Tags
access to information
AI
AIDA
AI governance
AI regulation
Ambush Marketing
artificial intelligence
big data
bill c11
Bill c27
copyright
data governance
data protection
data strategy
freedom of expression
Geospatial
geospatial data
intellectual property
Internet
internet law
IP
open courts
open data
open government
personal information
pipeda
Privacy
smart cities
trademarks
transparency
|
Thursday, 22 November 2018 06:51
Bill C-86 proposes positive changes for Official Marks regime: Comments to the Standing Senate Committee on Banking, Trade and Commerce,
The following is a copy of remarks I made in an appearance before the Standing Senate Committee on Banking, Trade and Commerce, on November 21, 2018. The remarks are about proposed amendments to the Trade-marks Act found in the omnibus Bill C-86. I realize that not everyone is as interested in official marks as I am. To get a sense of what the fuss is all about, you can have a look at my posts about the overprotection of Olympic Marks, the dispute over the Spirit Bear mark, the struggle to register a trademark in the face of an wrongly granted and since abandoned official mark, Canada Post’s official marks for POSTAL CODE and CODE POSTAL and a previous private member’s bill to reform official marks. Canada’s official marks regime has long been criticized by lawyers, academics and the Federal Court. In fact, it is the Federal Court that has, over the years, created some much needed boundaries for these “super marks”. The problems with official marks are well known, but they have largely been ignored by Parliament. It is therefore refreshing to see the proposed changes in ss. 215 and 216 of Bill C-86. Sections 215 and 216 address only one of the problems with the official marks regime. Although it is an important one, it is worth noting that there is more that could be done. The goal of my remarks will be to identify what I see as two shortfalls of ss 215 and 216. Official marks are a subcategory of prohibited marks, which may not be adopted, used or registered unless consent is provided. They are available to “public authorities”. A public authority need only ask the Registrar of Trademarks to give public notice of its adoption and use of an official mark for that mark to be protected. There are no limits to what can be adopted. There are no registration formalities, no examination or opposition proceedings. Until the very recent decision of the Federal Court in Quality Program Services Inc. v. Canada, it seemed nothing prevented a public authority from obtaining an official mark that was identical to or confusing with an already registered trademark. While Quality Program Services at least provides consequences for adopting a confusing official mark, it is currently under appeal and it is not certain that the decision will be upheld. This is another instance of the Federal Court trying to set boundaries for official marks that simply have not been set in the legislation. Official marks are theoretically perpetual in duration. They remain on the register until they are either voluntarily withdrawn by the ‘owner’ (and owners rarely think to do this), or until a successful (and costly) action for judicial review results in one being struck from the Register. Until the Ontario Association of Architects decision in 2002 tightened up the meaning of ‘public authority’, official marks were handed out like Halloween candy, and many entities that were not ‘public authorities’ were able to obtain official marks. Many of these erroneously-issued official marks continue to exist today; in fact, the Register of Trademarks has become cluttered with official marks that are either invalid or no longer in use. Sections 215 and 216 address at least part of this last problem. They provide an administrative process through which either the Registrar or any person prepared to pay the prescribed fee can have an official mark invalidated if the entity that obtained the mark “is not a public authority or no longer exists.” This is a good thing. I would, however, suggest one amendment to the proposed new s. 9(4). Where it is the case (as per the new s. 9(3)) that the entity that obtained the official mark was not a public authority or has ceased to exist, s. 9(4) allows the Registrar to give public notice that subparagraph (1)(n)(iii) “does not apply with respect to the badge, crest, emblem or mark”. As it is currently worded, this is permissive– the Registrar “may” give public notice of non-application. In my view, it should read: (4) In the circumstances set out in subsection (3), the Registrar may, on his or her own initiative or shall, at the request of a person who pays a prescribed fee, give public notice that subparagraph (1)(n)(iii) does not apply with respect to the badge, crest, emblem or mark. There is no reason why a person who has paid a fee to have established the invalidity of an official mark should not have the Registrar give public notice of this. I would also suggest that the process for invalidating official marks should extend to those that have not been used within the preceding three years – in other words, something parallel to s. 45 of the Trade-marks Act which provides an administrative procedure to remove unused registered trademarks from the Register. There are hundreds of ‘public authorities’ at federal and provincial levels across Canada, and they adopt official marks for all sorts of programs and initiatives, many of which are relatively transient. There should be a means by which official marks can simply be cleared from the Register when they are no longer used. Thus, I would recommend adding new subsections 9(5) and (6) to the effect that: (5) The Registrar may at any time – and, at the written request of a person who pays a prescribed fee, made after three years from the date that public notice was given of an official mark, shall, unless the Registrar sees good reason to the contrary – give notice to the public authority requiring it to furnish, within three months, an affidavit or a statutory declaration showing that the official mark was in use at any time during the three year period immediately preceding the date of the notice and, if not, the date when it was last so in use and the reason for the absence of such use since that date. (6) Where, by reason of the evidence furnished to the Registrar or the failure to furnish any evidence, it appears to the Registrar that an official mark was not used at any time during the three year period immediately preceding the date of the notice and that the absence of use has not been due to special circumstances that excuse it, the Registrar shall give public notice that subparagraph (1)(n)(iii) does not apply with respect to the badge, crest, emblem or mark. These are my comments on changes to the official marks regime that most closely relate to the amendments in Bill C-86. The regime has other deficiencies which I would be happy to discuss.
Published in
Trademarks
Thursday, 04 October 2018 05:43
Artist sued in Canada for copyright infringement for AI-related art project
A law suit filed in Montreal this summer raises novel copyright arguments regarding AI-generated works. The plaintiffs are artist Amel Chamandy and Galerie NuEdge Fine Arts (which sells and exhibits her art). They are suing artist Adam Basanta for copyright and trademark infringement. (The trademark infringement arguments are not discussed in this post). Mr Basanta is a world renowned new media artist who experiments with AI in his work. (See the Globe and Mail story by Chris Hannay on this law suit here). According to a letter dated July 4, filed with the court, Mr. Basanta’s current project is “to explore connections between mass technologies, using those technologies themselves.” He explains his process in a video which can be found here. Essentially, he has created what he describes as an “art-factory” that randomly generates images without human input. The images created are then “analyzed by a series of deep-learning algorithms trained on a database of contemporary artworks in economic and institutional circulation” (see artist’s website). The images used in the database of artworks are found online. Where the analysis finds a match of more than 83% between one of the randomly generated images and an image in the database, the randomly generated image is presented online with the percentage match, the title of the painting it matches, and the artist’s name. This information is also tweeted out. The image of the painting that matches the AI image is not reproduced or displayed on the website or on Twitter. One of Mr Basanta’s images was an 85.81% match with a painting by Ms Chamandy titled “Your World Without Paper”. This information was reported on Mr Basanta’s website and Twitter accounts along with the machine-generated image which resulted in the match. The copyright infringement allegation is essentially that “the process used by the Defendant to compare his computer generated images to Amel Chamandy’s work necessarily required an unauthorized copy of such a work to be made.” (Statement of Claim, para 30). Ms Chamandy claims statutory damages of up to $20,000 for the commercial use of her work. Mr Basanta, for his part, argues that there is no display of Ms Chamandy’s work, and therefore no infringement. AI has been generating much attention in the copyright world. AI algorithms need to be ‘trained’ and this training requires that they be fed a constant supply of text, data or images, depending upon the algorithm. Rights holders argue that the use of their works in this way without consent is infringement. The argument is that the process requires unauthorized copies to be fed into the system for algorithmic analysis. Debates have raged in the EU over a text-and-data mining exception to copyright infringement which would make this type of use of copyright protected works acceptable so long as it is for research purposes. Other uses would require clearance for a fee. There has already been considerable debate in Europe over whether research is a broad enough basis for the exception and what activities it would include. If a similar exception is to be adopted in Canada in the next round of copyright reform, we will face similar challenges in defining its boundaries. Of course, the Chamandy case is not the conventional text and data mining situation. The copied image is not used to train algorithms. Rather, it is used in an analysis to assess similarities with another image. But such uses are not unknown in the AI world. Facial recognition technologies match live captured images with stored face prints. In this case, the third party artwork images are like the stored face prints. It is AI, just not the usual text and data mining paradigm. This should also raise questions about how to draft exceptions or to interpret existing exceptions to address AI-related creativity and innovation. In the US, some argue that the ‘fair use’ exception to infringement is broad enough to support text and data mining uses of copyright protected works since the resulting AI output is transformative. Canada’s fair dealing provisions are less generous than U.S. fair use, but it is still possible to argue that text and data mining uses might be ‘fair’. Canadian law recognizes fair dealing for the purposes of research or private study, so if an activity qualifies as ‘research’ it might be fair dealing. The fairness of any dealing requires a contextual analysis. In this case the dealing might be considered fair since the end result only reports on similarities but does not reproduce any of the protected images for public view. The problem, of course, with fair dealing defences is that each case turns on its own facts. The fact-dependent inquiry necessary for a fair dealing defense could be a major brake on innovation and creativity – either by dissuading uses out of fear of costly infringement claims or by driving up the cost of innovation by requiring rights clearance in order to avoid being sued. The claim of statutory damages here is also interesting. Statutory damages were introduced in s. 38.1 of the Copyright Act to give plaintiffs an alternative to proving actual damage. For commercial infringements, statutory damages can range from $500 to $20,000 per work infringed; for non-commercial infringement the range is $100 to $5,000 for all infringements and all works involved. A judge’s actual award of damages within these ranges is guided by factors that include the need for deterrence, and the conduct of the parties. Ms Chamandy asserts that Mr Basanda’s infringement is commercial, even though the commercial dimension is difficult to see. It would be interesting to consider whether the enhancement of his reputation or profile as an artist or any increase in his ability to obtain grants would be considered “commercial”. Beyond the challenge of identifying what is commercial activity in this context, it opens a window into the potential impact of statutory damages in text and data mining activities. If such activities are considered to infringe copyright and are not clearly within an exception, then in Canada, a commercial text and data miner who consumes – say 500,000 different images to train an algorithm – might find themselves, even on the low end of the spectrum, liable for $250 million dollars in statutory damages. Admittedly, the Act contains a clause that gives a judge the discretion to reduce an award of statutory damages if it is “grossly out of proportion to the infringement”. However, not knowing what a court might do or by how much the damages might be reduced creates uncertainty that can place a chill on innovation. Although in this case, there may well be a good fair dealing defence, the realities of AI would seem to require either a clear set of exceptions to clarify infringement issues, or some other scheme to compensate creators which expressly excludes resort to statutory damages. The vast number of works that might be consumed to train an algorithm for commercial purposes makes statutory damages, even at the low end of the scale, potentially devastating and creates a chill.
Published in
Copyright Law
Monday, 04 December 2017 14:37
Controlling Data: Competition, Privacy and Copyright Issues Raised by Real Estate Data
Canada’s Federal Court of Appeal has handed down a decision that addresses important issues regarding control over commercially valuable data. The decision results from an appeal of an earlier ruling of the Competition Tribunal regarding the ability of the Toronto Real Estate Board (TREB) to limit the uses to which its compilation of current and historical property listings in the Greater Toronto Area (GTA) can be put. Through its operations, the TREB compiles a vast database of real estate listings. Information is added to the database on an ongoing basis by real estate brokers who contribute data each time a property is listed with them. Real estate agents who are members of TREB in turn receive access to a subset of this data via an electronic feed. They are permitted to make this data available through their individual websites. However, the TREB does not permit all of its data to be shared through this feed; some data is available only through other means such as in-person consultation, or communications of snippets of data via email or fax. The dispute arose after the Competition Commissioner applied to the Competition Tribunal for a ruling as to whether the limits imposed by the TREB on the data available through the electronic feed inhibited the ability of “virtual office websites” (VOWs) to compete with more conventional real estate brokerages. The tribunal ruled that they did, and the matter was appealed to the Federal Court of Appeal. Although the primary focus of the Court’s decision was on the competition issues, it also addressed questions of privacy and copyright law. The Federal Court of Appeal found that the TREB’s practices of restricting available data – including information on the selling price of homes – had anticompetitive effects that limited the range of broker services that were available in the GTA, limited innovation, and had an adverse impact on entry into and expansion of relevant markets. This aspect of the decision highlights how controlling key data in a sector of the economy can amount to anti-competitive behavior. Data are often valuable commercial assets; too much exclusivity over data may, however, pose problems. Understanding the limits of control over data is therefore an important and challenging issue for businesses and regulators alike. The TREB had argued that one of the reasons why it could not provide certain data through its digital feed was because these data were personal information and it had obligations under the Personal Information Protection and Electronic Documents Act to not disclose this information without appropriate consent. The TREB relied on a finding of the Office of the Privacy Commissioner of Canada that the selling price of a home (among those data held back by TREB) was personal information because it could lead to inferences about the individual who sold the house (e.g.: their negotiating skills, the pressure on them to sell, etc.). The Court noted that the TREB already shared the information it collected with its members. Information that was not made available through the digital feed was still available through more conventional methods. In fact, the Court noted that the information was very widely shared. It ruled that the consent provided by individuals to this sharing of information would apply to the sharing of the same information through a digital feed. It stated: “PIPEDA only requires new consent where information is used for a new purpose, not where it is distributed via new methods. The introduction of VOWs is not a new purpose – the purpose remains to provide residential real estate services [. . .].” (at para 165) The Court’s decision was influenced by the fact that the consent form was very broadly worded. Through it, TREB obtained consent to the use and dissemination of the data “during the term of the listing and thereafter.” This conclusion is interesting, as many have argued that the privacy impacts are different depending on how information is shared or disseminated. In other words, it could have a significant impact on privacy if information that is originally shared only on request, is later published on the Internet. Consent to disclosure of the information using one medium might not translate into consent to a much broader disclosure. However, the Court’s decision should be read in the context of both the very broad terms of the consent form and the very significant level of disclosure that was already taking place. The court’s statement that “PIPEDA only requires new consent where information is used for a new purpose, not where it is distributed via new methods” should not be taken to mean that new methods of distribution do not necessarily reflect new purposes that go beyond the original consent. The Federal Court of Appeal also took note of the Supreme Court of Canada’s recent decision in Royal Bank of Canada v. Trang. In the course of deciding whether to find implied consent to a disclosure of personal information, the Supreme Court of Canada had ruled that while the balance owing on a mortgage was personal information, it was less sensitive than other financial information because the original amount of the mortgage, the rate of interest and the due date for the mortgage were all publicly available information from which an estimate of the amount owing could be derived. The Federal Court of Appeal found that the selling price of a home was similarly capable of being derived from other publicly available data sources and was thus not particularly sensitive personal information. In addition to finding that there would be no breach of PIPEDA, the Federal Court of Appeal seemed to accept the Tribunal’s view that the TREB was using PIPEDA in an attempt to avoid wider sharing of its data, not because of concerns for privacy, but in order to maintain its control over the data. It found that TREBs conduct was “consistent with the conclusion that it considered the consents were sufficiently specific to be compliant with PIPEDA in the electronic distribution of the disputed data on a VOW, and that it drew no distinction between the means of distribution.” (at para 171) Finally, the Competition Tribunal had ruled that the TREB did not have copyright in its compilation of data because the compilation lacked sufficient originality in the selection or arrangement of the underlying data. Copyright in a compilation depends upon this originality in selection or arrangement because facts themselves are in the public domain. The Federal Court of Appeal declined to decide the copyright issue since the finding that the VOW policy was anti-competitive meant that copyright could not be relied upon as a defence. Nevertheless, it addressed the copyright question in obiter (meaning that its comments are merely opinion and not binding precedent). The Federal Court of Appeal noted that the issue of whether there is copyright in a compilation of facts is a “highly contextual and factual determination” (at para 186). The Court of Appeal took note of the Tribunal’s findings that “TREB’s specific compilation of data from real estate listings amounts to a mechanical exercise” (at para 194), and agreed that the threshold for originality was not met. The Federal Court of Appeal dismissed the relevance of TREB’s arguments about the ways in which its database was used, noting that “how a “work” is used casts little light on the question of originality.” (at para 195) The Court also found no relevance to the claims made in TREB’s contracts to copyright in its database. Claiming copyright is one thing, establishing it in law is quite another.
Note that leave to appeal this decision to the Supreme Court of Canada was denied on August 23, 2018.
Published in
Copyright Law
Thursday, 14 September 2017 12:42
Data visualization removed over threat of legal action
A recent incident raises important issues about excessive control over data and information. Open data activists, who have long battled to liberate government data will recognize the principles at play here. The difference in this case, is that the data over which control is being asserted are in private sector hands. Yet while the law necessarily provides means for private sector organizations to exercise control over data and information in appropriate circumstances, this control is not without its limits. In this case, the limits may have been seriously overstepped. A Toronto-area man who posted his own data-visualization based on Toronto real estate data hasreceived a blunt cease and desist notice from counsel for the Toronto Real Estate Board (TREB). The story, which was reported by David Hains in Metro News, explains that the 26 year old data analyst named Shafquat Arefeen created the visualization as a personal project and posted it on his non-commercial website. The visualization, which is no longer available as a result of the letter, provided an overview of housing sales activity in Toronto and contained data that, among other things, showed the differences between listing and selling prices, including data for specific house sales. The cease and desist letter makes it clear that the TREB believes that the data were taken from its proprietary website. There are some interesting issues around accessing and using data hosted on a web platform, including whether any terms of use associated with the site are binding on the user. The letter, however, does not raise any contractual claims. Instead, it asserts copyright – apparently in the data. Mr. Arefeen denies that he obtained the data from the TREB site. Whether he did or not, the copyright claims are independently worth considering. It is a basic and fundamental principle of copyright law that facts and information are in the public domain. The Federal Court of Canada has clearly stated: “there can be no copyright in information.”( Nautical Data International Inc v. C-Map USA Inc. at para 11), as has the Supreme Court of Canada: “copyright protection does not extend to facts or ideas but is limited to the expression of ideas.” (CCH Canadian Ltd. v. Law Society of Upper Canada, at para 22). Copyright law does protect compilations, including compilations of data. Yet, where data are part of a compilation, all that is protected is the original selection or arrangement of the data. In a 2016 decision of the Competition Tribunal regarding the TREB’s data, the Tribunal stated that it was: “not persuaded that TREB owns copyright in the MLS Database, including the Disputed Data. In brief, the Tribunal has concluded that TREB has not led sufficient evidence to establish the level of skill, judgment and labour required for the MLS Database to benefit from copyright protection.” (at para 731) Let’s assume for the sake of argument that there is a copyright claim to be made. Even in those cases where copyright in a compilation of data is found to exist, a second user who does not take a substantial part of the selection or arrangement of the data does not infringe copyright. If Mr. Arefeen’s visualization was his own original expression of the data that he used, then it would be very difficult to sustain an argument that there was a substantial taking of the arrangement of the TREB’s data. It is not clear whether it constituted substantial taking of any original selection of the data – this is far from an open and shut issue. Yet even if a court were to find substantial taking of a selection, Mr. Arefeen would be entitled to rely upon the fair dealing exceptions in the Copyright Act. The Supreme Court of Canada has mandated a generous approach to fair dealing, and there is every possibility that this non-commercial use might be considered fair – in other words, not infringing. The bottom line is that any claim to either copyright in the data or infringement of any such copyright would appear to be very weak. The cease and desist letter also contains strong language alleging that Mr. Arefeen’s use of the data was a violation of the Personal Information Protection and Electronic Documents Act (PIPEDA). PIPEDA applies only to personal information that is collected, used or disclosed in the course of commercial activity. Mr. Arefeen’s website appears to be non-commercial – it does not even contain advertising. If this is the case, PIPEDA does not apply. There are also exceptions to the application of PIPEDA where information is collected, used or disclosed for journalistic or artistic purposes. Frankly, it’s hard to see how PIPEDA would apply in this instance. The cease and desist letter achieved its objective in that Mr. Arefeen took down his data visualization, and it is no longer available to a public according to the newspaper coverage found it useful and interesting. This allows the TREB to maintain control over its closely controlled data about the real estate market in Ontario. It also enables it to restrict public engagement with data that are relevant, interesting and important to Toronto residents. The outcome highlights the imbalance between well-resourced data ‘owners’ and data users – particularly those who act in the public interest. Such users often have limited resources either to pay for data licences or to hire lawyers to push back against excessive claims. The result is far from being in the public interest.
Published in
Copyright Law
Tuesday, 08 August 2017 08:39
On data ownership rights
In early July 2017 I attended an excellent workshop hosted by researchers at the Centre for Information Technology, Society and Law at the University of Zurich. The objective of the workshop was to bring a group of academic experts together to discuss data ownership rights. It is perhaps not surprising that the issue of ownership rights in data is bubbling to the surface as we move further into the evolving big data environment. Data have been described as the new “oil” of the information society. They have a tremendous value and are strongly linked to innovation. One of the ways in which industrialized nations have nurtured innovation has been through the creation of intangible property rights such as intellectual property rights. Data ownership rights flow from that same industrial era mind set. However, it is far from clear that this paradigm is a good fit for data and data-related innovation. The concept of a data ownership right was raised in the EU in the European Data Market Study, Second Interim Report, June 2016. At page 146, it states: In fact, the way data are made available and the extent to which data are flowing across sectors and organizations, play a fundamental role in sustaining and developing the emergence of a European data-driven economy. In defining and specifying the rights to create, edit, modify, share and restrict access to data, data ownership becomes a pivotal factor affecting a growing number of potential data users and an increasing range of data-related activities. One might perhaps be forgiven for thinking that there are already data ownership rights; for example, terms of service for websites frequently state that the company behind the website “owns” its data. Canada’s federal government even got its knuckles rapped by the Federal Court of Appeal for making a similar copyright-based claim in one of its data licences (see my post on this decision here). And, while the law of confidential information could be argued to provide a kind of property right in data or information, in reality what is protected by this body of law is the confidentiality of the information. Once confidentiality is lost, it is clear that there is no underlying ‘property’ right in the data. Policy makers have long been wary of extending IP rights to data – and for some very good reasons. Copyright law, for example, does not protect “facts”, viewing them instead as the building blocks for creativity and expression, and therefore part of the public domain. Of course, copyright law does protect the original selection and arrangement that goes into creating a compilation of facts (i.e. a data set). How extensive this protection ultimately is depends on what a court sees as the taking of a substantial part of that selection or arrangement. It is this protection for compilations of data that no doubt supports those Terms-of-Service claims to ownership of data mentioned above, but the scope and extent of copyright protection in such circumstances is nevertheless limited and uncertain. In the EU, database rights have provided a broader protection for databases, but it still, fundamentally, is not a protection for the data that make up the database. It is difficult to see where the interest in a data ownership right is coming from. No clear or pressing need to enhance the protection available for data has been identified. Data ownership rights might be more likely to create confusion and uncertainty – and to increase transaction costs and slow innovation – than to improve the current situation. It would be difficult – and hugely problematic – even to begin to try to identify the ‘owners’ of rights in data and to manage the potential competing interests. And while there are undoubtedly issues around the fairness of particular uses of data, or the legitimacy of means used to acquire data, existing laws already offer a range of recourses and remedies that may be applicable in any given case. The brief summary of our meetings on data ownership is now publicly available, and it addresses these and many other issues relating to data ownership rights. Our conclusions – that there is no evident need for a new data ownership right and that it would be impossibly difficult to define or constrain – offer a caution to those who regard property rights as a panacea in marketplaces of all kinds.
Published in
Copyright Law
Tuesday, 14 February 2017 14:11
Smart Cities: Data Ownership and Privacy Issues
Note: the following are my speaking notes for my appearance before the Standing Committee on Transport, Infrastructure and Communities, February 14, 2017. The Committee is exploring issues relating Infrastructure and Smart Communities. I have added hyperlinks to relevant research papers or reports. Thank you for the opportunity to address the Standing Committee on Transport, Infrastructure and Communities on the issue of smart cities. My research on smart cities is from a law and policy perspective. I have focused on issues around data ownership and control and the related issues of transparency, accountability and privacy. The “smart” in “smart cities” is shorthand for the generation and analysis of data from sensor-laden cities. The data and its accompanying analytics are meant to enable better decision-making around planning and resource-allocation. But the smart city does not arise in a public policy vacuum. Almost in parallel to the development of so-called smart cities, is the growing open government movement that champions open data and open information as keys to greater transparency, civic engagement and innovation. My comments speak to the importance of ensuring that the development of smart cities is consistent with the goals of open government. In the big data environment, data is a resource. Where the collection or generation of data is paid by taxpayers it is surely a public resource. My research has considered the location of rights of ownership and control over data in a variety of smart-cities contexts, and raises concerns over the potential loss of control over such data, particularly rights to re-use the data whether it is for innovation, civic engagement or transparency purposes. Smart cities innovation will result in the collection of massive quantities of data and these data will be analyzed to generate predictions, visualizations, and other analytics. For the purposes of this very brief presentation, I will characterize this data as having 3 potential sources: 1) newly embedded sensor technologies that become part of smart cities infrastructure; 2) already existing systems by which cities collect and process data; and 3) citizen-generated data (in other words, data that is produced by citizens as a result of their daily activities and captured by some form of portable technology). Let me briefly provide examples of these three situations. The first scenario involves newly embedded sensors that become part of smart cities infrastructure. Assume that a municipal transit authority contracts with a private sector company for hardware and software services for the collection and processing of real-time GPS data from public transit vehicles. Who will own the data that is generated through these services? Will it be the municipality that owns and operates the fleet of vehicles, or the company that owns the sensors and the proprietary algorithms that process the data? The answer, which will be governed by the terms of the contract between the parties, will determine whether the transit authority is able to share this data with the public as open data. This example raises the issue of the extent to which ‘data sovereignty’ should be part of any smart cities plan. In other words, should policies be in place to ensure that cities own and/or control the data which they collect in relation to their operations. To go a step further, should federal funding for smart infrastructure be tied to obligations to make non-personal data available as open data? The second scenario is where cities take their existing data and contract with the private sector for its analysis. For example, a municipal police service provides their crime incident data to a private sector company that offers analytics services such as publicly accessible crime maps. Opting to use the pre-packaged private sector platform may have implications for the availability of the same data as open data (which in turn has implications for transparency, civic engagement and innovation). It may also result in the use of data analytics services that are not appropriately customized to the particular Canadian local, regional or national contexts. In the third scenario, a government contracts for data that has been gathered by sensors owned by private sector companies. The data may come from GPS systems installed in cars, from smart phones or their associated apps, from fitness devices, and so on. Depending upon the terms of the contract, the municipality may not be allowed to share the data upon which it is making its planning decisions. This will have important implications for the transparency of planning processes. There are also other issues. Is the city responsible for vetting the privacy policies and practices of the app companies from which they will be purchasing their data? Is there a minimum privacy standard that governments should insist upon when contracting for data collected from individuals by private sector companies? How can we reconcile private sector and public sector data protection laws where the public sector increasingly relies upon the private sector for the collection and processing of its smart cities data? Which normative regime should prevail and in what circumstances? Finally, I would like to touch on a different yet related issue. This involves the situation where a city that collects a large volume of data – including personal information – through its operation of smart services is approached by the private sector to share or sell that data in exchange for either money or services. This could be very tempting for cash-strapped municipalities. For example, a large volume of data about the movement and daily travel habits of urban residents is collected through smart card payment systems. Under what circumstances is it appropriate for governments to monetize this type of data?
Published in
Geospatial Data/Digital Cartography
Sunday, 12 February 2017 11:24
B.C. Court of Appeal Addresses Online Passing Off (Part 2)
Note: This is Part 2 of my discussion of the B.C. Court of Appeal’s decision in Vancouver Community College v. Vancouver Career College (Burnaby) Inc. Part 1 can be found here. The initial post considered issues around official marks as well as the first element of the tort of passing off in which the plaintiff must establish that they have acquired goodwill/reputation in a mark. This post considers the remaining two elements: the likelihood of confusion and the likelihood of damage. As noted in my earlier post, the B.C. Court of Appeal found that the appellant, the Vancouver Community College had considerable goodwill in the mark VCC. I was critical of this decision as it seems to conflate the official marks protection obtained by the Community College with the acronym as a trademark for the purposes of the passing off analysis. The Court of Appeal’s finding regarding the scope of the Community College’s rights in the mark VCC influences its reasoning with respect to the issue of confusion, which is the second element in the tort of passing off. The alleged passing off in this case arose out of new marketing strategies adopted by the respondent Vancouver Career College in 2009. At that point it adopted VCCollege as a trademark and registered VCCollege.ca as a domain name for its website. The appellant Vancouver Community College objected to the use by the respondent of the acronym VCC in its “internet presence”. It also objected to the Career College’s bidding on keywords that included “VCC” and “Vancouver Community College.” It argued that the result of these activities was passing off. Because the official marks arguments had been separated from the passing off claim, the Court of Appeal considered only the issue of passing off with respect to the use of “VCC”. The Court of Appeal summarized the essence of keyword advertising for the purposes of this case in these terms: “a bid on a keyword will make it more likely that the bidder’s advertisement with its domain name, linking to its website, will appear on the first search page revealed to the searcher.” (at para 19). The Court acknowledged that bidding on keywords in order to drive traffic to one’s website is legitimate, so long as it stays within the bounds of what is permissible. In the passing off context, the issue is whether the use of the keywords results in consumer confusion. When presented with a link in an ad on a search results page, the searcher has the option of following the link to the website to which it resolves. American case law on keyword advertising has focussed on the issue of confusion, rather than on the simple use of protected words as keywords. These cases have considered how the resulting ads are displayed on the page (e.g., whether they are in a location or font that distinguishes them from search results) and whether their content or presentation is misleading. The Court of Appeal does not address this case law or these issues in its confusion analysis. The Court of Appeal found that “”VCC” was the keyword that generated the most “clicks” to the respondent’s website, such that the respondent’s advertisements appeared almost always in searches for VCC (over 97% of the time), and the respondent’s text advertisements always displayed VCCollege.ca in the web address line of the advertisement.” (at para 22) However, as I noted in Part 1 of my discussion of this case, VCC is an acronym shared by both the respondent and the appellant. As an acronym, it is a weak mark. The Community College led evidence of confusion among some students searching for the Community College. The trial judge had given this evidence relatively little weight, particularly in a context in which VCC was also the acronym for the respondent’s name. He noted that the respondent’s web site made it evident that it was the site for the Career College. The Court of Appeal was critical of the trial judge for assessing confusion at the moment at which a student searching for VCC arrived at the landing page for the Career College – as opposed to when the student received the results of a browser search using VCC as a key word. According to the Court of Appeal, the authorities support a finding that first impressions are what matters in the confusion analysis. The Court of Appeal relied heavily upon Masterpiece Inc. v. Alavida Lifestyles Inc., a Supreme Court of Canada (SCC) decision involving an assessment of confusion under the Trade-marks Act. In that case, the SCC appeared to confirm that so-called “initial interest confusion” was actionable. In the internet context, initial interest confusion arises where a party’s trademarks have been used in such a way (in domain names or metatags, for example) that a person searching for the products or services of one company ends up at the website of another by mistake. In the early days of the internet, courts were more likely to find initial interest confusion to be actionable per se; more recently, courts in the United States have given searchers more credit for being able to find their way around the internet, and have looked for other evidence of uses of the marks that contribute to confusion. A searcher who quickly realizes they have made their way to a website other than the one for which they were searching is not confused. However, some have still maintained that initial interest confusion should be actionable because even if the consumer is not confused, they might still decide, once presented with similar goods or services from an alternate source, that they are happy enough to acquire them from that source rather than the one for which they were originally searching. In such circumstances, the use by a defendant of a competitor’s trademarks to draw business away from them is said to cause harm that should be actionable. In Masterpiece, the SCC stated that the diversion of consumers “diminishes the value of the goodwill associated with the trademark and business the consumer initially thought he or she was encountering in seeing the trademark. Leading consumers astray in this way is one of the evils that trademark law seeks to remedy.” (at para 73) While this has been taken by some to address initial interest confusion on the internet, it should be noted that Masterpiece did not deal with either the internet context or with passing off. Whichever view one takes on initial interest confusion, the problem in this case is that the appellant used the appropriate acronym for its name as a key word and its domain name was one in which it would doubtless be found to have a ‘legitimate interest’ under domain name dispute resolution policies. According to the Court of Appeal, the confusion required for a finding of passing off “is fully established by proof that the respondent’s domain name is equally descriptive of the appellant and contains the same acronym long associated to it.” (at para 71) This approach gives excessive scope to what should be – in the context of passing off – relatively weak rights in VCC. The acronym is obvious and appropriate for both the Vancouver Community College and the Vancouver Career College. While the Community College may have acquired goodwill in the acronym, its highly descriptive nature necessarily limits the scope of the goodwill and does not, without more, allow it to preclude its use by the Career College, itself in business for 20 years. Weak marks deserve limited protection in passing off. Having tolerated the presence of the Vancouver Career College since 1997, the action in passing off with respect to its use of its acronym online seems misdirected. The Court of Appeal found that the appellant had suffered damage to its goodwill. This flowed in part from “the lack of power to control the use of the marks to which the goodwill attached by unauthorized users” (at para 75). To characterize the Vancouver Career College as an “unauthorized user” of the appropriate acronym for its own name seems problematic. Essentially, the Court of Appeal would carve out an absolute monopoly for the appellant in VCC for use in association with education services notwithstanding the fact that the acronym is shared by two parties with names that are highly descriptive of similar services and that share the same acronym. In such circumstances, it is appropriate to require something more in the respondent’s conduct on which to base a finding of passing off. Of course, the appellant is not without its nuclear option – the VCC official mark, although it is clear that there are other difficulties with the official marks claim. As noted in Part 1, official marks give the kind of absolute protection for entirely descriptive marks that the appellant is clearly seeking. While the official mark issues in this case have yet to be resolved, it is unfortunate that the passing off analysis seems to have been carried out under the shadow of the official mark. The result is an analysis peculiar to the circumstances of this case that would be dangerous to extend to other cases of passing off.
Published in
Trademarks
Wednesday, 01 February 2017 13:34
Tobacco Trademarks and Plain Packaging Legislation in Canada
This post is based upon a presentation I gave at a panel organized jointly by the Centre for Law, Technology and Society and the Centre for Health Law, Policy and Ethics at the University of Ottawa on February 1, 2017. Canada is on the cusp of passing new legislation and enacting new regulations that will put us among a growing number of countries that have made plain packaging mandatory for tobacco products. Bill S-5, currently before the Senate, will amend the Tobacco Act to enable regulations to dictate the appearance of tobacco packaging. While the regulations are not currently available, it is to be expected that they will contain measures similar to those already introduced in Australia and Britain. Essentially plain packaging means prescribing a plain colour, size and configuration for all tobacco packages. In addition, packages will be used to convey graphic images and public health warnings. The only permitted use of tobacco trademarks will be of word marks consisting of the brand name and sub name in a prescribed font, colour and type-size. Tobacco trademarks consisting of logos, crests, images, colour, shape, configuration, or design will no longer be capable of use on tobacco product packaging. Plain packaging is a movement driven by the World Health Organization’s Framework Convention on Tobacco Control, of which Canada is a signatory. Interestingly, however, the treaty does not require signatories to implement plain packaging. Article 11 of the Convention addresses packaging, but merely requires that false and deceptive elements on packaging be banned (e.g. using “mild” to designate cigarettes that are every bit as harmful as regular cigarettes); that health warnings take up 30-50% of packaging surface; and that packages contain information about constituent ingredients and product emissions. Canada’s current packaging regulations are consistent with these requirements. Plain packaging is merely mentioned as something that signatory states “should consider” in paragraph 46 of the Guidelines for Implementing Article 11 of the Convention. Thus, it is important to underline that Canada is not under an international obligation to introduce plain packaging legislation. While the link between smoking and serious illness/death seems uncontestable, and the reduction of smoking is clearly an important public health objective, there is reason to question the wisdom of the plain packaging approach. Australia was the first country to introduce plain packaging in 2011. Its legislation survived a constitutional challenge (it was argued to be an illegal expropriation of trademark owners’ rights), and is currently being challenged before the World Trade Organization (WTO) as a violation of Australia’s obligations under the TRIPS Agreement. Although considerable sums of money have been spent on defending Australia’s statute, the evidence emerging as to the beneficial impact of the legislation is ambivalent. Plain packaging measures in Canada are also likely to face legal challenges. Restrictions on the use of trademarks in the 1988 Tobacco Products Control Act were found by the Supreme Court of Canada to be a violation of the freedom of expression of trademark owners that could not be justified under s. 1 of the Canadian Charter of Rights and Freedoms. These provisions were struck down by the Court. Provisions related to the use of tobacco trademarks in sponsorship activities in a reconstituted Tobacco Act were also challenged for violating the freedom of expression, but the Supreme Court in 2007 found that the violation was justified as rationally connected to a pressing and substantial government objective, and that it minimally impaired the rights concerned. The takeaway from these cases is that restrictions on the use of tobacco trademarks (such as those necessary to implement plain packaging) clearly violate the freedom of expression. In any court challenge, therefore, the issue will be whether the measures can be justified under s. 1 of the Charter as a “reasonable limit, demonstrably justified in a free and democratic society”. It is important to remember that plain packaging restrictions are extreme and the evidence linking plain packaging to harm reduction is ambivalent. It is not obvious at the outset that such measures would survive a Charter challenge. Trademark owners have also objected that the restrictions will harm their ability to acquire and maintain trademark rights in relation to tobacco products. Bill S-5 contains provisions that indicate that non-use of tobacco trademarks resulting from plain packaging regulations will not be a basis for the invalidation of existing registered trademarks. However, this does not settle the question. Trademark rights cannot be acquired (or maintained) at common law without use, and the law does nothing to address this category of rights. Further, certain kinds of trademarks (distinguishing guises, three-dimensional marks and other non-traditional subject marks soon to become registrable in Canada) cannot be registered until they have acquired distinctiveness through use. Plain packaging regulations might therefore constitute a bar to the registration of certain types of trademarks for use in relation to tobacco products. Canada’s existing international obligations under both the TRIPS Agreement and the NAFTA may lead to further challenges to the introduction of plain packaging. The creation of an impediment to the registration of certain types of trademarks for tobacco products may violate Article 15 of TRIPS, and there is an open and ongoing debate as to whether plain packaging laws also violate Article 20 which provides that “The use of a trademark in the course of trade shall not be unjustifiably encumbered by special requirements, such as use with another trademark, use in a special form or use in a manner detrimental to its capability to distinguish the goods or services of one undertaking from those of other undertakings.. . “. Australia’s legislation has been challenged under TRIPS, and a decision on its compliance with that treaty may be imminent. For its part, Article 1110 of the NAFTA provides that no member state can take a measure that is “tantamount to expropriation” of an investment except in limited circumstances which include a requirement to pay compensation. It is not clear whether a U.S.-based tobacco company could succeed before a NAFTA tribunal in arguing that the plain packaging laws amounted to an expropriation of their investment in their trademarks. The domestic challenge to Australia’s legislation turned on a property rights clause in the Australian constitution, and raised the question of whether the plain packaging was an expropriation of trademark rights. The majority of the court found that it did not, but of course that decision would not be binding on a NAFTA tribunal. The plain packaging regulations on the horizon for Canada are being introduced in the face of considerable uncertainty as to their legality both under Canada’s constitution and Canada’s international trade obligations. The extensive resources required to defend such measures should be weighed carefully not just against the likelihood of success of any challenges, but also against the public health benefits that are likely to flow from further changes to how tobacco products are packaged in Canada. It is perhaps also worth noting that there have been rumblings about plain packaging measures for other products considered harmful to public health, such as alcoholic beverages and junk food. The issues raised in relation to tobacco products have much broader implications, making this file one to watch.
Published in
Trademarks
Monday, 19 December 2016 08:52
Open licensing of real time data
Municipalities are under growing pressure to become “smart”. In other words, they will reap the benefits of sophisticated data analytics carried out on more and better data collected via sensors embedded throughout the urban environment. As municipalities embrace smart cities technology, a growing number of the new sensors will capture data in real time. Municipalities are also increasingly making their data open to developers and civil society alike. If municipal governments decide to make real-time data available as open data, what should an open real-time data license look like? This is a question Alexandra Diebel and I explore in a new paper just published in the Journal of e-Democracy. Our paper looks at how ten North American public transit authorities (6 in the U.S. and 4 in Canada) currently make real-time GPS public transit data available as open data. We examine the licenses used by these municipalities both for static transit data (timetables, route data) and for real-time GPS data (for example data about where transit vehicles are along their routes in real-time). Our research reveals differences in how these types of data are licensed, even when both types of data are referred to as “open” data. There is no complete consensus on the essential characteristics of open data. Nevertheless, most definitions require that to be open, data must be: (1) made available in a reusable format; (2) prepared according to certain standards; and (3) available under an open license with minimal restrictions or conditions imposed on reuse. In our paper, we focus on the third element – open licensing. To date, most of what has been written about open licensing in general and the licensing of open data in particular, has focused on the licensing of static data. Static data sets are typically downloaded through an open data portal in a one-time operation (although static data sets may still be periodically updated). By contrast, real-time data must be accessed on an ongoing basis and often at fairly short intervals such as every few seconds. The need to access data from a host server at frequent intervals places a greater demand on the resources of the data custodian – in this case often cash-strapped municipalities or public agencies. The frequent access required may also present security challenges, as servers may be vulnerable to distributed denial-of-service attacks. In addition, where municipal governments or their agencies have negotiated with private sector companies for the hardware and software to collect and process real-time data, the contracts with those companies may require certain terms and conditions to find their way into open licenses. Each of these factors may have implications for how real-time data is made available as open data. The greater commercial value of real-time data may also motivate some public agencies to alter how they make such data available to the public. While our paper focuses on real-time GPS public transit data, similar issues will likely arise in a variety of other contexts where ‘open’ real-time data are at issue. We consider how real-time data is licensed, and we identify additional terms and conditions that are imposed on users of ‘open’ real-time data. While some of these terms and conditions might be explained by the particular exigencies of real-time data (such as requirements to register for the API to access the data), others are more difficult to explain. Our paper concludes with some recommendations for the development of a standard for open real-time data licensing. This paper is part of ongoing research carried out as part of Geothink, a partnership grant project funded by the Social Sciences and Humanities Research Council of Canada.
Published in
Geospatial Data/Digital Cartography
Thursday, 10 November 2016 13:55
Federal Court affirms "right to read" in paywall copyright case
The Federal Court has just released a decision in a case that raised issues of fair dealing and copyright abuse. Blacklock’s, an Ottawa-based online news agency, had argued that officials at the Department of Finance breached its copyright in news articles when these articles were circulated internally. The decision is an important confirmation of the ‘right to read’ in Canada and may go some way to dispelling the aftertaste of an earlier flawed decision by the Ontario Small Claims Court in a similar dispute. Blacklock’s business model is to offer its news content on a subscription-only basis. Its articles are behind a paywall, and only subscribers, equipped with a password, can gain access to them. Individual subscriptions are available for $148 a year, whereas institutional subscription rates range between $11,470 and $15,670. In this case, a reporter from Blacklock’s had interviewed the President of the Canadian Sugar Institute, Sandra Marsden, for a story relating to sugar tariff changes. The same reporter had sought comments from Department of Finance officials and ultimately had an exchange of email correspondence with the Department’s media relations officer. In what appears to be Blacklock’s practice, teasers about the story were sent out to Marsden by email and by Twitter. Based on the teasers Marsden became concerned about the accuracy of the article. She paid for an individual subscription in order to access it. After reading the article her concerns grew and she cut and pasted the article into an email, to a Department official. The same reporter wrote a follow up piece which Marsden also found problematic; she forward this piece to the Department of Finance as well. The two articles were circulated between a total of 6 Finance employees who discussed amongst themselves whether any follow-up with Blacklock’s was required. In the end it was decided that the matter should be dropped. Justice Barnes found that there was no disputing that the Finance officials had used Blacklock’s copyright-protected material without paying for it or seeking Blacklock’s consent. The key issue was whether the use fell within the fair dealing exception for research or private study in s. 29 of the Copyright Act. After reviewing the Supreme Court of Canada’s landmark fair dealing decision in CCH Canadian v. Law Society of Upper Canada and its more recent decision in SOCAN v. Bell Canada, he concluded that the use constituted fair dealing. He noted that, according to the case law, “research” does not have to lead to the creation of a new work of authorship; it can be ““piecemeal, informal, exploratory, or confirmatory”, and can be undertaken for no purpose except personal interest.” (at para 31) Justice Barnes found that the Finance officials “had legitimate concerns about the fairness and accuracy” of the reporting in the article. Her further found the internal circulation of the piece was justified on the basis that “[e]veryone involved had a legitimate need to be aware in the event that further action was deemed necessary”. (at para 35) He identified a number of considerations that influenced his conclusion that the officials’ dealing with the material was fair. He noted that the articles had not been obtained by illegal means such as hacking the website; rather, they had been provided by a subscriber to the site who had legally accessed them and had forwarded them for “a legitimate business reason”. (at para 36) The articles had been sent to the Finance officials and not solicited by them; they received limited circulation; and they were not republished or used for any commercial purpose. The court also found that the two articles were a tiny fraction of the content available from the Blacklock’s site. Further, Justice Barnes opined that “a finding of copyright infringement against a news source for the simple act of reading the resulting copy is likely to have a chilling effect on the ability of the press to gather information.” (at para 36). Justice Barnes also stated that “copyright should not be a device that serves to protect the press from accountability for its errors and omissions.” (at para 36). Blacklock’s had argued that its terms and conditions for access to its paywalled content had been breached when the material was forwarded to Finance officials, and that this breach should serve to negate a finding of fair dealing. Justice Barnes appeared sympathetic to this argument on its face, stating that it was a “relevant consideration” (though he did not state that it would necessarily be determinative). However, he cautioned that for this factor to be taken into account, the copyright owner would have to demonstrate that the user was aware of the terms and conditions and that the terms and conditions actually barred the conduct at issue. In this case, he found that none of the parties involved had either read or even been aware of Blacklock’s terms and conditions which were not readily part of the process for signing up for an individual subscription. He also found that the terms and conditions were not clear, stating: “On the one hand they seemingly prohibit distribution by subscribers but, on the other, they permit it for personal, or non-commercial uses.” (at para 42). Blacklock’s also objected that a finding of fair dealing would undermine its business model – selling online news through a subscriber-only paywall. Justice Barnes was not particularly sympathetic, noting that “All subscription-based news agencies suffer from work-product leakage.” (at para 45) Further, he stated that “whatever business model Blacklock’s employs it is always subject to the fair dealing rights of third parties.” (at para 45) At the same time, he noted that by so stating, he was not endorsing “blameworthy conduct in the form of unlawful technological breaches of a paywall, misuse of passwords or the widespread exploitation of copyright material to obtain a commercial or business advantage.” (at para 45) As I noted in an earlier comment on this case, the defendants had argued that Blacklock’s was engaged in copyright misuse and was acting as a kind of “copyright troll”. In fact, there are 9 other suits brought by Blacklock’s against the federal government on similar sets of facts. Noting that “there are certainly some troubling aspects to Blacklock’s business practices”, Justice Barnes nevertheless found it unnecessary to rule on the copyright abuse and trolling arguments in light of his findings on fair dealing. The other cases, which were stayed pending the resolution of this first dispute, may now end up being settled out of court. In the course of his decision, Justice Barnes referred to what occurred in this case as “no more than the simple act of reading by persons with an immediate interest in the material.” (at para 36) This right to read is fundamentally important in a society that values knowledge and the freedom of expression. The decision makes it clear that business models for content distribution cannot run roughshod over certain fundamental users rights.
Published in
Copyright Law
|
Electronic Commerce and Internet Law in Canada, 2nd EditionPublished in 2012 by CCH Canadian Ltd. Intellectual Property for the 21st CenturyIntellectual Property Law for the 21st Century: Interdisciplinary Approaches |