The recent Federal Court of Canada decsion in Warman v Fournier (2012 FC 803 – for reasons which elude me, the decision is not yet available on the Federal Court’s website, but Howard Knopf has generously uploaded a copy) has occasioned a host of thoughtful observations (see Knopf, Trosow, Geist), which I can only summarize and add a few hopefully trenchant elements of my own analysis.
The dispute involved claims of copyright infringement relating to three separate works, one a speech, one an article and the last a photograph. The defendants (who also co-defendants in the Baglow v Smith case, discussed here) operate a website described in the judgment as “an online political news discussion forum”, and each of the works in question was uploaded to the defendant’s website.
The action for infringement of copyright in the speech was time-barred due to the expiry of the limitations period set out in the Copyright Act (Canada), and so need not concern us further, except to note that the plaintiff’s attempt to ground his claim on an instance of infringement which occurred within the limitations period failed for lack of evidence – which the court indicated could have been resolved by provision of “a printout … with [a new] URL listed at the bottom owhich would have been evidence that there was an additional uploaded copy”. So, reminder: print out hard copy evidence of infringements bearing timestamps.
No Infringement Because a “Substantial Part” Was Not Copied
With respect to the purportedly infringing article, a few things are worth remarking on. First, the article was not written by Richard Warman, the plaintiff – instead, it was written by Jonathan Kay, and Warman acquired an exclusive license in respect of the work as a result of the settlement of an earlier defamation claim advanced by Warman against Kay and the National Post (the paper which had originally published the article). I had previously remarked on assignments of copyright as a remedy for defamation (see here) and reiterate now my thought at the time that while this seems like an extremely useful remedy for a plaintiff to obtain, it still strikes me as problematic from a rights management perspective (why should the plaintiff get copyright in an entire work, as opposed to simply the defamatory elements?).
In any event, the court rejected the infringement claim because the amount copied on the defendant’s website did not amount to a “substantial part” of the article. To quote from the judgment (para. 23, all emphasis added):
Whether a substantial part of a work has been reproduced is a question of fact and involves a qualitative rather than quantitative analysis. The relevant factors to be considered include:
a. the quality and quantity of the material taken;
b. the extent to which the respondent’s use adversely affects the applicant’s activities and diminishes the value of the applicant’s copyright;
c. whether the material taken is the proper subject-matter of a copyright;
d. whether the respondent intentionally appropriated the applicant’s work to save time and effort; and
e. whether the material taken is used in the same or a similar fashion as the applicant’s:
U & R Tax Services Ltd v H & R Block Canada Inc,  FCJ No 961, at para 35.
While, because people may be tempted to derive “bright line” tests, it can be dangerous to quote from particular infringement analyses, this one seems worth reproducing (paras. 25-27, all emphasis added):
Quantitatively, the reproduction constitutes less than half of the work. The Kay Work itself consists of a headline and eleven paragraphs. The reproduction on Free Dominion included the headline, three complete paragraphs and part of a fourth. Qualitatively, the portions reproduced are the opening “hook” of the article, and the summary of the facts on which the article was based. Most of the commentary and original thought expressed by the author is not reproduced.
… It does not appear that the excerpts of the Kay Work were reproduced to “save time and effort”. Based on the context of the posting, the respondents reproduced portions of the Kay Work to preserve a record of the facts summarized in the article, so that members of Free Dominion could continue to discuss those facts on the forum. Also, contrary to the applicant’s argument, the reproduction does include a summary or paraphrase of part of the work, specifically the second paragraph.
(I also note, parenthetically, that Warman’s acquisition of the copyright in the article appears to have cut against him in the court’s infringement analysis: at para. 26, the court notes that the plaintiff does not “use” the article in any traditional sense, but rather he acquired copyright “in order to prevent its further publication”.)
Back in the olden days when I was mucking about in the political blogosphere, an oft-discussed (or at least nodded at) question was exactly where the contours lay when quoting extensively from other blogs or news outlets – while no precise answer is possible, Warman v Fournier confirms that the traditional “substantial part” analysis applies to online fora such as blogs and discussion boards.
No Infringement Because the Use Constituted Fair Dealing
The Warman v Fournier decision is notable for being one of the few decisions to pay attention to the “news reporting” category of fair dealing found in Section 29.2 of the Copyright Act (Canada). This element of the decision makes it worthy of attention from anyone engaging in online use of copyrighted materials. Reiterating that the fair dealing provisions need to be, pace the Supreme Court, interpreted in a “large and liberal” manner, the court set out the fair dealing test:
(1) the purpose of the dealing; (2) the character of the dealing; (3) the amount of the dealing; (4) alternatives to the dealing; (5) the nature of the work; and (6) the effect of the dealing on the work
In assessing the use of the work in this case, the court found as follows:
|Factors Weighing in Favour of FD
||Factors Weighing Against FD
|(1) purpose – news reporting (interpreted purposively)
||(2) character of the dealing – distributed on internet
|(3) amount of the dealing – “very limited”, copied mostly facts and not most of the author’s original commentary
||(4) alternatives to the dealing – could have provided a summary, rather than quotations
|(5) nature of the work – because withdrawn from publication, finding of FD serves purpose of copyright law of disseminating works
|(6) effect of dealing on the work – plaintiff was not exploiting the work
One could quibble with the analysis (e.g., it’s not clear that the court gave enough weight to the fact that the work had been withdrawn due to defamation concerns, which perhaps should have shifted item (5) to the other column), but again it is useful as an indicator of how courts approach these issues.
Linking is not Infringement (But Are All Links Created Equal?)
The aspect of the decision which has prompted much of the notice it has received relates to the issue of hyperlinking and its relationship to copyright infringement. This portion of the decision receives the shortest treatment from the court, which is regrettable, as it appears no less deserving of extended attention from the court and from readers.
According to the decision, the plaintiff posted the photograph in question (which included the plaintiff’s image) on his website and the defendant’s website contained a link to the photograph. The court held that the creation of the link could not constitute infringement because the plaintiff had already authorized telecommunication of the work to the public by the act of posting the photo on his website, and that the defendants merely also “telecommunicated the work to the public”. But the analysis seems too brief. It fails to distinguish, for example, between the nature of different types of hyperlinks: the defendant’s website did not contain simply a “hot” text link which could be clicked on and which would take the viewer to the plaintiff’s website – rather, according to para. 8 of the decision, the website contained an “inline link” to the plaintiff’s website, meaning that visitors to the defendant’s website saw the photograph at the defendant’s website, and likely were unaware that the photograph was being hosted at the plaintiff’s website. That sort of “linking” seems qualitatively different from a simple hypertext link which would require the visitor to the defendant’s website to “click through” in order to view the photograph.
The court’s observation about the linking (“the [photograph] was within the applicant’s full control and if he did not wish it to be communicated by telecommunication, he could remove it from his website, as he eventually did”), while not wrong, seems slightly glib. While it’s clear that the plaintiff had implicitly authorized others to “access” the photograph by putting it online, I’m not sure it automatically follows that he had also implicity authorized them to replicate it at another website by inline linking (technically, inline linking can be prevented on the host’s side (i.e., completely removing the photo is not only the option to prevent inline linking), but the details of that are well above my pay grade).
The difference between a simple hyperlink and an inline link seems meaningful – as an example, when the Supreme Court of Canada in its Crookes v Newton decision held that a hyperlink did not constitute publication of a defamatory statement, Justice Abella’s decision drew an analogy between a hyperlink and a “reference” (or footnote), stating “Both communicate that something exists, but do not, by themselves, communicate its content. And they both require some act on the part of a third party before he or she gains access to the content.” (para. 30). The decision draws repeated distinctions between “mere” hyperlinks and other types of links, saying “Embedded or automatic links, for example, may well prove to be of consequence in future cases, but these differences were not argued in this case or addressed in the courts below, and therefore need not be addressed here.” Regrettably, the Warman v Fournier trial decision did not spend much time interrogating the differences in types of links.
Warman v Fournier is likely to be appealed – it strikes me that if it is, the appellate decision will need to spend a little more time on the nature of different types of hyperlinks and their relevance to the copyright infringement analysis.