![[Ad box removed; this image serves to flag pages that need to be updated in my log file.]](/i/boxremoved.png)
Any discussion of copyright must begin and end with freedom of expression. That principle has guided my participation in Canada's copyright reform process, including my lengthy initial written comment, New Media Copyright Extensions Would Harm Canada; my three written response comments (to DirecTV and Lahnakoski, Smith and Henshaw, and Aliant et al.); and my interventions at the Toronto and Ottawa consultation meetings. My previous submissions speak for themselves. In this comment of 28 April 2002 I will highlight only a few points that I think were missed, or not covered as fully as they deserve, in the consultation meetings. Here are the points I will highlight:
The discussion of ISP liability has been pre-empted: while we were talking, Parliament was legislating. In particular, Parliament has passed Bill C-15A, a criminal law reform bill which creates a new offense for "accessing" child pornography. That bill is interesting because it creates what copyright reformers have been calling a "notice and notice" (or "notice, notice, and takedown") regime for child pornography. If there is material on an interactive computer system (for instance, a Web server) that is alleged to be illegal child pornography, then the police must send a notice to the owner of the computer to be forwarded to the poster of the material, then the poster has a chance to show cause in court for why the material should not be removed, then the court may issue an order for takedown, and all of that must happen before the material can be legally forced off the Web. Parliament chose that process apparently in order to respect the due process guarantee of the Canadian Constitution.
Tuesday, 23 April 2002 was a busy day in the House of Commons. For one thing, that day was Canada Book Day. Canada Book Day was chosen to mark several important literary anniversaries - for instance, the birth of Vladimir Nabokov. Several Members of Parliament rose to make statements about the importance of literacy and the freedom to read. That day was also an "alloted day" for the Canadian Alliance; the Alliance chose to use the day to debate a motion to (among other things) further tighten Canada's child pornography laws by banning "materials that appear to depict or describe children engaged in sexual activity" - a motion that I take to include not only the writings of the notorious John Robin Sharpe, but also the Christian Holy Bible, and such classics as Catch-22, Dune, Snow Crash, and World of Wonders (third volume of Robertson Davies' Deptford trilogy). I posted a longer list, and commentary, on my Web site. The motion outraged me, as a book-lover, but that is not why I mention it here: in this comment I wish to point out that even on Canada Book Day, when many views were expressed in Parliament that I would consider insanely extreme, nonetheless nobody ever suggested that it should be possible to destroy written material as "child pornography" without a court order. Any effort to ban books would still have to gain at least an air of legitimacy by going through a court.
So even the extremists agree: child pornography may be an evil thing, but due process of law must still be followed when dealing with it. Also worth noting is that amendments to C-15A, introduced by the Senate and passed by the House after some debate, exempt Internet service providers (ISPs) from almost all liability for the new "access" offense, even if they were aware that their customers might be committing the offense through their facilities. As far as "access" of child pornography is concerned, the message is clear: it is a serious crime, but accused criminals retain their due process rights, and ISPs need not shoulder a ruinously unreasonable liability load.
Compare the new child pornography law with the position of some privilege holder groups in the written comment process and consultation meetings. Some, not all, privilege holder groups claim that "piracy" is so evil, and so immediately dangerous to society, that it must be stopped as quickly as possible at any cost. In particular, they demand a "notice and takedown" regulatory regime, in which someone claiming to be a privilege holder could force the removal of any material from the Web immediately, by filing an ex parte complaint, without going through a court, and leaving the poster of the material assumed guilty and forced to prove his or her own innocence after having already been punished for the crime. Notice and takedown might better be called the Wonderland regulatory regime: sentence first, verdict afterwards! If Parliament would not even consider such a regime for child pornography, how can it be justified for "piracy"? Supporters of "notice and takedown" are claiming that copyright infringement is worse than child pornography. That is unthinkable.
My second point was raised in several slightly different forms by various speakers at the Ottawa consultation meeting. The point is that if privilege holders choose to use technological protection measures (TPMs) on their work, then they should forfeit some or all of the protections of copyright. That may sound like an absurd idea, but it is actually well in keeping with the existing theory of patents. An inventor with a new idea for a device or process may seek patent protection for it; in that case, the inventor gets a legislatively guaranteed monopoly for a limited time, is required to disclose it to the public immediately, and at the expiry of the patent, the idea becomes public domain. Alternatively, the inventor may choose to keep the idea as a trade secret; then he or she must rely on the much weaker legal protection of trade secrets, but there is no expiry, and the inventor may use whatever lawful technological means can be devised to protect the idea for as long as may be possible.
Why shouldn't copyright law work the same way as patent law? If TPMs operate as advertised, to enforce reasonable limitations congruent with those of existing copyright law, then privilege holders should jump at the chance to exchange existing law, with its time limit and the spectre of fair dealing, for TPMs which do not have either flaw. Just as with patents, immediate disclosure, and eventual release to the public domain, should be conditions for copyright protection. A work encumbered by TPMs is not "disclosed" and may never be available in the public domain; thus it is not eligible for legally-protected monopoly status. Privilege holders ought choose one or the other. At the very least, as described in one of my previous comments, works encumbered by TPMs should not be eligible for compensation under the blank media levy.
Extension of the term of copyright is not officially one of the issues in the present consultation, but it has been brought up by some intervenors, particularly by representatives of "content industries" in Toronto (in favour of extension), and independent representative in Ottawa (against). The topic is in everyone's mind partly because of concern over the U.S. "Sonny Bono Copyright Term Extension Act" (CTEA), which extends copyright in many cases from the Berne Convention limit of "life+50", to "life+70" - that is, the copyright lasts until the end of the calendar year 70 years after the original creator died. Some interests are calling for a similar extension here in Canada. No good reason has been suggested for such an extension.
The term of copyright does not need to be extended, and should not be extended. Even the existing limit of "life+50" is arguably too much, especially in the case of computer software. Computer software becomes obsolete so quickly that any term longer than ten years might as well be forever. And it is clear that although they only demand an additional twenty years at this stage, "forever" is really the term of protection the content industries intend to achieve. Copyright in the U.S., originally limited to 28 years from publication by a law modelled on the 1709 copyright statue of Queen Anne, was steadily increased during the 20th Century, and there is no indication that the most recent increase will be the last attempt. Every time the privileges held by the major media conglomerates (most notably the Disney corporation) draw close to expiry, they buy a new extension. Disney did not even attempt to conceal its lobbying for the Sonny Bono law, nor attempt to justify it on any grounds other than its own business interests.
For out of the old fieldes, as men saithe,
Cometh al this new corne fro yere to yere;
And out of old bookes, in good faithe,
Cometh al this new science that men lere.- Geoffrey Chaucer, The Assembly of Fowles
But if there are few good reasons to extend the term of copyright, there are plenty of good reasons not to extend it. As suggested by the above quote, new creation depends upon the availability of older work; copyright, especially when backed by TPMs, hinders the ability of creators to use existing work. For some classes of work, such as computer programs, even the existing limit of life+50 is far too much. For works such as classical music, the existing artistic community depends on the constant stream of works that are elevated to the public domain as copyrights expire. Term extension, especially if retroactive like the Sonny Bono law, threatens the existence of community orchestras (who cannot afford licenses for copyrighted material and depend on public domain work) in order to confer the dubious benefit of encouraging long-dead composers to write more music. Other arguments against term extension are presented by other intervenors and in my other writings; I will close my comments here on this subject by drawing attention to a speech of Thomas Babington Macaulay before the British Parliament in 1841, in which he presented many of the same arguments that opponents of term extension put forward today.
I brought up my next point during the "extra topics" session in Ottawa and some delegates commented on it; I repeat it here because it is important. No liability can attach to persons who post "links" to allegedly illegal material. The concept of linking is familiar to any user of the World Wide Web; documents on the Web can contain the name and location of other documents, in a concise machine-readable form, so that a writer can easily cite a document posted by someone else and a reader can easily "follow" the link to look up the cited document.
If someone writes an illegal book, is it illegal to cite that book by listing its title, author, and publisher? Certainly not. Neither is it illegal for a newspaper to report the address where a crime took place. We can talk about the actions of others, without being participants in those actions. In just the same way, publishing a link to a document on the Web does not create any participation or endorsement relationship between the poster of the link and the poster of the linked document. There should be no liability.
The poster of a link cannot even be responsible for knowing the content of the linked document, because documents can be freely changed. My own Web site at ansuz.sooke.bc.ca contains at last count links to 536 different documents on other Web sites (including 84 documents on various Government of Canada Web sites), as well as numerous internal links among documents on my own site. Any of the external documents could be changed at any time. There is no way I can be constantly aware of the contents of all the documents I link to; if someone should post something illegal on one of those Web sites (one could hope that would not happen on a Government site!), then it is unreasonable to hold me responsible. The Web sites I link to are totally outside my control, and almost totally outside my knowledge.
At the Toronto meeting, David Basskin of the Canadian Music Publisher's Association attempted to allay my fears by drawing a distinction between links like mine, which are mostly to text documents for one-time viewing on a computer screen, and "direct" links to "pirated" media files; the difference being that a "direct" link causes the immediate download of a file to the user's hard disk. (In the meeting he also used the term "deep links" to describe links causing immediate download, but as that term has a very different meaning to most Web users, I avoid it here.) Unfortunately, the distinction between "direct" and other links is not technologically useful. All links cause the immediate download of a file when activated; that is simply how Web browsers work. HTTP, the protocol used by the World Wide Web, is a file transfer protocol. It exists to transfer files, always and only, and downloading a file is a necessary prerequisite to displaying it on a computer screen.
Most Web browsers do not save downloaded files beyond the first viewing unless the user explicitly asks for the file to be saved, but that is a configurable behaviour of the Web browser, controlled by the user. As publisher of a link, I have no control over whether that link will "cause the immediate download" of a file to the user's hard drive for permanent storage. It is true that the behaviour will generally be determined by the type of file - text is not usually saved permanently, while most audio and video files are usually saved permanently. But that behaviour is only the default, for most but not all software, and is fully controlled by the user. A user with appropriate plug-in software modules could easily listen to or watch an audio or video file in the browser without permanently saving it; a user could also permanently and automatically save text files, using the "crawling" features of a browser like Lynx or Internet Explorer.
As publisher of a link, I cannot predict the behaviour even in the default case, because I cannot predict the format of a document I might link to. The format of a document can easily change from one moment to the next. Publishing a link is like saying, "Phone such-and-such phone number and listen to what the person on the other end says." If the phone number is not my own, then I can make no promises about the results of such a call. There is no way I can be responsible for whether my link is or is not "direct".
Just as I cannot control Web sites I may link to, I cannot control Web sites that may link to me. This is not an issue that has been raised in the current round of Canadian copyright debates, but it has been visited by certain U.S. court cases. Some posters of Web sites have claimed that because they own the copyright privileges to their own Web sites, they should be allowed to specify who may or may not publish links to their Web sites. Often this manifests as a situation where the privilege holders welcome links, but only to a special "entry" page; they want to prohibit links to other documents on the site, bypassing the entry page. They apparently want to create a new privilege in the copyright bundle: a privilege of restricting citation.
I suggest that any such privilege would be a big mistake. We do not allow authors of hardcopy books to specify who is or is not allowed to write reviews of those books. To allow such a restriction would destroy the valuable process of intellectual debate, because it would give authors a veto power over critical reviews. Restricting links on the Web would be similarly destructive. Furthermore, any restriction on Web links would become a restriction on all forms of citation, because software exists or can easily be written to translate any standardised citation format into Web links, for documents which are available on the Web. Thus, any formal citation of a work can be seen as a "link" to it if the work is available in any form on the Web; giving privilege holders power over links to their work is giving them power over who may cite their work, and that is far too much power.
Note that as discussed above, no useful distinction can be drawn between "direct" links and other links; the difference is determined entirely by the viewer's configuration and is not under the control of the poster of the source or destination Web site.
Creators of work have a moral right to be identified as such, but they have an equally important right to remain anonymous if they so choose. Anonymous publishing is especially important for works of a socially or politically sensitive nature - the very works that might most likely be targets of suppression efforts such as false infringement claims. In the Toronto meeting there were many comments about it being easy to trace the poster of a document on the Web. It is true that we can easily find someone with the power to remove any given document from the Web, but it is neither easy nor desirable to always identify the original poster of a document. Systems like Freenet are designed to guarantee the anonymity of document posters through technological means, and the "moral rights" of copyright law suggest that guaranteeing anonymity is a worthwhile pursuit. Any talk of regulatory regimes that would require the identification of posters of documents, seems worrisome.
If I post a document on the Web anonymously, and my opponents file a false claim of copyright infringement against it, they should not be able to use that complaint as a means to discover my identity. If the complaint were genuine then we might argue that I have no right to anonymity, but until the complaint has been finally judged valid, we must operate on the assumption that I am innocent, and protect my right to anonymous posting. Protecting anonymity is especially important because if my identity is revealed unjustly, there is no meaningful way to take that back or compensate me for the breach.
Protection of anonymity need not be a problem in a "notice and notice" or even "notice and takedown" regime. The ISP is required to act as a go-between in either case, and it need not reveal the identity of the poster of a document to the person making the complaint. I highlight this point only to prevent any attempt to slip a requirement for revealing identity into the legislation: spurious copyright complaints must not become a tool for violating anonymity. Revealing the identity of an anonymous poster must require a court order. Since material may be found infringing and removed from the Web perfectly well without knowing who posted it, the order to reveal the poster's identity should in fact require additional justification, over and above the legal requirements for a takedown order.
Many Canadian families choose to educate children at home instead of sending them to school. I was a homeschooled student myself. During this consultation process we have heard from many educational groups demanding special privileges and copyright exemptions for educational purposes, and for the most part I am in favour of those proposals. I am concerned, however, that any "educational amendment" to copyright law must be worded in such a way as to include homeschoolers. Otherwise, homeschooled students will be placed at an unfair disadvantage relative to students of more formal educational institutions.
My final point is one suggested by Tom Trottier at the Ottawa consultation meeting; I think it deserves additional emphasis. We have heard, in interventions from the copyright industries, the theory that fair dealing can be adequately served by extremely limited access to copyrighted works. For instance, a movie stored on DVD in a high-quality digital format might be available for the purposes of quoting in a critical review, only by circumventing the TPMs or in an analog format encumbered by watermarks, Macrovision distortion, and deliberately reduced quality. Privilege holders suggest that that - access to the movie at a significantly reduced quality from the level used for ordinary viewing - is good enough for legitimate fair dealing.
Actually, the opposite is true. Academic fair dealing will often require better access to work than that used by ordinary viewers, listeners, or readers. For instance, the discipline of "stylometry" combines literary criticism with statistics, to build profiles of individual authors, examine questionable attributions, and compare works and authors in a precise mathematical way. Researchers wishing to apply stylometric techniques to a novel published in electronic form will generally need to obtain the work in an unencumbered text format, suitable for input to their computerized analysis tools. Ordinary readers, on the other hand, might be satisfied with an encrypted copy of the novel, only readable with a special client that prevented cutting and pasting to unencumbered formats. Even a special "fair dealing" feature in the viewer, to (for instance) allow excerpts up to a limited number of words, would be insufficient; legitimate stylometric research requires access to the whole thing, in an unencumbered format.
Although its roots go far back into the past, stylometry as it is practiced today is a recent development, because it depends on recently-invented computer techniques. By the nature of science, we cannot predict what other new developments lie in the future, and we cannot foresee all the forms of access that researchers may legitimately need towards work that we "protect" today. Barring a regime like that suggested by Chris Friesen, in which all works are deposited in unencumbered format in a central registry, we must permit circumvention for research purposes without limiting fair dealing to deliberately-degraded "analog" or "excerpt-quality" copies of work.
My general views on copyright reform have already been stated many times in written submission and on my Web site. I would refer readers looking for a fast summary to my list of the top 12 copyright reform issues. In this comment I have described several issues that arose from the consultation process to date, guided by the basic principles already set out. Any discussion of copyright must begin and end with freedom of expression.
This form is for posting public comments to be read by other people who visit this Web site. If you have a software support question, or other material directed to the page author instead of to the general public, please send email instead.
All the data you enter, and your IP address, will be saved and displayed. Don't enter secret information. HTML is not accepted; it will be displayed as plain text. Your comment will only be added if you enter valid data in all required fields; if it isn't, use the back button and try again.
I, and I alone, reserve the right to remove postings for any reason.
![[RSS syndication file]](/i/rss.png){kind=small}
No comments yet.