There is no easy way to manage private facts in public documents, especially when the documents are posted online. For more than a decade, Canadian courts, tribunals and legal publishers have followed recommendations from the Canadian Judicial Council to shield court records from being easily accessed by search engines. This, it is felt, provides the needed balance between accessibility and privacy.
Earlier this week, a Federal Court ruled in A.T. v. Globe24h.com that a Romanian website violated Canadian privacy laws, and the accepted protocol for online publication of court records, when it scraped Canadian court decisions from CanLII and other sites and then republished them in a manner searchable by Google. The operator of this occasionally ad-supported website did more than depart from the norms – — he sought payment to remove or anonymize the documents when contacted by people named in those decisions. The court observed that in addition to violating the publishing norms, there was no factual basis upon which the site operator could rely, for example, on otherwise available legal defences such as claiming its reproductions were “journalistic” in nature or were posted for the purpose of informing the public on matters of public interest.
Fair enough. We don’t seek to defend the site operator. In fact, I [Colin] was CEO of CanLII during the period in which the scraping and republication took place and the public uproar began. I personally fielded dozens of calls from worried individuals. I also directly engaged with the offender and repeatedly reviewed the matter in detail with national media and with courts and tribunals across the country. This does not leave one with much sympathy for the author of everyone’s troubles.
We are concerned, however, that despite the desirable result in this case, the Federal Court has narrowed the open- court principle and Parliament’s intent concerning republication of the law. This was a ruling made in consideration of one issue – — privacy – — and not of the balance between privacy and accessibility.
When the court said: “In this instance, there is no need to republish the decisions to make them accessible as they are already available on Canadian websites for free,” it created a hurdle – — one not found in the preamble to the still-relevant 1997 Federal Reproduction of Law Order that provides: “the Government of Canada wishes to facilitate access to its law by licensing the reproduction of federal law without charge or permission.”
When the court said the Romanian site could not benefit from the privacy regulation provisions permitting collection, use and disclosure of private information in a record or document of a judicial or quasi-judicial body that is available to the public, it created another hurdle — t. This time on the basis that the purpose of the publication did not “relate directly to the purpose for which the information appears in the record or document.” In looking to the financial motives of the site operator, the court didn’t address the fact that the reproduction was complete – — that is, private facts weren’t mined and reposted, but appeared in their original context alongside the complete judicial record.
We say again that we are not trying to defend the site operator or his actions. But when the court accepted the privacy commissioner’s arguments that the offender’s actions could undermine the administration of justice, harm participants in the justice system and potentially discourage people from accessing the justice system, there was no acknowledgement that the same concerns go directly to the heart of the open- court principle and balanced interests that supported the first complete publication of the same document.
At what point does something posted online become too public? No one wants to return to the days of paper-only judgments – — the ones that were kept in the basement of local courthouses, where people had to line up during “special” hours to access them. So, presumably, most people can agree that publishing judgments online is a good thing.
With very few exceptions, our courts are public, the process is public and the results are public. The court in this case stated that the details of Canada’s judgments were too easily found because, essentially, they didn’t make the public work hard enough to find them. At first glance, stating that the details of a particular judgment are overly accessible when they’re already publically available doesn’t make a lot of sense. And we’re not sure it makes sense at second glance, either. If the information is online and available to all, then it is accessible to all. And as such, it is available to Canadians and Russians and Australians and Romanians. That is the nature of the net. And that is what the word public has meant for the past 20 years.
Does something posted on YouTube become too public when it is also posted on Vimeo? Does a judgment published on CanLII become too public if it’s also posted in its entirety on Facebook? The content doesn’t change – — it’s just that more people are likely to see it. And isn’t that the point of publishing our judgments in the first place? We don’t want to make them available and hope nobody sees them. We want everyone to be able to access court records so they can see, without formality or cost, how our judicial system thinks and works.
So, yes, this particular respondent used the publically available content in ways he should not have. Yes, he is a bad guy. But in the end, the process is public, the judgments are public and so are we all.
Finding the balance is not easy. The original Judicial Council guidance concerning online posting of judgments came in 2005. New guidance is long overdue. Nearly everything else that we understand about the Internet has changed since that era. We simply can’t leave the re-definition of the open- court principle to random cases, bad facts and judgments that never fully engage the question.
Colin Lachance is an Ottawa-based lawyer and the CEO of legal publisher Compass.law.
Ava Chisling is a Montreal-based media lawyer and an award-winning writer and editor.
