hereby placed into the public domain. Industry Canada and Canadian Heritage are the source institutions for the quoted portions.

Return to list of submissions      

Share this page

To share this page, just select the social network of your choice:

• Delicious
• Facebook
• Twitter

No endorsement of any products or services is expressed or implied.

x Close