Stopping online piracy by Anthony Farris (CC BY-NC-ND 2.0) https://flic.kr/p/bf4uTz

Stopping online piracy by Anthony Farris (CC BY-NC-ND 2.0) https://flic.kr/p/bf4uTz

News

The Case Against the Bell Coalition’s Website Blocking Plan, The Finale

Nearly one month ago, I set out to outline the case against the Bell coalition’s website blocking plan. Sixteen instalments later (plus bonus posts on Bell’s astroturfing campaign and the remarkable success of the day of action opposing the plan), I have examined the myriad of problems with the proposal. The objective was never to justify piracy. Rather, it was to conclusively demonstrate that the proposal is disproportionate, harmful, offside international standards, violates Canadian norms, and does not come close to meeting the CRTC’s requirements for approval of website blocking. This post summarizes the key points for each of these five sources of concern. The CRTC is accepting interventions until March 29th, leaving Canadians with several more weeks to speak out to the Commission, their Member of Parliament, and the Innovation, Science, and Economic Development Minister Navdeep Bains.

i.    The Plan is a Disproportionate Response to Piracy Concerns

As ISED Minister Navdeep Bains rightly noted when the Bell coalition filed its proposal with the CRTC, Canada already has “numerous legal provisions and tools to help copyright owners protect their intellectual property.” Indeed, Canada has some of the world’s toughest anti-piracy provisions, which Bell and others have actively used in recent years. This includes lawsuits against set-top box distributors, mod-chip sellers, and websites such as TVAddons. Some of these lawsuits have resulted in massive damage awards running into the millions of dollars. Further, Canadian copyright law has also been used to shut down websites whose primary purpose is to enable infringement with rights holders relying on an “enabler provision” contained in the 2012 copyright reforms that can be used to target online sites that provide services primarily for the purpose of infringement. Copyright owners are seeking to create their own system at the CRTC without direct court involvement or policy review by Parliament. Before entertaining such a possibility, they should surely be required to test the effectiveness of existing law.

Moreover, the piracy evidence in Canada remains inconclusive at best as the data consistently shows that Canada is not a global leader when it comes to piracy. This includes data from Music Canada on stream ripping, from CEG-TEK on the impact of anti-piracy notices, and from the Business Software Alliance on software piracy rates. Even the MUSO report, which provides the key data points for the Bell’s coalition’s proposal, plainly states that Canadian piracy rates declined during the study period.

The limited impact of Canadian piracy is reflected in the data on the digital economy and Canadian creative sector, which show a thriving industry. The total value of the film and television sector exceeded $8 billion last year, over than a billion more than has been recorded over the past decade. In fact, last year everything increased: Canadian television, Canadian feature film, foreign location and service production, and broadcaster in-house production.

The Canadian data on digital business models also points to a steady stream of success stories that refute claims that it difficult if not impossible to create successful business models in Canada. Online video services, which the Bell coalition suggests are harmed by streaming sites, are experiencing rapidly expanding revenues, now generating more than $1 billion per year. In fact, two Canadian online video services – CraveTV and Club illico – are estimated to have earned $373 million last year, up from just $13 million four years earlier. The Canadian success story is not limited to online video as the online music market has experienced similar growth. According to industry data, the Canadian music market is growing much faster than the world average (12.8% in 2016 vs. 5.9% globally), streaming revenues more than doubled last year to US$127.9 million (up from US$49.82 million) growing far faster than the world average of 60.4%, the Canadian digital share of revenues of 63% is far above the global average of 50%, and Canada has leaped past Australia to become the 6th largest music market in the world.

The lack of proportionality stems not only from the limited impact of piracy in the Canadian market, but also from evidence that site blocking is far less effective than its proponents claim. For example, INCOPRO, which sells site blocking services including lists of sites to block (and therefore has an obvious vested interest in promoting their effectiveness) has issued several studies on blocking. A 2017 INCOPRO study on the effectiveness of Australian website blocking points to reduction in piracy rates but also examined usage of a list of 250 unauthorized sites:

Usage of the top 250 sites in Australia decreased by 4% (204,843) when comparing March 2017 to October 2016. Usage of the same sites reduced by 13% for the global (excluding Australia) group and by 10.8% for the global control group.

The study attributes the fact that Australian declines with site blocking were lower than global averages by acknowledging that “there may have been an increase in the usage of some unblocked sites as a result of the most popular site being blocked.” Further, INCOPRO studies contain a key exclusion:

General purpose VPN and proxy services have been excluded because they allow users to access any website of their choice. As a result, it cannot be definitively concluded that they are being used to access unauthorised sites.

Given that it does not account for users who shift their Internet usage to VPNs, the reliability of the data is questionable. This post identifies many more studies and court rulings that have concluded site blocking has limited long-term impact.

ii.    The Plan is Offside International Standards

One of the most obvious problems with the site blocking plan – indeed one that is fatal – is the absence of court orders for website blocking. The attempt to avoid direct court involvement means the proposal suffers from an absence of full due process, raising a myriad of legal concerns. If adopted, the coalition website plan would put Canada at odds with almost everyone since the data is unequivocal: the overwhelming majority of countries require a court order for site blocking.

Just how rare is non-court ordered blocking? Working with Amira Zubairi, a University of Ottawa law student, we examined 22 countries that have or have had some form of copyright-related website blocking. Some groups say that there are 27 countries with website blocking, but we excluded five countries due to widespread censorship: Saudi Arabia (which features government-backed Internet blocking), Indonesia (which has blocked 800,000 sites), Malaysia (which regularly uses the power to block legitimate sites), Turkey (which uses real-time large scale blocking of sites including Wikipedia) and South Korea (which uses censors to block access to thousands of web pages). Our research shows that of the 22 countries that have site blocked for copyright purposes, 20 use or have used court orders (the exceptions are Italy and Portugal). The full comparative details can be found here.

Notwithstanding assurances that there are many systems that do not depend on court orders, the reality is that almost everyone with a free and open Internet only engages in the possibility of website blocking with a court order. The failure to include one – indeed the very point of the Bell coalition proposal seems to be to avoid the courts – would put Canada at odds with almost all our allies and likely be subject to an immediate legal challenge given our emphasis on openness, net neutrality, and due process.

iii.    The Plan Would Create Serious Harms to Internet Access

The widespread opposition to the Bell coalition website blocking stems in large measure due to the many harms it would be create for those accessing online content and the Internet itself. Chief among them is the likelihood of blocking access to legitimate content and creating a framework that steadily expands in scope. The danger of over-blocking legitimate websites raises serious freedom of expression concerns, particularly since experience suggests that over-blocking is a likely outcome of blocking systems. The Council of Europe Commissioner for Human Rights issued a report in 2014 on the rule of law on the Internet in the wider digital world, noting:

blocking is inherently likely to produce unintentional false positives (blocking sites with no prohibited material) and false negatives (when sites with prohibited material slip through the filter). From the point of view of freedom of expression, the most problematic is widespread over-blocking: the blocking of access to sites that are not in any way illegal, even by the standards supposedly applied.

One of the best-known cases of over-blocking arose in Canada in 2005, when Telus unilaterally blocked access to a pro-union website without a court order during a labour dispute. In doing so, it simultaneously blocked access to an additional 766 websites hosted on the same computer server.

The real danger is that this is not ancient history. Working with University of Ottawa law students Tanvi Medhekar and Matt Westwell, we identified numerous instances around the world in recent years where anti-piracy blocking resulted in over-blocking of legitimate sites. For example, in 2013, UK ISPs blocked access to around 200 legitimate websites including Radio Times. The blocking occurred as a result of a court order targeting two file sharing websites. There have been many similar instances in the UK including the 2012 blocking of the Promo Bay and the 2015 blocking of CloudFlare customers. In fact, OFCOM, the UK regulator, anticipated the over-blocking issue in 2010 study that noted:

We believe that IP address based site blocking is not granular and is likely to lead to over-blocking. This may undermine the confidence in any site-blocking scheme, and create significant liability risks for service providers. The over blocking property is a by-product of sites sharing IP addresses.

The report noted risks of over-blocking with all technical approaches to site blocking. This post provides further detail on how the UK experience has been replicated in other countries.

Over-blocking becomes even more likely as the definition of “piracy” expands and the blocking extends to other issues. For example, the use of VPNs, which enhance privacy but also allow users to access out-of-market content, has been sore spot for the companies for many years. VPN services are already targeted by IP lobby groups such as the IIPA and can be expected to face demands for blocking (similar to the way Netflix and Hulu have cracked down on VPN use). Beyond VPNs, it would not be surprising to find legitimate services streaming unlicensed content as the next target. With Bell characterizing accessing U.S. Netflix as stealing, the company may call for blocking of content from foreign services without Canadian rights.

If the CRTC were to create a system for mandated website blocking of intellectual property issues, it would surely soon face requests for far more. For example, the first request for mandated website blocking involved a request in 2006 from Richard Warman to block two hate sites. Warman provided the Commission with expert evidence that the sites violated the Criminal Code. The CRTC refused to issue the order, noting that it did not think it had the legislative power under Section 36 to issue blocking orders. With the floodgates opened, however, hate speech sites would quickly give way to online gambling and other regulated activities.

The harms from a site blocking system involve more than just blocking legitimate content. The costs associated with site blocking with can run into the thousands of dollars per site with significant investment in blocking services, employee time to implement blocking mandates, and deal with associated service issues. A mandated blocking system applied to all ISPs in Canada would have an uneven impact: larger ISPs will face new costs but may find it easier to integrate into existing systems (some already block child pornography images), whereas hundreds of smaller ISPs would face significant new costs that would affect their marketplace competitiveness.

The cost concerns are not theoretical. Larger ISPs in the UK disclosed their approximate costs in a 2014 case. For example, Sky Broadband spent over 100,000 pounds to develop a website blocking system in 2011, solely for IP right infringing website injunctions and spent thousands more each month on monitoring costs. British Telecom spent over a million pounds on a DNS web-blocking system in 2012 and required more than two months of employee time on implementation. EE spent more than a million pounds on its web-blocking system and over 100,000 pounds every month for operations. While companies such as Bell would stand to gain from blocking with higher fees passed along to subscribers and reduced competition, smaller ISPs would face a difficult economic challenge, leaving all Canadians facing higher monthly Internet bills.

iv.    The Plan Violates Canadian Norms on Net Neutrality, Freedom of Expression, and Privacy

Of all the claims that accompanied the launch of the Bell coalition’s website blocking plan, the most audacious may be the repeated assurances that site blocking does not raise net neutrality issues. Given that the starting principle for net neutrality is the right for users to access content and applications of their choice, blocking content is prima facie a net neutrality violation. The fact that Bell argues that its site blocking plan does not implicate net neutrality should not come as a surprise. Bell, more than any other Canadian company, has spent more than a decade arguing that practically nothing is covered by net neutrality.

In this case, the Bell coalition argues that net neutrality is limited to “lawful content” and that its plan therefore falls outside the rules. In its application, however, it does not cite the Canadian rules. That too is unsurprising, since Canada’s net neutrality framework was never limited in application to content that is “lawful.” The 2009 CRTC net neutrality decision says the following about blocking:

The Commission notes that the majority of parties are in agreement that actions by ISPs that result in outright blocking of access to content would be prohibited under section 36 unless prior approval was obtained from the Commission. The Commission finds that where an ITMP would lead to blocking the delivery of content to an end-user, it cannot be implemented without prior Commission approval. Approval under section 36 would only be granted if it would further the telecommunications policy objectives set out in section 7 of the Act. Interpreted in light of these policy objectives, ITMPs that result in blocking Internet traffic would only be approved in exceptional circumstances, as they involve denying access to telecommunications services.

In other words, blocking may only be permitted under exceptional circumstances where it furthers the telecommunications policy objectives. There is no reference to lawful content. In fact, the word “lawful” does not appear in the decision.

The Bell coalition website blocking plan may violate more than just Canadian net neutrality rules. As currently framed, it may also violate human rights norms. Website blocking or other measures to limit access to the Internet raises obvious freedom of expression concerns that has sparked commentary from many international organizations. Frank LaRue, the former U.N. Rapporteur on Freedom of Expression, was one of several experts on freedom of expression, including representatives from the Organization for Security and Co-operation in Europe, the Organization of American States, and the African Commission on Human and Peoples’ Rights, who issued a joint declaration in 2011 on freedom of expression and the Internet. It states the following on blocking:

Mandatory blocking of entire websites, IP addresses, ports, network protocols or types of uses (such as social networking) is an extreme measure – analogous to banning a newspaper or broadcaster – which can only be justified in accordance with international standards, for example where necessary to protect children against sexual abuse.

In 2012, a further declaration from LaRue and the IACHR-OAS Special Rapporteur on Freedom of Expression states:

all restrictions on freedom of expression, including those that affect speech on the Internet, should be clearly and precisely established by law, proportionate to the legitimate aims pursued, and based on a judicial determination in adversarial proceedings. In this regard, legislation regulating the Internet should not contain vague and sweeping definitions or disproportionately affect legitimate websites and services.

International human rights rules and declarations leave the Bell coalition website blocking plan vulnerable to challenge in at least two respects: the absence of court orders and the proportionality of the measures relative to harm.

The Bell website blocking coalition cites privacy protection as a reason to support its plan, noting the privacy risks that can arise from unauthorized streaming sites. Even if Bell was as an exemplary company with respect to privacy protection (which it is not), its website blocking proposal would still pose significant privacy risks. First, the use of virtual private networks is an increasingly important mechanism for users to safeguard their privacy online. Since the response to site blocking from some Internet users will surely involve using VPNs to evade the blocks, the attempt to characterize VPNs as services engaged in piracy will only increase. Further, the identification of piracy sites and usage by subscribers depends upon snooping into Internet users’ online activities. In fact, certain website blocking technologies raise serious privacy concerns. An Ofcom review of site blocking noted:

To be successful, any process also needs to acknowledge and seek to address concerns from citizens and legitimate users, for example that site blocking could ultimately have an adverse impact on privacy and freedom of expression.

Rather than enhancing privacy protection, the Bell coalition proposal puts it at greater risk, with the possibility of VPN blocking, incentives to monitor customer traffic, and the potential adoption of invasive site blocking technologies.

v.    The Plan Does Not Comply With CRTC Rules For Approval of Site Blocking

Even if the CRTC were to still think this terrible idea is worth supporting, it would fall outside its stated rules on website blocking. The Commission has made it clear that it will only permit blocking in “exceptional circumstances” and only where doing so would further the objectives found in the Telecommunications Act. The Bell coalition website blocking proposal does not come close to meeting the standards established by the CRTC, applicable legislation, and the federal government.

The CRTC said the following in its September 2016 letter arising out of the Quebec law mandating the blocking of access to unlicensed gambling sites:

the Commission is of the preliminary view that the Act prohibits the blocking by Canadian carriers of access by end-users to specific websites on the Internet, whether or not this blocking is the result of an ITMP. Consequently, any such blocking is unlawful without prior Commission approval, which would only be given where it would further the telecommunications policy objectives. Accordingly, compliance with other legal or juridical requirements – whether municipal, provincial or foreign – does not in and of itself justify the blocking of specific websites by Canadian carriers, in the absence of Commission approval under the Act.

The Bell website blocking proposal must therefore do more than simply raise concerns with respect to copyright law or cultural policies found in the Broadcasting Act objectives. Rather, it must convince the CRTC that website blocking would further the telecommunications policy objectives.  The Bell coalition application cites three objectives in support: that piracy “threatens the social and economic fabric of Canada” (subsection a), that the telecommunications system should “encourage compliance with Canadian laws” (subsection h), and that website blocking “will significantly contribute toward the protection of the privacy of Canadian Internet users” (subsection i).

The Bell coalition case is exceptionally weak on all counts. As discussed in great detail earlier in the series, there is no compelling evidence that piracy is threatening the social and economic fabric of Canada. The argument around encouraging compliance with the law is even weaker as the Commission has already stated that compliance with other legal or juridical requirements does not justify site blocking. As for privacy, if anything, the proposal places privacy at risk.

In fact, not only does the Bell proposal fail to make the case that it furthers the Telecommunications Act objectives, but there is a far better argument that it undermines them:

  • Subsection (a) references the “orderly development throughout Canada” of the telecommunications system, but the creation of a blocking system applied to hundreds of ISPs and wireless carriers of all sizes across the country would undermine that goal as it would likely lead to the implementation of differing blocking technologies, inconsistent over-blocking of legitimate content, and a non-neutral Internet in Canada. Access to content could differ from ISP to ISP.
  • Subsection (b) identifies reliability and affordability as objectives, but both would be undermined by website blocking. The reliability of the telecommunications services would be harmed by over-blocking of legitimate content and by the violation of net neutrality norms. The goal of better affordability would be undermined by the increased costs that would be passed along to subscribers to fund site blocking technologies and services.
  • Subsection (c) focuses on competitiveness of telecommunications services, yet the website blocking plan would have an uneven impact: larger ISPs will face new costs but may find it easier to integrate into existing systems (some already block child pornography images), whereas hundreds of smaller ISPs would face significant new costs that would affect their marketplace competitiveness. In fact, while companies such as Bell would stand to gain from blocking with higher fees passed along to subscribers and reduced competition, smaller ISPs would face a difficult economic challenge, leaving all Canadians facing higher monthly Internet bills and reduced competition.
  • Subsection (f) emphasizes the need for efficiency and reliance on market forces. As discussed last week in a post on why the plan is inconsistent with the CRTC Policy Direction, website blocking could not be more inconsistent with that objective. Indeed, with courts around the world concluding that site blocking is a disproportionate remedy, evidence that it is likely to lead to overblocking and is ineffective, and that it risks violating net neutrality and privacy rights, the plan fails to meet the Telecommunications Act objective of “efficient and effective” regulation.
  • Subsection (h) speaks to the economic and social requirements of users of telecommunications services, but users of those services are largely absent from the proposal. Instead, broadcasters, broadcast distributors, movie theaters, and some creator groups seek to impose new restrictions on users in the form of a regulatory framework without court orders. If anything, the social requirements of users should include compliance with net neutrality, human rights, and privacy norms, a standard the Bell coalition proposal fails to meet.
  • Subsection (i) focuses on contributing to privacy protection, but rather than enhancing privacy protection, the Bell coalition proposal puts it at greater risk, with the possibility of VPN blocking, incentives to monitor customer traffic, and the potential adoption of invasive site blocking technologies.

Moreover, the Bell coalition proposal fails to comply with the policy direction to the CRTC on implementing Canadian telecommunications policy objectives. Despite years of insistence that the CRTC follow the policy direction, Bell has now proposed regulatory intervention that could not be more inconsistent with that direction. After invoking the policy direction at seemingly every opportunity, the Bell application suddenly goes silent with respect to the issue (the legal opinion references a separate aspect of direction to argue that non-economic regulations are permitted).

However, with courts around the world concluding that site blocking is a disproportionate remedy, evidence that it is likely to lead to overblocking, and risks that violates net neutrality and privacy rights, the plan fails to meet the policy direction’s requirement of “efficient and proportionate” regulation. Further, the evidence on the current state of the Canadian market with record revenues and Canadian production reinforces that market forces are working as consumers embrace well-priced, convenient authorized services. Indeed, there is no market failure that requires a heavy-handed regulatory approach that promises to yield years of litigation over its very existence.

Finally, the proposal runs counter to the CRTC’s longstanding emphasis that it does not licence or judge Internet content nor is it empowered by legislation to do so. The regulatory framework for telecommunications – whether in the Act’s objectives, the government’s policy direction, or in the Supreme Court’s clear separation of broadcasting and telecom – all point to policy priorities premised on efficiency, affordability, and competitiveness. The CRTC recognizes that content regulation of telecommunications is incompatible with those priorities since its primary role is to help facilitate a competitive marketplace. To engage in content regulation on the Internet would turn it into an Internet content regulatory authority, opening the door to licensing or regulating Internet streaming services, content that runs through ISP networks, and the legality of web-based content wherever it may be located.

Supporters of the Bell proposal downplay these concerns, arguing that it is a narrowly tailored approach to address piracy. This series has identified why the blocking system is likely to lead to overblocking and expanded scope of coverage for both IP and non-IP issues. But even more fundamentally, implementing a blocking plan without court order under the auspices of the CRTC turns the Commission (and by extension the government) into a regulator of Internet content in direct contradiction to the CRTC’s stated approach to online content.

For further discussion on all of these issues, links to the full series is posted below:

2 Comments

  1. Guillaume says:

    Thanks Michael, we’re lucky to have someone like you actively watching Canadian privacy issues.

    To support your point #1, I read through Bell’s yearly financial report to see how much of a strategic and financial issue piracy had become for them.

    I was surprised to find a single reference to Piracy in the whole report – citing a 150 million $ supreme court settlement that Bell lost to Videotron because they were turning a blind eye to piracy of their own satellite signals in the early 2000’s.

  2. Pingback: This Week’s [in]Security – Issue 50 - Control Gap | Control Gap