Date: 20210708

Docket: T-1779-18

Citation: 2021 FC 723

Ottawa, Ontario, July 8, 2021

PRESENT: The Associate Chief Justice Gagné

IN THE MATTER OF a reference pursuant to subsection 18.3(1) of the Federal Courts Act, R.S.C. 1985, c. F-7 of questions or issues of law and jurisdiction concerning the Personal Information Protection and Electronic Documents Act, S.C. 2000, c. 5 that have arisen in the course of an investigation into a complaint before the Privacy Commissioner of Canada

BETWEEN:

THE PRIVACY COMMISSIONER OF CANADA

Applicant

JUDGMENT AND REASONS

I. Overview

[1] The Privacy Commissioner of Canada [Commissioner] has brought a Reference pursuant to subsection 18.3(1) of the Federal Courts Act, RSC 1985, c F-7, which allows a federal office to bring a question or issue of law, of jurisdiction or of practice and procedure to this Court, at any stage of its own proceedings.

[2] This Reference is brought on the fringe of the Commissioner’s investigation of a complaint made in June 2017 against Google LLC [Google]. The Complainant states that Google contravenes the Personal Information Protection and Electronic Documents Act, SC 2000, c 5 [PIPEDA] by displaying links to news articles that contain personal and sensitive information about him, when his name is searched using Google’s search engine. Information in the materials filed by the parties that could identify the Complainant will remain confidential in accordance with the Order of Madam Prothonotary Tabib dated November 2, 2018.

[3] The Commissioner submits two Reference questions pertaining to the application of PIPEDA to Google’s search engine service:

1. Does Google, in the operation of its search engine service, collect, use or disclose personal information in the course of commercial activities within the meaning of paragraph 4(1)(a) of PIPEDA when it indexes webpages and presents search results in response to searches of an individual’s name?

2. Is the operation of Google’s search engine service excluded from the application of Part 1 of PIPEDA by virtue of paragraph 4(2)(c) of PIPEDA because it involves the collection, use or disclosure of personal information for journalistic, artistic or literary purposes and for no other purpose?

[4] The parties on this Reference are, on one side, the Commissioner, the Attorney General of Canada [AGC] and the Complainant, and, on the other side, Google.

[5] The Canadian Broadcasting Corporation/Société Radio-Canada [CBC] and the Samuelson-Glushko Canadian Internet Policy and Public Interest Clinic [CIPPIC] were granted intervener status. Generally speaking, CBC sides with Google whereas CIPPIC sides with the Commissioner.

II. Facts

A. Background

[6] Although this Reference is factually linked to the June 2017 complaint, the Commissioner had a vested interest in the issue of online privacy prior to receiving the complaint. In 2016, the Commissioner began to research issues of online privacy and sought consultation and commentary on whether a “right to be forgotten” could apply in the Canadian context.

[7] In his complaint, the Complainant alleges that the news articles Google displays in its search results contain outdated and inaccurate information and they disclose sensitive information. The Complainant says that the fact that Google prominently links these articles to his name in search results has caused, and continues to cause him, direct harm, including physical assault, lost employment opportunities and severe social stigma. At the Commissioner’s suggestion, the Complainant approached Google to request that it remove the links in question from searches for his name. Google declined to do so and provided the following advice to the Complainant:

We encourage you to resolve any disputes directly with the owner of the website in question. Visit [hyperlink omitted] to learn how to contact a site’s webmaster and request a change. If you pursue legal action against this site that results in the removal of the material, our search results will display this change after we next crawl the site. If the webmaster makes these changes and you need us to expedite the removal of the cached copy, please submit your request using our webpage removal request tool.

[Emphasis omitted.]

[8] In November 2017, the Commissioner notified Google of the complaint and asked for a written response.

[9] In January 2018, the Commissioner published a Draft Position Paper on Online Reputation expressing the view that, in certain circumstances, PIPEDA applies to search engines like Google. A consequence of PIPEDA’s application would be that Google might be required to remove links to content containing personal information. The Commissioner invited interested parties to provide feedback. At the time of hearing this Reference, the Draft Position Paper was not yet finalized.

[10] In March 2018, Google responded to the complaint. Google stated that PIPEDA did not apply to its search engine, which is not a commercial activity within the meaning of paragraph 4(1)(a) of PIPEDA. In the alternative, argued Google, its search engine is a journalistic or literary operation within the meaning of paragraph 4(2)(c) of PIPEDA. Google further submitted that an interpretation of PIPEDA that required it to delist lawful public content is contrary to freedom of expression as enshrined in the Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11 [the Charter].

[11] Considering Google’s response to the complaint and the responses received thus far to the Draft Position Paper, the Commissioner prepared a draft Preliminary Findings of Fact on the complaint and decided to refer the jurisdictional issues raised in Google’s response to this Court. The Commissioner chose not to refer Google’s Charter argument.

[12] Google brought a motion to determine whether the Reference questions should be reframed more broadly to encompass its Charter argument. Prothonotary Tabib dismissed Google’s motion to expand the Reference questions and I upheld Prothonotary Tabib’s judgment on July 22, 2019 [July Order]. I noted:

[66] It is difficult to understand how Google can claim that PIPEDA applies in a way that breaches its constitutional rights, before PIPEDA has even been applied to it. In my view, the fundamental flaw with this position is that Google is attempting to rely on its constitutional arguments as a response to the jurisdictional issue raised in the reference. Google’s argument is circular, because for PIPEDA to infringe Google’s rights, it must first apply to Google in one way or another, so as to limit a Charter right.

[13] The Reference questions will thus be answered within the parameters proposed by the Commissioner.

B. How does Google Search Work?

[14] Google Search operates through three basic functions: crawling, indexing, and displaying search results. Crawling is an automated process that involves the use of software called a “crawler” that continuously accesses publicly available webpages and transmits information from those webpages to be indexed or referenced. As pages are updated over time, Google’s crawlers access the updated version of the pages. Information identified by the crawler is then added to an index where Google organizes the information. This index displays an entry for each word on each webpage indexed. The index is updated if a new webpage appears or an existing webpage is altered or removed.

[15] When an individual enters a search query, Google Search uses algorithms to display search results linking to the relevant web pages in the index, ranked from most to least relevant. Google Search displays the title of the webpages, links to the webpages, and automatically generated short textual “snippets” from the webpages. The information displayed is content from the webpage itself and is subject to the instructions of website operators.

[16] Google displays responses to a user search query in the order that Google considers of likely interest to the user as determined by algorithms maintained by Google, which analyze many different factors, including how recent the content is and the number of times it has been linked to by prominent websites.

[17] Website operators have control over whether their content is displayed by Google Search. Website operators may configure their servers to refuse to respond to requests for access from one of Google’s crawlers, preventing the contents of that URL from being indexed and displayed by Google Search. Operators can also provide more detailed instructions to Google on whether and how to capture particular content through files titled “robots.txt”.

[18] For news articles, like the content at issue here, news organizations control what stories appear in Google Search as part of their overall journalistic mission: first, by deciding what to publish on their website; then, by deciding whether to remove or change any information on their website; and finally, they can use robots.txt files to direct Google on which stories from their websites to include in Google Search. The evidence before the Court is that all of the news organizations at issue in the complaint allowed Google to include their webpages in Google Search.

[19] Google generates revenue when users click on advertisements displayed in a search result. Although ads are not displayed in response to all queries, whether ads appear will depend on keywords selected by advertisers and not on categories of search that Google chose to exempt from advertisement. An advertiser creates the text of the advertisement and selects keywords in respect of which it wants its advertisement to be displayed. That advertisement will then be displayed in response to a query for those keywords. On a search result where advertisements are displayed, they are given the “Ad” label and appear before the search results.

[20] According to Google, it is highly unlikely – although not impossible – that advertisers would select the name of a private individual like the Complainant as a keyword for their advertisements.

III. Issues

[21] Thus, this Reference raises the following issues:

1. Does Google, in the operation of its search engine service, collect, use or disclose personal information in the course of commercial activities within the meaning of paragraph 4(1)(a) of PIPEDA when it indexes webpages and presents search results in response to searches of an individual’s name?

2. Is the operation of Google’s search engine service excluded from the application of Part 1 of PIPEDA by virtue of paragraph 4(2)(c) of PIPEDA because it involves the collection, use or disclosure of personal information for journalistic, artistic or literary purposes and for no other purpose?

[22] Google raises a third issue which is:

3. Should this Court simply decline to answer the Reference questions or dismiss the Reference because the questions cannot/should not be answered without addressing the constitutional issues and/or because there is an inadequate evidentiary record before this Court?

[23] In my view, the third issue contains a contradiction. Courts should, in fact, refrain from addressing constitutional issues without an adequate evidentiary record. As indicated in the July Order, these questions are left for the Commissioner who will benefit from a complete evidentiary record and will be in a better position to assess whether PIPEDA can be applied in the way the Complainant wishes it to apply, without violating Charter values. Therefore, the Court will answer the Reference questions, and only those questions.

IV. Analysis

A. Reference Question Number 1

(1) The Commissioner’s position

[24] The Commissioner first notes that Google does not deny being an “organization” that “collects”, “uses” or “discloses” personal information as required by paragraph 4(1)(a) of PIPEDA. Google acquires personal information, for instance, by copying the content from webpages and storing it within its index, which information Google then displays during a search for a person’s name. The Commissioner submits that this Court in AT v Globe24h.com, 2017 FC 114 [Globe24h] held that an organization is caught by paragraph 4(1)(a) when it copies and republishes personal information from other websites.

[25] The Commissioner adds that the personal information is under Google’s control since Google can determine how often or whether it crawls a given webpage with personal information, and, for example, in what order that information will appear in search results. The Commissioner argues that Google’s control over personal information and its ability to make personal information readily accessible make Google subject to legal obligations (Google Inc v Equustek Solutions Inc, 2017 SCC 34).

[26] Commercial activity is broadly defined in PIPEDA as “any particular transaction, act or conduct or any regular course of conduct that is of a commercial character, including the selling, bartering or leasing of donor, membership or other fundraising lists” (s 2(1)). The Commissioner notes that Google sells advertising space associated with searches and Google earns money when users click on an ad; this is clearly commercial activity. Further still, Google collects data from users, such as through their past searches, to disclose more targeted ads in the future. The Commissioner relies on Equustek Solutions Inc v Jack, 2014 BCSC 1063 for the assertion that Google’s advertising and search engine service are “inextricably linked” (at para 60).

[27] The Commissioner submits that this analysis is not altered by the fact that some Google searches yield ads and others do not, as the term “commercial activity” in PIPEDA includes “any regular course of conduct that is of a commercial character” (s 2(1)). The Commissioner submits that “the activity as a whole”, being usage of Google’s search engine service, is the activity that must be assessed.

(2) The Complainant’s position

[28] The Complainant echoes the Commissioner’s submissions. He further submits that the Court should apply a “large and liberal” interpretation of “commercial activity” in the present context. First, PIPEDA’s role is to protect autonomy, privacy and dignity by providing individuals with control over their personal information. Second, PIPEDA has a role in a global network of information sharing practices by ensuring public confidence in the protection of personal information. Third, PIPEDA’s purpose would be undermined if companies could innovate business models to escape its application.

[29] The Complainant notes that Globe24h and past decisions of the Commissioner militate in favour of the finding that the operation of Google’s search engine, when a user searches for a person’s name, is a commercial activity.

(3) The AGC’s position

[30] The AGC largely agrees with the submissions of the Commissioner and the Complainant, but differs in one respect. The AGC submits that the Court should employ the established approach to statutory interpretation (as found in Canada Trustco Mortgage Co v Canada, 2005 SCC 54 at para 10 [Canada Trustco]) and should not defer to Charter values or PIPEDA’s quasi-constitutional status in doing so. The quasi-constitutional nature of PIPEDA does not transform or alter the proper approach to statutory interpretation (Lavigne v Canada (Office of the Commissioner of Official Languages), 2002 SCC 53 at para 25 [Lavigne]).

(4) CIPPIC’s position

[31] CIPPIC also submits that the first Reference question should be answered in the affirmative. Like the Complainant, CIPPIC argues that an interpretation of commercial activity that excludes Google is an interpretation that frustrates PIPEDA’s purpose, as it would then exclude modern e-commerce business models. The thesis of CIPPIC’s position is that commercial activity must be understood in relation to Google’s business model. CIPPIC submits that a prevalent modern e-commerce model, and the one adopted by Google, is to provide low or no-cost services, but to accumulate data in order to sell targeted advertising. CIPPIC submits that every search query contributes to Google’s commercial value because searches improve personalization by contributing to an individual’s data profile.

[32] CIPPIC also argues that PIPEDA’s purpose would be frustrated by an interpretation that excluded Google searches for a person’s name because the legislature intended for PIPEDA to apply to “‘software tracking agents’ that prey on the ‘data trail’ created when individuals browse the Internet” (citing to House of Commons, Standing Committee on Industry, Evidence, 36-1, No 76 (1 December 1998) at 1545 (Hon John Manley)). CIPPIC therefore argues that it would be contrary to the legislature’s intention for PIPEDA to apply to some Google searches and not others.

(5) Google’s position

[33] Google submits that PIPEDA’s purpose is not simply to protect personal information, but rather to protect some personal information while also allowing industry to collect and use information as a component of success in the “information economy”. As seen from the Hansard debates, PIPEDA also aims at protecting free speech and freedom of the press (House of Commons, Standing Committee on Industry, Evidence, 36-1, No 77 (2 December 1998) at 1645 (Bruce Phillips); House of Commons, Standing Committee on Industry, Evidence, 36-1, No 83 (3 February 1999) at 1615 (Hon Anne McLellan)).

[34] Google disputes that “collects”, “uses” or “discloses” as written in paragraph 4(1)(a) of PIPEDA was meant to apply to online intermediaries who convey or disseminate the free speech of third parties. Google submits the publishers, and not the intermediary, are the ones who collect, use or disclose content.

[35] With further reference to the Hansard debates, Google submits that, with respect to commercial activity, one must look at the specific transaction at stake, and not the organization (House of Commons Debates, 36-2, No 7 (20 October 1999) at 1755 (Hon Susan Whelan)). Google notes that the Commissioner’s Preliminary Findings of Fact acknowledges that no advertisements were generated alongside a search for the Complainant’s name. In the absence of any evidence that Google displays advertisements in connection with a search for an individual’s name, the Commissioner cannot override Parliament’s intention to look at the transactional level by looking at its entire business model instead.

[36] Google relies on this Court’s decision in State Farm Mutual Automobile Insurance Company v Privacy Commissioner of Canada, 2010 FC 736 [State Farm] in support of the argument that one must look at the nature of the activity in question to ascertain whether it is of a commercial character.

[37] Google similarly relies on Professor Teresa Scassa’s blog post entitled “OPC Report on Online Reputation Misses the Mark on the application of PIPEDA to search engines” (31 January 2018) in further support of the position that the relevant activity that may be the subject of PIPEDA is the user’s activity and not that of Google.

(6) The Court’s Answer to Reference Question Number 1

[38] The issue raised by this Reference is one of statutory interpretation. The modern approach to statutory interpretation is well established in Supreme Court jurisprudence. As stated in Canada Trustco:

10 It has been long established as a matter of statutory interpretation that “the words of an Act are to be read in their entire context and in their grammatical and ordinary sense harmoniously with the scheme of the Act, the object of the Act, and the intention of Parliament”: see 65302 British Columbia Ltd. v. Canada, [1999] 3 S.C.R. 804, at para. 50. The interpretation of a statutory provision must be made according to a textual, contextual and purposive analysis to find a meaning that is harmonious with the Act as a whole. When the words of a provision are precise and unequivocal, the ordinary meaning of the words play a dominant role in the interpretive process. On the other hand, where the words can support more than one reasonable meaning, the ordinary meaning of the words plays a lesser role. The relative effects of ordinary meaning, context and purpose on the interpretive process may vary, but in all cases the court must seek to read the provisions of an Act as a harmonious whole.

(See also Canada (Canadian Human Rights Commission) v Canada (Attorney General), 2011 SCC 53 at para 33 and Rizzo & Rizzo Shoes Ltd (Re), [1998] 1 SCR 27 at para 21).

[39] PIPEDA is quasi-constitutional legislation because its focus is on ensuring that individuals can control their personal information, which is intimately connected to their individual autonomy, dignity and privacy (Alberta (Information and Privacy Commissioner) v United Food and Commercial Workers, Local 401, 2013 SCC 62 at para 19 [United Food]). However, this does not change the exercise of statutory interpretation to be undertaken by the Court (Lavigne at para 25).

[40] The sub-questions that are relevant to this first Reference question are whether a) in the operation of Google’s search engine service, when it indexes webpages and presents search results in response to searches of an individual’s name, Google discloses, collects or uses personal information, and b) whether it does so in the course of commercial activities.

[41] As to the first sub-question, Google accepts that the plain and ordinary meaning of the words “collect”, “use” and “disclose” could capture indexing and displaying search results. However, Google is of the view that having in mind the purpose and object of PIPEDA, it is not clear that “collect, use and disclose” was ever intended to apply to online intermediaries conveying the speech of third parties to users. Google compares itself to a library that creates a user-friendly index of works published by others. According to Google, publishers use, collect and disclose personal information as contemplated by PIPEDA.

[42] In its written submissions, Google almost systematically uses the terms “collect, use, and disclose”, whereas Part 1 of PIPEDA applies to organizations who “collect, use or disclose” personal information. It might be true that it is the publisher (or any commercial or non-commercial website operator) that primarily collects the personal information. However, in order to copy the content of webpages, Google uses its automated “crawlers”. The crawlers access and copy the content found on publicly accessible webpages, including personal information, and transmit it to Google’s servers for indexing. Google determines which websites it crawls and the frequency with which it does so. That is a collection activity.

[43] Furthermore, in light of the factual background that is before the Court, Google uses personal information – Google needs as much information as possible to make its search engine as comprehensive and valuable as possible for users, and consequently for advertisers. Google also discloses the information; Google has control over its “snippets” and the order in which the information appears in the Google search results.

[44] In addition to using and disclosing the personal information of the subject of the search, Google readily admits that it also collects, uses and discloses the personal information of the user (the individual performing the search).

[45] As for the second sub-question, Google argues that in light of the factual background presented by the Commissioner, the only possible answer to the first Reference question is no: Google does not collect, use or disclose personal information in the course of commercial activities within the meaning of paragraph 4(1)(a) of PIPEDA, when it indexes webpages and presents search results in response to searches of an individual’s name.

[46] In Google’s view, the focus of the Court should be on the free service Google provides to the content providers (the media in this case) and the user of the search engine; Google acts on behalf of the latter to connect the user, without charge, to the information displayed by the former.

[47] This aspect of the analysis is important: it ensures that Parliament remains within the boundaries of its power to legislate. PIPEDA’s privacy protections were created within a regulatory scheme relying on the federal “general trade and commerce” power. Provincial legislatures may exclusively make laws in relation to property and civil rights, pursuant to section 92 of the Constitution Act, 1867 (UK), 30 & 31 Vict, c 3, reprinted in RSC 1985, Appendix II, No 5. The requirement that PIPEDA only apply to commercial activity is therefore a critical constitutional guardrail. That is why subsection 4(1) specifically excludes non-commercial activities, provincially regulated workplaces and other entities beyond the reach of federal regulation. As held in State Farm at paragraph 40, under PIPEDA, “personal information is regulated only insofar as it relates to how the Canadian economy functions and operates.”

[48] Considering the Commissioner’s admission in its Preliminary Findings of Fact on the complaint that, “there is no evidence that advertisements appear alongside the search results generated by a search of the Complainant’s name”, Google states that the activity under review is not commercial in the usual and traditional sense of buying and selling of goods and services (3510395 Canada Inc v Canada (Attorney General), 2020 FCA 103 at paras 140-141).

[49] With respect, I believe this is over simplifying the activity at stake here.

[50] Google’s search engine allows individuals to search for content, including text, images and videos, found on webpages that make up the World Wide Web. When a user enters a search query, Google’s proprietary computer algorithms cross-reference its index and identifies the content that Google considers relevant to the query. Its algorithms take into account a number of factors, including the search terms used, a user’s past searches, the user’s location, and the perceived quality of a webpage to determine which search results are, in Google’s view, most relevant to a user.

[51] Google is a for-profit corporation and one of the most successful technology businesses of the modern era. Its search engine is far and away the most dominant with some estimates suggesting that it is used to conduct 70-75% of all internet searches globally. According to Google, its search engine is used to conduct millions of searches each day. The popularity of its search engine and its other products have made Google one of the most profitable corporations today. Google’s parent company reported that it earned approximately $63.5 billion (USD) from Google in the first half of 2018 alone.

[52] The bulk of Google’s revenue comes from advertising (roughly $54.7 billion in the first half of 2018). Most of Google’s advertising revenue is in turn generated from Google’s search engine and other online services. As Google notes, “[m]uch of our business is based on showing ads, both on Google services and on websites and mobile apps that partner with us” (Exhibit S-17 to the Lachance affidavit). Google’s growth in revenue for the second quarter of 2018 was “primarily driven by increases in mobile search resulting from ongoing growth in user adoption and usage, as well as continued growth in advertiser activity” (Exhibit S-25 to the Lachance affidavit). Google also experienced “growth in desktop search due to improvements in ad formats and delivery” (Exhibit S-25 to the Lachance affidavit).

[53] Advertisers pay Google a fee each time a user clicks on an ad in Google search results or takes an action having seen an ad, such as downloading an app.

[54] Because Google also delivers ads on third-party websites, Google may earn revenue if a user navigates to a webpage listed in Google search results and then clicks on or views a Google-delivered ad displayed on that page.

[55] Google promotes its advertising business by highlighting the popularity of its search engine. As its promotional materials note, “[w]ith millions of searches per day on Google, you can make sure your customers notice your brand, consider your offerings, and take action” if you advertise in Google search results (Exhibit S-21 to the Lachance affidavit). Google also promotes to advertisers its ability to target ads to users of its search engine based on personal information it has about them (Exhibit S-19 to the Lachance affidavit).

[56] This business model was foreseen when Parliament enacted PIPEDA. The Government of Canada’s consultation paper that preceded PIPEDA noted that, “[t]he challenge of the electronic age is that with each transaction we leave a data trail that can be compiled to provide a detailed record of our personal history and preferences” (Exhibit D-1 to the Ballott affidavit p 23). In this new environment, personal information itself becomes a commodity, to be mined and used for profit (Exhibit D-1 to the Ballott affidavit 23).

[57] That is to say that even if Google provides free services to the content providers and the user of the search engine, it has a flagrant commercial interest in connecting these two players. There is a real trade between Google and the users of its search engine. In exchange for the information displayed in the search results, the users provide a variety of personal information (their location, preferences, interests, consumption patterns etc.). That personal information is used for profit.

[58] And, in order to attract the users, Google needs to feed them with the most accurate and customized information they are searching for.

[59] Therefore, unless it is forced to do so, Google has no commercial interest in de-indexing or de-listing information from its search engine. In my view, every component of that business model is a commercial activity as contemplated by PIPEDA. To have a microscopic look at the free aspect (i.e. no payment in money made) of the search for the user, or to the free aspect of the “library service” provided to news media would be, in my respectful view, a misunderstanding of Google’s business model. All these activities are intertwined, they depend on one another, and they are all necessary components of that business model.

[60] I would therefore answer the first Reference question in the affirmative.

B. Reference Question Number 2

(1) The Commissioner’s position

[61] The Commissioner submits that journalism, though not defined in PIPEDA, can be given a wide meaning, but not so wide as to encompass any type of communication (United Food and Commercial Workers, Local 401 v Alberta (Attorney General), 2012 ABCA 130 at para 56 [United Food ABCA]).

[62] The Commissioner refers to the meetings of the Standing Committee on Industry to argue that the journalistic exemption was included in paragraph 4(2)(c) of PIPEDA out of concern that PIPEDA could impede freedom of the press otherwise. Further, that the words “for no other purpose” were included in the journalism exemption to ensure that even media companies had to comply with PIPEDA when engaged in purely commercial activity.

[63] The Commissioner also notes that this Court applied a three-pronged test to identify journalism in Globe24h for determining whether the Globe24h.com website was engaged in journalism when it republished court and tribunal decisions. An activity is journalism “where its purpose is to (1) inform the community on issues the community values, (2) it involves an element of original production, and (3) it involves a ‘self-conscious discipline calculated to provide an accurate and fair description of facts, opinion and debate at play within a situation’” (Globe24h at para 68). The Commissioner submits that Google does not satisfy this test. First, Google makes information universally accessible, which is broader than informing a community about issues the community values. Second, Google does not create any information. Finally, searches are returned as the result of search terms and there is no self-conscious discipline on the part of Google to ensure that the results are fair and accurate. Google’s purpose should not be confused with the purpose of the websites listed in the search results.

(2) The Complainant’s position

[64] The Complainant also submits that Google does not meet the test for journalism set out in Globe24h, for the same reasons as argued by the Commissioner. The Complainant also argues that the purposes of journalistic material on websites like that of CBC are distinct from Google’s purpose.

[65] The Complainant submits that the presence of plainly non-journalistic search results is irreconcilable with the position that Google’s purpose is journalistic when it indexes webpages and presents search results in response to searches of an individual’s name.

(3) The AGC’s position

[66] The AGC submits that, “the legal definition of what constitutes journalism cannot be so wide as to include a company whose role, at most, is to act as a conduit through which some users gain access to journalism created and hosted by third parties.” The AGC also relies on the test for journalism from Globe24h to argue that Google is not engaged in journalism.

(4) CIPPIC’s position

[67] CIPPIC argues that paragraph 4(2)(c) of PIPEDA protects freedom of the press – the freedom of the press recently described by the Supreme Court in Denis v Côté, 2019 SCC 44. Parliament clearly did not intend to exempt Google by enacting paragraph 4(2)(c) of PIPEDA. CIPPIC also relies on the test for journalism found in Globe24h and argues that journalists are liable for what they publish in a way that internet intermediaries are not.

(5) CBC’s position

[68] CBC submits that the journalistic exemption found in paragraph 4(2)(c) exists to prevent the Commissioner or this Court from making editorial decisions about what should be in the news. CBC notes there are a variety of other checks on journalistic activity that exist at law – in statute or the common law.

[69] CBC submits this Reference has the potential to expose journalism to the Office of the Privacy Commissioner, which can then make one-sided decisions about journalistic publications without input from the content creator.

[70] CBC proposes the following test: First, the Court should determine whether the material disclosed by Google in search results is the work-product of a recognized news media. If it is, then the exclusion found in paragraph 4(2)(c) of PIPEDA should apply without further analysis. Second, if it is alleged that the work-product is journalistic in nature despite not coming from recognized news media, the finder of fact should ask whether it is the product of journalism and if so, the exclusion should apply. For the second prong of CBC’s test, which requires defining journalism, CBC relies upon the three pronged journalism test from Globe24h. However, CBC applies the three pronged test from the perspective of the content creator, and not from Google’s perspective. Essentially, the Court must look at whether Google is engaging in journalism in respect of the personal information it returns in search results.

[71] CBC notes that Google plays an essential role in the dissemination of journalistic information and removing that content from Google search results is tantamount to removing the information from the public realm.

[72] CBC submits that applying the journalistic exemption to Google strikes the proper balance between Charter rights, like freedom of expression, and privacy interests, which are personal in nature. CBC submits that Canadians courts have preferred prioritizing the public interest rather than purely personal interests like embarrassment (citing MEH v Williams, 2012 ONCA 35).

(6) Google’s position

[73] Google submits that the purpose of the journalism exemption found in paragraph 4(2)(c) of PIPEDA is to protect Charter rights like freedom of speech.

[74] Google further submits that the journalism exemption must be broader than “simply reporting and publishing news stories”, but should also include indexing and displaying content for journalistic purposes. Google argues that the dissemination of journalistic content is part of the ordinary meaning of the word journalism as found in the Encyclopedia Britannica and by way of the definition of journalism found in the Canada Evidence Act, RSC 1985, c C-5.

[75] Google asks this Court to focus on the Reference question, being whether the exemption applies to personal information that Google collects, uses or discloses when indexing and presenting webpages in response to searches for an individual’s name. Google also asks this Court to note the personal information that is “at issue”, being the information about the Complainant that was the focus of his complaint. Considering the Reference question, and the personal information at issue, the Court is tasked with determining whether Google is acting with a journalistic purpose when it links to news articles containing personal information. Google submits that the Office of the Privacy Commissioner will be given the power to regulate journalism – something that it cannot do directly – if this Court interprets that paragraph 4(2)(c) of PIPEDA encompasses Google’s acts of indexing and displaying news media websites in its search results.

[76] Google argues that this Court should apply a Charter-compliant interpretation of paragraph 4(2)(c) of PIPEDA. Relying on United Food, Google argues that the Supreme Court has already found that an analogous provision in Personal Information Protection Act, SA 2003, c P-6.5 violated the Charter when construed such that the impugned activity had to be 100% journalistic in order to qualify for the journalism exemption. Google submits this Court should adopt a Charter-compliant understanding of PIPEDA instead, one where disseminating news articles is protected by the Charter.

[77] Google does not agree that the three part test from Globe24h is applicable here. In that case, the test was applied to a publisher of content, whereas Google is only an intermediary. Further, there was no party that was carrying on journalism in Globe24h; the court decisions that the website republished were not journalism and the website was not engaged in a journalistic endeavour by republishing them. Rather, in this matter, Google submits that it is disseminating recognized journalistic content.

(7) The Court’s answer to Reference Question Number 2

[78] Again, I believe Google and CBC are taking a somewhat microscopic look at the situation. They urge the Court to only look at the object of the Complainant’s complaint: a handful of articles published by recognized news media.

[79] Yet, the second Reference question is whether the operation of Google’s search engine is excluded from the application of Part I of PIPEDA by virtue of paragraph 4(2)(c) of PIPEDA, because it involves the collection, use or disclosure of personal information for journalistic, artistic or literary purposes and for no other purposes.

[80] In addition, even if the Court’s analysis was limited to searches for an individual’s name, the evidence shows that Google still displays what it considers to be relevant information linked to that name in the form of search results. A search for an individual’s name may return, for instance, content from personal blogs and websites, chatrooms, social media sites, websites of businesses, governments, non-governmental organizations, as well as news organizations (exhibit R to the Lachance affidavit at para 40). The personal information that can be displayed in response to such searches is thus wide and varied, it is not limited to media content, and it can lead to a detailed portrait of an individual.

[81] I agree that Google facilitates access to information, such as news media. I also agree that facilitating access to information is often associated with “publishing” said information. This point was made by the Supreme Court in Crookes v Newton, 2011 SCC 47 at paragraph 29 [Crookes] in the context of the defamation framework, where publication is one element. However, facilitation is also just one indicator of publication. In Crookes, the Supreme Court concluded that hyperlinks were not publication for the purposes of defamation even though they did facilitate access to information (at para 29). The Court’s analysis in Crookes suggested that hyperlinks were not publication because the person who refers to content using a hyperlink does not participate in the content’s creation (at para 28), the person who hyperlinks to the content has no control over the content (at para 27), and the hyperlinks express no opinion (at para 30).

[82] In my view, the reasons why hyperlinks are not publication for the purposes of defamation are also relevant to search results – Google has no control over the content of search results, the search results themselves express no opinion, and Google does not create the content of the search results.

[83] An ordinary understanding of the word journalism encompasses content creation and content control as shown by the definition for journalism found in Globe24h. It is relevant to note that the test used in that case was proposed by the Canadian Association of Journalists: “an activity should qualify as journalism only where its purpose is to (1) inform the community on issues the community values, (2) it involves an element of original production, and (3) it involves a ‘self-conscious discipline calculated to provide an accurate and fair description of facts, opinion and debate at play within a situation’” (Globe24h at para 68). In the matter before me, it is likewise noteworthy that CBC agrees it is an adequate definition or test.

[84] Google does not think the test from Globe24h is appropriate in this matter because Google is an intermediary for news articles, not a publisher. With respect, I am not sure where Google derived that distinction. I do not dispute that Google is an intermediary, but that distinction is not relevant to the test. Before introducing the test in Globe24h, Justice Mosley stated: “[t]he ‘journalistic’ purpose exception is not defined in PIPEDA and it has not received substantive treatment in the jurisprudence” (at para 68). By introducing the three pronged test or definition from the Canadian Association of Journalists, this Court was thereby attempting to provide insight into a previously undefined area. The Court did not say, as Google argues, that this was a test for publishers only. It is a test for aiding in the identification of what journalism encompasses.

[85] Applying the test from Globe24h to the matter before me: first, Google makes information universally accessible, which is much broader than informing a community about issues the community values; second, Google does not create or produce anything – it only displays search results; and third, there is no effort on the part of Google to determine the fairness or the accuracy of the search results. The publishers would be accountable for the accuracy of the content of a search result, not Google.

[86] In my view, the operation of Google’s search engine, even when we only consider the search results for the Complainant’s name, does not satisfy the definition of journalism accepted by the Court in Globe24h.

[87] The final and determinative aspect of this analysis is the part of paragraph 4(2)(c) which says that it only applies to organizations that collect, use or disclose personal information for journalistic purposes, “for no other purpose”. Google argues that “for no other purpose” does not exclude commercial organizations because in order to apply the journalistic exemption, an organization must already be commercial under paragraph 4(1)(a). I agree. However, that does not mean that “for no other purpose” is useless or meaningless. The legislature does not speak in vain and the presumption against tautology means that Courts should avoid adopting interpretations that render any portion of a statute meaningless or redundant (Bell ExpressVu Limited Partnership v Rex, 2002 SCC 42 at para 37 and Placer Dome Canada Ltd v Ontario (Minister of Finance), 2006 SCC 20 at para 45).

[88] Justice Mosley’s analysis in Globe24h is applicable here. He wrote:

[71] The evidence indicates that the respondent’s primary purpose is to incentivize individuals to pay to have their personal information removed from the website. A secondary purpose, until very recently, was to generate advertising revenue by driving traffic to his website through the increased exposure of personal information in search engines. There is no evidence that the respondent’s intention is to inform the public on matters of public interest.

[72] Even if the respondent’s activities could be considered journalistic in part, the exemption under paragraph 4(2)(c) only applies where the information is collected, used or disclosed exclusively for journalistic purposes. It is clear from the record that Globe24h.com’s purposes extend beyond journalism.

[Emphasis in original.]

[89] It is useful to apply a similar analysis to Google. The primary purpose of Google’s search engine service is to index and present search results. This is not a primarily journalistic purpose because although it may facilitate access to information, it contains no other defining feature of journalism, such as content control or content creation. Even though Google returns some journalism in its search results, its search results clearly extend beyond journalism.

[90] In sum, Google’s search engine service does not operate for a journalistic purpose at all, or at least it does not operate for an exclusively journalistic purpose.

[91] I now turn specifically to CBC’s submissions. CBC proposes a two-part test that focuses on the nature of each search result. First, the Court should determine whether the material disclosed by Google in search results is the work-product of a recognized news media. If it is, then the exclusion found in paragraph 4(2)(c) of PIPEDA should apply without further analysis. If a search result is not from recognized news media, the Court should determine whether it is journalism.

[92] With respect, CBC’s test does not answer the Reference question. It does not analyse whether Google’s search engine service collects, uses or discloses personal information for journalistic purposes. It only focuses on whether recognized news media and other journalists collect, use or disclose personal information. That is not helpful to the matter before the Court.

[93] CBC also takes the position that it makes sense to apply PIPEDA to Google’s dissemination of news media in a manner that protects freedom of expression as enshrined in the Charter.

[94] Yet, it is not necessary to resort to Charter values unless there is “genuine ambiguity” in interpretation of a statute (Wilson v British Columbia (Superintendent of Motor Vehicles), 2015 SCC 47 at para 25). And in my view, there is no ambiguity here. First, Parliament limited PIPEDA to only protecting “journalism” and not freedom of expression broadly speaking. As stated by the Court of Appeal of Alberta in United Food ABCA at paragraph 56, “[i]t is unreasonable to think that the Legislature intended the term ‘journalism’ to be so wide as to encompass everything within the phrase ‘freedom of opinion and expression’.” Second, Parliament limited PIPEDA to protecting the collection, disclosure and use of personal information for exclusively journalistic purposes. Third, a popular or ordinary understanding of journalism, as provided by journalists themselves, does not encompass Google’s search engine service.

[95] Therefore, I find that Google’s purposes for collecting, using and disclosing personal information for its search engine service are not journalistic, and they are certainly not exclusively so.

[96] The Court’s answer to the second Reference question is no.

V. Conclusion

[97] For all of these reasons, the Court answers the first question in the affirmative and the second question in the negative. However, this does not determine the outcome of the Complainant’s complaint, the power of the Commissioner to recommend deindexing, the constitutionality of PIPEDA, or any other non-reference question that is better left to the Commissioner’s proceedings.

 

JUDGMENT in T-1779-18

THIS COURT’S JUDGMENT is that:

1. This Reference is granted;

2. To the question:

Does Google, in the operation of its search engine service, collect, use or disclose personal information in the course of commercial activities within the meaning of paragraph 4(1)(a) of PIPEDA when it indexes webpages and presents search results in response to searches of an individual’s name?

The Court’s answer is: Yes

3. To the question:

Is the operation of Google’s search engine service excluded from the application of Part 1 of PIPEDA by virtue of paragraph 4(2)(c) of PIPEDA because it involves the collection, use or disclosure of personal information for journalistic, artistic or literary purposes and for no other purpose?

The Court’s answer is: No

4. The parties shall provide submissions regarding costs, not exceeding five pages, within 30 days of this judgment.

“Jocelyne Gagné”

 

FEDERAL COURT

SOLICITORS OF RECORD

DOCKET:	T-1779-18
STYLE OF CAUSE:	IN THE MATTER OF a reference pursuant to subsection 18.3(1) of the Federal Courts Act, R.S.C. 1985, c. F-7 of questions or issues of law and jurisdiction concerning the Personal Information Protection and Electronic Documents Act, S.C. 2000, c. 5 that have arisen in the course of an investigation into a complaint before the Privacy Commissioner of Canada
PLACE OF HEARING:	HELD BY VIDEOCONFERENCE
DATE OF HEARING:	January 26, 2021
JUDGMENT AND REASONS:	GAGNÉ A.C.J.
DATED:	JULY 8, 2021

APPEARANCES:

Peter Engelmann Colleen Bauman Regan Morris Kelly Stephens	For The Applicant
Mark Phillips Michael Fenrick Charlotte Calon	For The COMPLAINANT
James D. Bunting Anisah Hassan Adam Soliman David TS Fraser	FOR GOOGLE LLC
Christian Leblanc Sean Moreman	FOR CBC/RADIO-CANADA
Christopher Rupar Kirk Shannon Fraser Harland	FOR THE ATTORNEY GENERAL OF CANADA
Alexandra (Lex) Gill Tamir Israel	FOR SAMUEL-GLUSHKO CANADIAN INTERNET POLICY AND PUBLIC INTEREST CLINIC

SOLICITORS OF RECORD:

Goldblatt Partners Ottawa, ON Office of the Privacy Commissioner of Canada Gatineau, QC	For The Applicant
Paliare Roland Rosenberg Rothstein LLP Toronto, ON	For The COMPLAINANT
Tyr LLP Toronto, ON McInnes Cooper Halifax, NS	FOR GOOGLE LLC
Fasken Martineau Dumoulin, S.E.N.C.R.L., S.R.L. Montreal, QC Canadian Broadcasting Corporation/Société Radio-Canada Toronto, ON	FOR CBC/RADIO-CANADA
Attorney General of Canada Ottawa, ON	For The Attorney general of canada
Trudel Johnston & Lespérance Montréal, QC Samuelson-Glushko Canadian Internet Policy & Public Interest Clinic (CIPPIC) Ottawa, ON	FOR SAMUEL-GLUSHKO CANADIAN INTERNET POLICY AND PUBLIC INTEREST CLINIC