Appearance before the Standing Committee on Ethics, Accountability & Privacy
May 31, 2012

Good morning.  My name is Michael Geist.  I am a law professor at the University of Ottawa, where I hold the Canada Research Chair in Internet and E-commerce Law. I was a member of the National Task Force on Spam struck by the Minister of Industry in 2004 and I currently serve on the Privacy Commissioner of Canada’s Expert Advisory Committee.

I appear before this committee today in a personal capacity representing only my own views.

My opening comments will identify several areas for potential government action, but I want to provide a bit of context with three key caveats.

First, it may be stating the obvious, but social media is an enormously important, positive development. The number of users is staggering and its role as a key source for communication, community, and political activity grows by the day. The opportunities presented by social media should be embraced, not demonized and government should be actively working to ensure that it incorporates social media into its policy consultation processes.

Second, Canada has been a leader in the use and regulation of social media. The Privacy Commissioner of Canada was the first to conduct a major privacy investigation into Facebook and has led on other issues involving social media and Internet companies.

Third, while we have had some influence through those investigations, Canada has not led in creating the social media services used by millions around the world. The failure to articulate and implement a national digital economy strategy comes back to haunt us in these circumstances, where the ability to place an unmistakable Canadian stamp on social media is undermined by the policy failures that have done little to encourage the development of Canadian social media sites.

What Is There to Be Done?

With those caveats, what is there to be done?  I'd like to point to four areas and issues.

1. Finish What We've Started

The government has introduced and even passed legislation that can be helpful in addressing some concerns that arise from social media, yet these initiatives have stalled short of the finish line. Anti-spam legislation, which received royal assent in 2010, has still not taken effect as final regulations have not been approved. In fact, Industry Canada officials now indicate it could go well into 2013 before the regulations take effect. Given the amount of work that went into the legislation, it is shocking that the law has been left in limbo.

Moreover, Bill C-12, the PIPEDA reform bill that seeks changes arising from the 2006 privacy review, lags in the House of Commons with seemingly no interest to move the bill forward. Indeed, the bill is now outdated and a full PIPEDA review to address emerging concerns such as order making power, statutory damages, and tougher security breach requirements than those found in the bill is needed. In fact, the C-12 security breach reporting rules are primarily bark with little bite given the absence of penalties for failure to comply.

The government has also promised a digital economy strategy for years and has failed to deliver. The strategy has come to be known as the Penske File, a reference to the Seinfeld episode that involves working on an imaginary file. While other countries are now years into implementing their strategies, Canada still lags behind.

It should also be noted that these issues must increasingly be addressed in concert with the provinces. The line between federal and provincial jurisdiction on many of these issues is blurry and legal challenges against federal legislation is a real possibility. Work is needed to begin to develop minimum standards that can be implemented at the provincial level should federal leadership be challenged in the courts by companies seeking to circumvent their privacy obligations.

2.    Devil is in the Defaults

In many respects, social media and Internet companies are the most powerful decision makers when it comes to privacy choices. As my colleague Ian Kerr says, the devil is in the defaults. In other words, the choices made by leading social media companies with respect to default privacy settings are the defacto privacy choices for millions of users. Given the increasing pressure to generate revenues, we can expect those default choices to change in aggressive ways to make greater use of user data.

There are examples of companies doing great work in the area. Twitter recently implemented do-not-track options that won plaudits from the Federal Trade Commission in the U.S. Similarly, Google offers its users transparency tools so that they can obtain detailed information about what information is collected, how it is used, and how they can modify their privacy choices. The company has also been transparent about law enforcement requests for information and copyright takedown demands. 

There needs to be continued work on these defaults, initiatives to provide users with greater information and transparency, and steps to ensure that companies live by their privacy commitments.

3.    Lawful Access

The introduction of Bill C-30 brought with it an avalanche of public outrage and concern over the proposed Internet surveillance legislation. While much of the focus was on the mandatory warrantless disclosure of subscriber information by telecom service providers, the potential for social media and big data Internet sites to serve much the same purpose cannot be overlooked.

A recent investigation by the Privacy Commissioner of Canada into Nexopia, a Canadian social network, identified hundreds of law enforcement requests for customer name and address information, frequently for accounts that should have been deleted months earlier.

Social media generates a treasure trove of personal information that must enjoy full privacy protection and court oversight before disclosure. Indeed, documents recently obtained under Access to Information indicate that Public Safety is thinking about how the rules are applied to social media sites and services. Bill C-30 needs to go back to the drawing board to effectively account for these issues.

4.    New Legal Issues

While I think that much can be done to use or augment existing rules, social media and the Internet do raise some unique issues that may require targeted responses.  In the interest of time, let me name two.  First, do-not-track. As you may know, cookies can be used to trace the web browsing habits of users, including when they visit third party sites. For example, Facebook inserts a cookie on user browsers that traces your activities as you surf the net. Any site with nothing more than a Facebook "like" button - as found on Conservative, NDP, and Liberal sites - means that Facebook records a visit to the site and retains the information for months.

A growing number of sites, including Yahoo, AOL, and Twitter, respect functionality found in Firefox browsers that allow a user to choose not to be tracked. Google has said it will implement similar technology in its Chrome browser.

However, many sites have been slow to adopt do-not-track and Facebook has thus far declined to do so.  Given the failure of industry to self-regulate, it is appropriate for government to step in with stronger measures to ensure that user choice is respected.

Second, there is a growing problem of social media misuse. For example, in recent months there have been an increasing number of stories of employers requiring employees to provide their Facebook user id and password as a condition of a job interview. Seeking the same information with direct questions would typically be prohibited, so this is used as a circumvention of longstanding standards and principles within employment law. In response, the State of Maryland has passed a law banning employers from requiring employees or job applicants to provide access to their personal digital/social media accounts. Several other states are working on similar legislation and Canada should consider following suit.

For example, in December 2010, the Australian Government's Productivity Commission, which is the government's independent research and advisory body on a range of economic, social and environmental issues affecting the welfare of Australians, released a report on the impact of bilateral and regional trade agreements. The report concluded:

The Commission considers that Australia should not generally seek to include IP provisions in further BRTAs, and that any IP provisions that are proposed for a particular agreement should only be included after an economic assessment of the impacts, including on consumers, in Australia and partner countries. To safeguard against the prospect that acceptance of ‘negative sum game’ proposals, the assessment would need to find that implementing the provisions would likely generate overall net benefits for members of the agreement.

In fact, the report was critical of the inclusion of IP in prior agreements as well:

The Commission is not convinced, however, that the approach adopted by Australia in relation to IP in trade agreements has always been in the best interests of either Australia or (most of) its trading partners. Among other things, there does not appear to have been any economic analysis of the specific provisions in AUSFTA undertaken prior to the finalisation of negotiations, nor incorporated in the government’s supporting documentation to the parliament. As noted above, the AUSFTA changes to copyright imposed net costs on Australia, and extending these changes to other countries would be expected to impose net costs on them, principally to the benefit of third parties.

Similar views were expressed in New Zealand in December 2010, when a leaked government document on the Trans Pacific Partnership expressed concern with the shift toward international pressure on IP with limited consideration of the domestic effects:

Traditionally, international intellectual property treaties have been based on two basic rules: the provision of substantive minima obligations and national treatment (see for example the Paris and Berne Conventions). The substantive minima have been periodically amended, mostly to require greater or different levels of protection. While traditionally, recognising the principles of self determination and diversity, a large part of law making would be left to national laws, stronger levels of international IP protection coupled with a larger coverage of issues is increasingly limiting the capacity of states to determine the optimal level and form of intellectual property protection, hence limiting room for cultural and economic difference and maximisation of social utility.

These developments are underpinned by an increasing pressure from rights holders to internationalise a larger array of issues and find international solutions to issues that have only had limited consideration at the national level. This is particularly true for the area of copyright, where rights holders have been seeking the adoption of more intrusive international rules with respect to a range of copyright issues at an early stage of norm development.

While there may be good arguments for an early and more rapid development of international solutions, such approach bears the risk of being premature, lacking evidence at the national level that would suggest a desirable international norm. 

Last year, the Canadian Standing Committee on Canadian Heritage adopted recommendations that were also critical of the inclusion of intellectual property within trade negotiations:

The Committee calls on the Government of Canada to ensure that domestic copyright policies are not part of any present or future trade negotiations; that Canada’s commitments to the implementation of the Anti-Counterfeiting Trade Agreement (ACTA) are limited to the agreement’s focus on combating international counterfeiting and commercial piracy efforts; and that the Government of Canada retains the right to maintain domestic copyright policies that have been developed within the framework of its commitments to the World Intellectual Property Organization and the Berne Convention.

The trend is global: in the United States, Senator Ron Wyden has proposed legislative language requiring Congressional approval of ACTA and requiring disclosure of the TPP intellectual property chapter. ACTA countries such as Mexico and Switzerland have not signed the agreement with a Mexican Senate resolution opposed to ACTA and the Swiss indicating that they will forgo signing for the moment. Most recently, the TPP negotiations included openly critical comments from several delegations who are increasingly opposed to the U.S.-imposed IP rules within the agreement.

The implications of this backlash is significant as it points to increasing discomfort with the inclusion of intellectual property chapters within large scale trade agreements. Indeed, intellectual property is invariably one of the major stumbling blocks within these agreements - whether the inclusion of the Internet provisions in ACTA, the TPP IP chapter, and the Canada-European Union Trade Agreement which is facing a major backlash over the IP rules. Note that these are not "anti-IP countries" but rather countries that recognize that trade agreements frequently do not yield intellectual property rules that serve their national interest. The development of global IP norms is important, but it belongs in open, transparent, inclusive multilateral institutions, not secretive trade deals like ACTA, the TPP, and CETA. meta http-equiv=

Given the trajectory of open access growth, the continual growth of materials available through site licences, and posting of materials online (subject to a Bill C-11 exception), newer materials frequently fall outside the Access Copyright licence. In other words, the size of the Access Copyright repertoire may be growing, but a sizable percentage of the new works are available through alternative means and therefore do not require an Access Copyright licence.

The one area where the Access Copyright licence might theoretically offer some value would be for older materials that are not readily available in digital form (and not typically available under open access or site licences).  But according to Access Copyright itself, older works are not likely to be copied under its licences. In its 2012 Payback FAQ to authors, the collective asks authors to only list works published within the last 20 years, noting:

Q. Why are you only asking for works published within the last 20 years?

A. Our statistical analysis of copying data shows that works published more than 20 years ago are unlikely to be copied under our licences.

This admission from Access Copyright shows how its repertoire is declining in value for licensees even as its size increases. A growing percentage of newer materials are available by alternative means, while the older materials may not be subject to an alternate licence, but they are unlikely to be copied. Over the coming years, the Access Copyright squeeze is only going to grow as the entire repertoire of materials likely to be copied - the materials published within the last 20 years - are all published in the digital/Internet era and available through alternative means. This unfolding scenario helps explain why Access Copyright jumped at the chance for a huge increase in fees that come with the model licence and why it is the wrong deal at the wrong time for universities.

What if the government had not rejected requests from leading cable companies such as Rogers and Shaw, who fear that the bill will block their ability to introduce network PVRs into the Canadian market and restrict innovative cloud-based computer services?

What if the government had not rejected requests from the Documentary Organization of Canada for a specific digital lock exception for documentary film making, given that U.S. film makers benefit from such an exception and that Canadians will be placed at a competitive disadvantage?

What if the government had not rejected requests from many Canadian library associations, who argued that new digital inter-library loans provisions are unusable in light of requirements to establish onerous restrictions limiting the use of works?

What if the government had not rejected proposed amendments from NDP, Liberal and Green Party MPs to add flexibility to the digital lock rules so that concerns that the approach trumps fair dealing and education rights could be addressed?

What if the government had not rejected concerns from teachers, who lamented new distance learning provisions that include requirements to destroy lessons 30 days after a course concludes?

What if the government had followed the lead of innovative economies such as South Korea and Israel, which both recently adopted U.S.-style fair use rules that are often credited with forming the copyright engine behind cutting-edge business models?

What if the government had not rejected an amendment to require labeling of digital locks so that consumers would have advance warning of the restrictions that come with the products they purchase?

What if the government had not rejected a proposal from the Canadian Federation for the Humanities and Social Sciences to eliminate crown copyright and give a boost to its open government data initiative?

What if the government had not rejected a plan raised by Project Gutenberg to create a legal safe harbour provision for the use of works where the copyright owner is unknown, thereby increasing digitization of Canadian materials?

What if the government had not rejected the advice of provincial ministers of education who called for an expansion of fair dealing to cover copies for classroom use as is the case in the U.S.?

What if the government had not rejected the Retail Council of Canada's recommendation to eliminate the iPod tax and thus avoid a hearing later this year on extending the private copying levy to microSD cards?

No bill is ever perfect and Bill C-11 is no exception. Yet as the government touts its success in navigating the challenging copyright waters to pass a bill after multiple false starts, it might temper its enthusiasm by recognizing that provisions that lock out the blind, create disadvantageous barriers for creators and educators, and stifle innovation are not a cause for celebration.

More relevant is the content of those submissions. As part of the 2009 copyright consultation, opposition to the digital lock approach found in Bill C-61 was the number one issue, yet the government rejected the views of thousands of Canadians who called for a more balanced, flexible approach. Similarly, the Bill C-32/C-11 committee consistently heard from business groups, creator associations, consumer groups, and education representatives, who all opposed the digital lock approach in the bill. More recently, Regan told the House of Commons that the most of the 80,000 emailed submissions also opposed the digital lock approach.

Given the consistent opposition to the digital lock approach, both the NDP and Liberals proposed numerous amendments to the digital lock rules, all of which were defeated. Those were followed by further digital lock amendments proposed by the Green Party's Elizabeth May, which were also defeated.

Paradis and Canadian Heritage Minister James Moore (who in the same Hill Times article is quoted as inaccurately saying that Canadian copyright has not been changed for 22 years) may point to the large number of Canadians that participated in the copyright debate over the past few years, but it is more accurate to acknowledge the large number of Canadians whose views the government rejected in adopting a digital lock approach in which the voice of one consultation - that with the United States - proved more influential than anything tens of thousands of Canadians had to say.

1.   The Licence is Unnecessary

As many have pointed out, the universities already pay for access to a wide range of materials that likely cover the majority of copyrighted materials used on campus. Campus wide electronic database licences offer access to thousands of journals and electronic books that can be incorporated directly into electronic coursepacks. Universities pay millions of dollars for these licences with the money flowing to database companies, publishers, and authors. For example, last year the Canadian Research Knowledge Network alone spent over $96 million in content licences that offer access to millions of articles to 900,000 students and researchers at 75 universities across the country.

Moreover, open access licensing, where research publications are freely available online, constitutes a growing percentage of published research (some estimate it at 30 percent of all scholarly research), with thousands of open access journals and hundreds of thousands of articles posted directly by the researchers themselves. As Cory Doctorow might say, there will never be less work available under open access than there is today.

Add public domain works, fair dealing, hundreds of millions spent on textbooks, and pay-per-use licences for the remaining works and an Access Copyright licence simply becomes unnecessary.

2.   The Licence Does Not Adequately Account for Bill C-11

One of the primary concerns with the Access Copyright model licence is the failure to account for forthcoming changes to Canadian copyright law under Bill C-11. The bill seems likely to become law by the summer and it includes several important provisions for Canadian education institutions.

First, the expansion of fair dealing to include a new category of education expands the scope of permitted copying without the need for permission or compensation. While the new education category will not mean that all copying on university campuses is covered by fair dealing (as some in the publishing community have misleadingly claimed), it will obviously include some educational copying that may not have been covered by research or private study.

Second, the inclusion of an exception for publicly available materials on the Internet covers the content found on millions of websites that can now be communicated and reproduced by educational institutions without the need for permission or compensation. The provision states:

30.04 (1) Subject to subsections (2) to (5), it is not an infringement of copyright for an educational institution, or a person acting under the authority of one, to do any of the following acts for educational or training purposes in respect of a work or other subject-matter that is available through the Internet:

(a) reproduce it;
(b) communicate it to the public by telecommunication, if that public primarily consists of students of the educational institution or other persons acting under its authority;
(c) perform it in public, if that public primarily consists of students of the educational institution or other persons acting under its authority; or
(d) do any other act that is necessary for the purpose of the acts referred to in paragraphs (a) to (c).

The subsections that follow create several conditions, including attribution, the absence of a digital lock, and the absence of a clear opt-out notification (that is more than just a copyright notice). As I noted in an earlier post, this provision is certainly applicable to linking to online content and will also cover many other online materials used in the classroom.

Third, the non-commercial user generated content provision may also prove relevant for some electronic casebooks that incorporate materials to create a new work for non-commercial purposes. The UGC provision (Section 29.21) includes four conditions including an analysis of "substantial adverse effect", but the provision may allow for the development of new materials where the evidence suggests that the new works do not substantially adversely affect the original works.

Fourth, the bill establishes a technology-neutral approach for the reproduction of materials for display purposes. The current law is limited to manual reproduction or on an overhead projector. With Bill C-11, the provision applies to any display technology:

It is not an infringement of copyright for an educational institution or a person acting under its authority for the purposes of education or training on its premises to reproduce a work, or do any other necessary act, in order to display it.

There are limits to this exception (it applies where the work is not commercially available in a medium that is appropriate for the purpose referred to in the exception), but it still may cover uses that Access Copyright would prefer to licence. For example, concerns about the inclusion of copyright-materials in powerpoint presentations may be covered by the new display provision.

Fifth, the digital inter-library loans provision, while restrictive, will open the door to digital transmission of materials on an inter-library basis, increasing access to materials that have been acquired by university libraries.

Sixth, the bill features changes to the statutory damages provision that decrease the risk of significant liability for educational institutions. Those changes are discussed further below.

All of these provisions represent significant changes in the law that provide new rights for education, yet the model licence seemingly acts is if they do not exist.

3.   The Legal Risk of Not Signing is Limited

The University of Victoria notes in its announcement that it is reluctantly signing the model licence stating "the decision to opt into the license was taken reluctantly as a strategy to mitigate the risk of litigation by Access Copyright in the immediate future." Yet the legal risk is already very limited. First, the Supreme Court of Canada's CCH decision, still the leading decision on fair dealing, speaks of the need for a "large and liberal" interpretation to fair dealing categories such as research "to ensure that users’ rights are not unduly constrained." Fair dealing certainly provides considerable latitude for copying at Canadian universities and would be the first line of defence against a claim of infringement. Indeed, the recent Georgia State University fair use case in the United States demonstrates how user rights/copyright exceptions can be used as an effective defense against many claims of infringement (Ariel Katz argues that Canadian defences would be even stronger than those in the U.S.).

Second, there remain considerable doubts about the Access Copyright repertoire, both with respect to its scope and the need for evidence of copyright ownership or rights in specific works. Access Copyright is viewed by many as vulnerable on both of these fronts, creating significant legal risk for the collective should it choose to pursue litigation against a university operating outside the model licence.

Third, Bill C-11's statutory damages reforms limit the potential liability for non-commercial infringement. The bill will establish a cap of $5,000 maximum for all infringements involved "if the infringements are for non-commercial purposes." Given the non-commercial status of educational institutions and the absence of any profit motive in these cases, the new statutory damages cap may be applicable. While Access Copyright could seek actual damages, the Georgia State University case demonstrates that actual damages may be even lower in many cases.

Fourth, concerns that universities will offload copyright liability onto faculty are unfounded. Universities rightly advise faculty to follow appropriate copyright guidelines to ensure that their policies are consistent with the law. However, should a faculty member overreach with their copying practices, it is incredibly unlikely that they will face a lawsuit (and even if they did, litigation insurance would cover the costs). The copying involved would be so small that there simply are insufficient damages to justify a lawsuit. Consider a faculty member that makes 50 or 100 copies of an article for their students. Leaving aside all the available exceptions, the damages from the copying would be tiny (capped at $5000 but likely to be much, much less). Given the legal costs and risks associated with losing such a suit (imagine a court ruling that education under fair dealing covered those classroom copies), these are lawsuits that will not happen.

While none of this suggests that universities can simply copy what they like without addressing the issue of permissions (they clearly cannot), assuming a university has developed an appropriate copyright policy as well as invested in site licences and the necessary copyright clearances, the risk of liability should Access Copyright sue is limited.

4.   The Licence is Inequitable

The Access Copyright decision to blend two separate fees into one creates significant inequalities among students. At the moment, all students pay $3.38 annually to cover general copying on campus (the interim tariff) and an additional 10 cents per page for coursepacks. The new licence establishes a single fee of $26 per student. This creates a huge jump in costs for the majority of students. The majority of faculties - sciences, law, health, medicine, dentistry, and engineering - make very little use of the Access Copyright licence for coursepacks since their materials are typically either texts, available under open access, or can be accessed through alternative licensed databases (as is the case in law). 

For those students, the only fee they currently pay is the $3.38 per year. If their university signs the model licence, they will be paying $26 next year for next to nothing. The incidental or general copying is almost certainly covered by fair dealing and these students don't use coursepacks with materials necessitating an Access Copyright licence. It is inequitable to compel these students to pay additional fees with no value in return, particularly as some of these students already pay the highest student fees on campus.

5.   The Licence is Harmful

Some institutions may believe signing the agreement is the best way to limit their legal risk, yet the reality is that the licence is not neutral in the sense that it simply results in higher fees for students (or costs for the institution) in return for reduced legal liability.  The model licence is harmful in two important respects.  First, as CAUT has pointed out, the licence contains very problematic language that raises the prospect of surveillance as well as restraints on scholarly communication, use of scholarly materials, and use of modern technologies. 

Second, providing Access Copyright with millions in additional funds that may not be necessary will serve to undermine educational concerns in current and future copyright reform. During the Bill C-32/C-11 debate, Access Copyright was the leading opponent of educational interests, opposing virtually all educational exceptions in the bill. In fact, Access Copyright spent millions last year (over 30% of its licensing revenues) on Copyright Board applications and professional fees, including lobbying on Bill C-11. When universities provide a windfall of millions to Access Copyright, much of that money is funneled toward advocacy opposing education concerns.

6.   The Licence Hurts Long Term Education

As I discussed in a column last week, the emerging education model flips the current approach of expensive textbooks, closed research, and limited access to classroom-based learning on its head, instead featuring open course materials, open access to scholarly research, and Internet-based courses that can simultaneously accommodate thousands of students. Over the past year, dozens of universities have spent thousands of dollars in shifting toward a more open approach by investing in open access and open educational resources.

The University of Victoria claims that it will continue that effort so that it can adopt an open access approach in 2015. Yet many universities signing the Access Copyright model licence will simply write-off the recent investments, particularly given the huge increase in costs they will face. The long-term impact on those schools will be significant. Rather than focusing on open materials and greater flexibility for faculty and students, they will lock into the Access Copyright model of high fees and limited rights to use course materials. As a model for the future, the model licence is a major step backward.

• The Interception Regulations, including specific details on interception capabilities (as many as 200 simultaneous interceptions), response time to interception requests (30 minutes for remote interceptions), confidentiality requirements, transmission capabilities (real time transmission of intercepted communications), and delivery of intercepted communications.
• The Interception Equipment Regulations, including very specific capabilities for simultaneous interceptions including multiple targets and providing intercepted communications to up to five different agencies at the same time. These regulations also identify requirements on service providers to increase their capacity (up to five business days).
• The Subscriber Information Disclosure Regulations, including the form to be used to request subscriber information (which can come from police, the Competition Bureau, or CSIS). These also discuss the concept of law enforcement providing at least one identifier (ie. a name, email address or IP address) in order to receive the other corresponding subscriber information. There are also confidentiality requirements and details on telecom provider record keeping. The regulations also identify timing requirements for disclosure, typically within two business days but within 30 minutes in exceptional circumstances.
• The Other Obligations Regulations, including location information disclosures that may require telecom companies to disclose location information such as street address, longitudinal and latitudinal coordinates or cell site. It is not clear whether such information would require a warrant. These regulations also will provide details on assisting law enforcement in testing equipment, the special rules for smaller providers, and categories for administrative monetary penalties for failing to comply with the law.
• The Payment to Providers Regulations identify when telecom providers will be compensated by law enforcement. These include (1) complying with a Ministerial Order to obtain equipment, software, or to modify existing equipment; (2) provide telecom support related to interceptions; and (3) providing subscriber information.

While the actual regulations may change, it is shocking that Public Safety has provided this information to dozens of companies but kept it secret from the Canadian public. The secrecy associated with the lawful access initiative certainly further undermines trust in Bill C-30 and highlights the need to scrap the bill and the two-tier policy process and start from scratch.

The secret working group includes virtually all the major telecom and cable companies, whose representatives have been granted Government of Canada Secret level security clearance and signed non-disclosure agreements. The group is led by Bell Canada on the industry side and Public Safety for the government.

The inaugural meeting, held just three weeks before Bill C-30 was introduced, included invitations to eleven companies (Bell Canada, Cogeco, Eagle, MTS Allstream, Quebecor, Research In Motion, Rogers, Sasktel, Telus, Vidéotron, and Wind Mobile) along with two industry associations (the Canadian Wireless Telecommunications Association and the Canadian Network Operators Consortium).

The secret working group is designed to create an open channel for discussion between telecom providers and government. As the uproar over Bill C-30 was generating front-page news across the country, Bell reached out to government to indicate that "it was working its way through C-30 with great interest" and expressed desire for a meeting to discuss disclosure of subscriber information. A few weeks later, it sent another request seeking details on equipment obligations to assist in its costing exercises.

Months before the January 2012 meeting, officials worked with the telecom companies to identify many concerns and provide guidance on the government's intent on Internet surveillance regulations, information that has never been publicly released.

For example, a December 2011 draft list of lawful access issues features questions about surveillance of social networks, cloud computing facilities, and Wi-Fi networks. The telecom companies raise many questions about compensation, such as "a formula for adequate compensation" for the disclosure of subscriber information as well as payment for testing surveillance capabilities and providing surveillance assistance.

At a September 2011 meeting that included Bell Canada, Cogeco, RIM, Telus, Rogers, Microsoft, and the Information Technology Association of Canada, government officials provided a lawful access regulations policy document that offered guidance on plans for extensive regulations that will ultimately accompany the Internet surveillance legislation.  

The 17-page document indicates that providers will be required to disclose certain subscriber information without a warrant within 48 hours and within 30 minutes in exceptional circumstances. Interceptions of communications may also need to be established within 30 minutes of a request with capabilities that include simultaneous interceptions for five law enforcement agencies.

The close cooperation between the government and telecom providers has created a two-tier approach to Internet surveillance policy, granting privileged access and information for telecom providers. Meanwhile, privacy and civil society groups, opposition MPs, and millions of interested Canadians are kept in the dark about the full extent of the government's plans. The public has already indicated its opposition to the bill. The secrecy and backroom industry talks associated with Bill C-30 provides yet another reason to hit the reset button.