Copyright Infringement on a Website: the risks of scraping and framing

By Richard Stobbe

If photos are available on the internet, then… they’re free for the taking, right?

Wait, that’s not how copyright law works? In the world of copyright, each original image theoretically has an “author” who created the image, and is the first owner of the copyright. The exception to this rule is that an image (or, indeed, any other copyright-protected work), which is created by an employee in the course of employment is owned by the employer. So, an image has an owner, even if that owner chose to post the image online. And copying that image without the permission of the owner could be an infringement of the owner’s copyright.

That seemingly simple question was the subject of a lawsuit between two rival companies who are in the business of listing online advertisements for new and used vehicles. Trader Corp. had a head start in the Canadian marketplace with its autotrader.ca website. Trader had the practice of training its employees and contractors to take vehicle photos in a certain way, with certain staging and lighting. A U.S. competitor, CarGurus, entered the market in 2015. It was CarGurus practice to obtain its vehicle images by “indexing” or “scraping” Dealers’ websites. Essentially, the CarGurus software would “crawl” an online image to identify data of interest, and then extract the data for use on the CarGurus site.

As part of its “indexing” or “scraping”, the CarGurus site apparently included some photos that were owned by Trader. Although some back-and-forth between the parties resulted in the takedown of a large number of images from the CarGurus site, the dispute boiled over into litigation in late 2015 – the lawsuit by Trader alleged copyright infringement in relation to thousands of photos over which Trader claimed ownership.

Some interesting points arise from the decision in Trader v. CarGurus, 2017 ONSC 1841 (CanLII):

Trader was only able to establish ownership in 152,532 photos. There were thousands of photos for which Trader could not show convincing evidence of ownership. This speaks to the inherent difficulty in establishing a solid evidentiary record of ownership of individual images across a complex business operation.

CarGurus raised a number of noteworthy defences:

  • First, CarGurus argued that in the case of some of the photos there was no actual “copying” or “reproduction” of the original image file. Rather, CarGurus argued that it merely framed the image files. Put another way, “although the images from Dealers’ websites appeared to be part of CarGurus’ website, they were not physically present on CarGurus’ server, but located on servers hosting the Dealers’ websites.”
    The court was not convinced by the novel argument. “In my view” the court declared, “when CarGurus displayed the photo on its website, it was ‘making it available’ to the public by telecommunication (in a way that allowed a member of the public to have access to it from a place and at a time individually chosen by that member), regardless of whether the photo was actually stored on CarGurus’ server or on a third party’s server.”
    The court decided that by making the images available to the public in through its framing technique, CarGurus infringed Trader’s copyright.
    This tells us that copyright infringement can occur even where the infringer is not storing or hosting the copyright-protected work on its own server.
  • Second, CarGurus attempted to mount a “fair dealing” defense. It is not an infringement if the copying is for the purpose of “research or private study.” The court also rejected this argument, saying that even if a consumer was engaged in “research” when viewing the images in the course of car shopping, it would be too much of a stretch to accept that CarGurus was engaged in research. Theirs was clearly a commercial purpose.
  • Lastly, the lawyers for CarGuru argued that, even if infringement did occur, CarGurus should be shielded from any damages award by virtue of section 41.27(1) of the Copyright Act. This provision was originally designed as a “safe harbour” for search engines and other network intermediaries who might inadvertently cache or reproduce copyright-protected works in the course of providing services, provided the search engine or intermediary met the definition of an “information location tool”. Although CarGurus does assist users with search functions (after all, it searches and finds vehicle listings), the court batted away this argument, pointing out that CarGurus cannot be considered an intermediary in the same way a search engine is. This particular subsection has never been the subject of judicial interpretation until now.

Having dismissed these defences, the Court assessed damages for copyright infringement at $2.00 per photo, for a total statutory damages award in the amount of $305,064.

CALGARY – 07:00 MT

No comments