Once again, Linking is NOT unauthorized copying
When Access Copyright included “linking” or “hyperlinking” in its definition of “copy” in its tariff application back in 2010, the provision became one of the most notable lightning rods of opposition for the proposed tariff; a paradigm example of over-reaching. After all, Section 3 of the Copyright Act, which sets forward the full and complete list of owners’ exclusive rights does not include an exclusive linking right. And for good reason; the functionality of the Internet is based on the ability to embed links to other works in urls which will point to other websites (yes, even those containing works in which copyright subsists). AUCC, ACCC , CAUT, CFS and dozens of other objectors pointed to this obvious problem with this and other AC requests; and a vigorous opposition to the tariff proposal was underway.
Yet when the University of Toronto and UWO made their bombshell announcement in late January that they had entered into side-agreements with Access Copyright, and that the agreements contained this very same treatment of linking, Access Copyright’s extra-legal demands began to take on a life of their own. Never mind that linking cannot be the basis for infringement under the act, two major educational institutions have just agreed that it does, or at least that it might. So the Western and UT capitulation to Access Copyright had an important effect, not only did it undermine the opposition efforts to the tariff, it further legitimated AC demands that otherwise had no other reasonable basis, much less justification in the law.
And when AUCC, and then as well ACCC, followed the bad example of UT and UWO, the legitimation of this and other unfounded claims about the scope of compensable activities just became further entrenched. What had started out as purely opportunistic and self-serving distortions of copyright law started to become real. Not because of any provision in the Copyright Act, or because of any judicial interpretation, but simply instead because of the acquiescence of powerful institutional actors.
In justification of its ill-conceived “model license” agreement with Access Copyright, a Q&A format memo was prepared by AUCC Counsel and sent to member institutions advising now that the infringement liability status of linking was still an open issue:
5Q: Does the definition of “Copy” in the AUCC model licence mean that AUCC accepts that posting a hyperlink to a digital copy is the same as authorizing the making of a copy and requires a licence?
A: Despite the ruling of the Supreme Court of Canada in a recent defamation case, Crookes v. Newton, it is still an open issue in Canadian law whether posting a hyperlink could make a person liable for authorizing the copying of the digital work. The definition of “Copy” in the model licence makes the licence and the indemnity very broad in scope. Another provision in the model licence clarifies that AUCC has accepted this definition on a“without prejudice” basis and reserves the right to take a different position on the meaning of the term in any other proceeding.
Having been circulated around to AUCC member institutions, this bad advice has re-emerged in the form of local Q&As issued in justification of campus-based decisions to sign on to the AUCC-endorsed agreement. So the McMaster Q&A, said:
Q2. The definition of “Copy” includes hyperlinking to a digital copy. Is this definition granting Access Copyright more rights than the law requires?
A2. Despite the ruling of the Supreme Court of Canada in Crookes v Newton, it is still unclear whether posting a hyperlink could make a person liable for authorizing the copying of a digital work under Canadian copyright law. The broad definition of “Copy” means that more activities will be covered under the licence agreement, which also broadens the scope of legal protection afforded to the McMaster community.
to which Ariel Katz responds:
This is simply false. The Model License defines “posting a link or a hyperlink to a Digital Copy” as an act of copying, not as an act of authorization. There can be no question that a link is not a copy since the link itself is not a reproduction of the work, and therefore posting it does not create an additional copy. Clicking on a link may or may not result in a reproduction of the linked-to work, and usually the person who posted the link will not be deemed to have authorized the reproduction, except, perhaps, in very specific and unusual circumstances. But this is irrelevant to the present case because a person cannot be held liable for authorizing the making of a non-infringing copy. Since the Model License permits the copying of Repertoire Works, an Authorized Person cannot be held liable for “authorizing” the making of such copies, by posting a hyperlink to it or otherwise. If for some reason McMaster is concerned that acts of “authorizing” may not be covered by the license, then it should insist on making it clear that the license explicitly covers the acts of “authorizing”.
And to only add insult to injury, AUCC and McMaster make the further argument that accepting the definition of linking is without prejudice because of language in the model license. McMaster’s Q&A again:
Q3. Wouldn’t McMaster be prejudicing itself in the future by agreeing to such a broaddefinition of “Copy”?
A3. The definition of “Copy” will be accepted by McMaster on a “without prejudice” basis and the licence agreement we sign with Access Copyright will clearly reflect that. McMaster reserves the right to take a different position on the meaning of the term “Copy” in any legal proceedings related to this licence agreement.
To which Prof. Katz responded:
R3. Incorrect. Under the Model License the definitions are binding with respect to proceedings related to this agreement. The “without prejudice” language is limited to proceedings “unrelatedto this agreement”. One possible implication of the definition is that AC would have a basis for demanding a survey that includes studying the linking behaviour on campus, or how many timesworks are “displayed” on computers or other devices.
Similar arguments in response to AUCC’s initial Q&A were made by Howard Knopf.
This notion of linking liability was wrong when Access Copyright first raised it in the tariff application, it was wrong when the University of Toronto and UWO agreed to it, it was wrong when AUCC reversed its earlier position and tried to justify it in their April Q&A, it was wrong when it was then parroted by the McMaster admin, and it will be wrong when you see the same argument repeated on your campus.
Last week’s decision from the Federal Court of Canada in Warman and National Post v Fornier simply reinforces what the many have been arguing all along, linking is not itself a copyright relevant activity and should not be taken into account by the Board in setting a tariff. Linking to a website containing copyright works is not an authorization to reproduce the works and it can not rise to the level of being included in any definition of “copy” or “copying”.
As Michael Geist said earlier today:
The Supreme Court of Canada has already ruled against attributing defamation to such links and now the Federal Court has concluded that links cannot be said to constitute unauthorized copying. The implications once again extend to forums, blogs, and other venues as well as the Access Copyright model licence.
Maybe now the failure of the University of Toronto, UWO and AUCC/ACCC licenses with Access Copyright to properly comprehend the relevance of linking will become a bit clearer. Maybe even enough for Toronto, Western, AUCC and ACCC to accept that they made a mistake.
But at the very least it should be enough to keep further institutions from repeating the mistake.