Centre for Internet and Society

An Evidence based Intermediary Liability Policy Framework: Workshop at IGF

jyoti — 2014-07-04T06:41:10Z

CIS is organising a workshop at the Internet Governance Forum 2014. The workshop will be an opportunity to present and discuss ongoing research on the changing definition of intermediaries and their responsibilities across jurisdictions and technologies and contribute to a comprehensible framework for liability that is consistent with the capacity of the intermediary and with international human-rights standards.

The Centre for Internet and Society, India and Centre for Internet and Society, Stanford Law School, USA, will be organising a workshop to analyse the role of intermediary platforms in relation to freedom of expression, freedom of information and freedom of association at the Internet Governance Forum 2014. The aim of the workshop is to highlight the increasing importance of digital rights and broad legal protections of stakeholders in an increasingly knowledge-based economy. The workshop will discuss public policy issues associated with Internet intermediaries, in particular their roles, legal responsibilities and related liability limitations in context of the evolving nature and role of intermediaries in the Internet ecosystem. distinct

Online Intermediaries: Setting the context

The Internet has facilitated unprecedented access to information and amplified avenues for expression and engagement by removing the limits of geographic boundaries and enabling diverse sources of information and online communities to coexist. Against the backdrop of a broadening base of users, the role of intermediaries that enable economic, social and political interactions between users in a global networked communication is ubiquitous. Intermediaries are essential to the functioning of the Internet as many producers and consumers of content on the internet rely on the action of some third party–the so called intermediary. Such intermediation ranges from the mere provision of connectivity, to more advanced services such as providing online storage spaces for data, acting as platforms for storage and sharing of user generated content (UGC), or platforms that provides links to other internet content.

Online intermediaries enhance economic activity by reducing costs, inducing competition by lowering the barriers for participation in the knowledge economy and fuelling innovation through their contribution to the wider ICT sector as well as through their key role in operating and maintaining Internet infrastructure to meet the network capacity demands of new applications and of an expanding base of users.

Intermediary platforms also provide social benefits, by empowering users and improving choice through social and participative networks, or web services that enable creativity and collaboration amongst individuals. By enabling platforms for self-expression and cooperation, intermediaries also play a critical role in establishing digital trust, protection of human rights such as freedom of speech and expression, privacy and upholding fundamental values such as freedom and democracy.

However, the economic and social benefits of online intermediaries are conditional to a framework for protection of intermediaries against legal liability for the communication and distribution of content which they enable.

Intermediary Liability

Over the last decade, right holders, service providers and Internet users have been locked in a debate on the potential liability of online intermediaries. The debate has raised global concerns on issues such as, the extent to which Internet intermediaries should be held responsible for content produced by third parties using their Internet infrastructure and how the resultant liability would affect online innovation and the free flow of knowledge in the information economy?

Given the impact of their services on communications, intermediaries find themselves as either directly liable for their actions, or indirectly (or “secondarily”) liable for the actions of their users. Requiring intermediaries to monitor the legality of the online content poses an insurmountable task. Even if monitoring the legality of content by intermediaries against all applicable legislations were possible, the costs of doing so would be prohibitively high. Therefore, placing liability on intermediaries can deter their willingness and ability to provide services, hindering the development of the internet itself.

Economics of intermediaries are dependent on scale and evaluating the legality of an individual post exceeds the profit from hosting the speech, and in the absence of judicial oversight can lead to a private censorship regime. Intermediaries that are liable for content or face legal exposure, have powerful incentives, to police content and limit user activity to protect themselves. The result is curtailing of legitimate expression especially where obligations related to and definition of illegal content is vague. Content policing mandates impose significant compliance costs limiting the innovation and competiveness of such platforms.

More importantly, placing liability on intermediaries has a chilling effect on freedom of expression online. Gate keeping obligations by service providers threaten democratic participation and expression of views online, limiting the potential of individuals and restricting freedoms. Imposing liability can also indirectly lead to the death of anonymity and pseudonymity, pervasive surveillance of users' activities, extensive collection of users' data and ultimately would undermine the digital trust between stakeholders.

Thus effectively, imposing liability for intermediaries creates a chilling effect on Internet activity and speech, create new barriers to innovation and stifles the Internet's potential to promote broader economic and social gains. To avoid these issues, legislators have defined 'safe harbours', limiting the liability of intermediaries under specific circumstances.

Online intermediaries do not have direct control of what information is or information are exchanged via their platform and might not be aware of illegal content per se. A key framework for online intermediaries, such limited liability regimes provide exceptions for third party intermediaries from liability rules to address this asymmetry of information that exists between content producers and intermediaries.

However, it is important to note, that significant differences exist concerning the subjects of these limitations, their scope of provisions and procedures and modes of operation. The 'notice and takedown' procedures are at the heart of the safe harbour model and can be subdivided into two approaches:

a. Vertical approach where liability regime applies to specific types of content exemplified in the US Digital Copyright Millennium Act

b. Horizontal approach based on the E-Commerce Directive (ECD) where different levels of immunity are granted depending on the type of activity at issue

Current framework

Globally, three broad but distinct models of liability for intermediaries have emerged within the Internet ecosystem:

1. Strict liability model under which intermediaries are liable for third party content used in countries such as China and Thailand

2. Safe harbour model granting intermediaries immunity, provided their compliance on certain requirements

3. Broad immunity model that grants intermediaries broad or conditional immunity from liability for third party content and exempts them from any general requirement to monitor content.

While the models described above can provide useful guidance for the drafting or the improvement of the current legislation, they are limited in their scope and application as they fail to account for the different roles and functions of intermediaries. Legislators and courts are facing increasing difficulties, in interpreting these regulations and adapting them to a new economic and technical landscape that involves unprecedented levels user generated content and new kinds of and online intermediaries.

The nature and role of intermediaries change considerably across jurisdictions, and in relation to the social, economic and technical contexts. In addition to the dynamic nature of intermediaries the different categories of Internet intermediaries‘ are frequently not clear-cut, with actors often playing more than one intermediation role. Several of these intermediaries offer a variety of products and services and may have number of roles, and conversely, several of these intermediaries perform the same function. For example , blogs, video services and social media platforms are considered to be 'hosts'. Search engine providers have been treated as 'hosts' and 'technical providers'.

This limitations of existing models in recognising that different types of intermediaries perform different functions or roles and therefore should have different liability, poses an interesting area for research and global deliberation. Establishing classification of intermediaries, will also help analyse existing patterns of influence in relation to content for example when the removal of content by upstream intermediaries results in undue over-blocking.

Distinguishing intermediaries on the basis of their roles and functions in the Internet ecosystem is critical to ensuring a balanced system of liability and addressing concerns for freedom of expression. Rather than the highly abstracted view of intermediaries as providing a single unified service of connecting third parties, the definition of intermediaries must expand to include the specific role and function they have in relation to users' rights. A successful intermediary liability regime must balance the needs of producers, consumers, affected parties and law enforcement, address the risk of abuses for political or commercial purposes, safeguard human rights and contribute to the evolution of uniform principles and safeguards.

Towards an evidence based intermediary liability policy framework

This workshop aims to bring together leading representatives from a broad spectrum of stakeholder groups to discuss liability related issues and ways to enhance Internet users’ trust.

Questions to address at the panel include:

1. What are the varying definitions of intermediaries across jurisdictions?

2. What are the specific roles and functions that allow for classification of intermediaries?

3. How can we ensure the legal framework keeps pace with technological advances and the changing roles of intermediaries?

4. What are the gaps in existing models in balancing innovation, economic growth and human rights?

5. What could be the respective role of law and industry self-regulation in enhancing trust?

6. How can we enhance multi-stakeholder cooperation in this space?

Confirmed Panel:

Technical Community: Malcolm Hutty: Internet Service Providers Association (ISPA)
Civil Society: Gabrielle Guillemin: Article19
Academic: Nicolo Zingales: Assistant Professor of Law at Tilburg University
Intergovernmental: Rebecca Mackinnon: Consent of the Networked, UNESCO project
Civil Society: Anriette Esterhuysen: Association for Progressive Communication (APC)
Civil Society: Francisco Vera: Advocacy Director: Derechos Digitale
Private Sector: Titi Akinsanmi: Policy and Government Relations Manager, Google Sub-Saharan Africa
Legal: Martin Husovec: MaxPlanck Institute

Moderator(s): Giancarlo Frosio, Centre for Internet and Society (CIS) and Jeremy Malcolm, Electronic Frontier Foundation

Remote Moderator: Anubha Sinha, New Delhi

For more details visit https://cis-india.org/internet-governance/blog/igf-workshop-an-evidence-based-intermediary-liability-policy-framework

A provisional definition for the Cultural Last Mile

nishant — 2011-08-02T08:57:07Z

In the first of his entries, Ashish Rajadhyaksha gives his own spin on the 'Last Mile' problem that has been at the crux of all public technologies. Shifting the terms of debate away from broadcast problems of distance and access, he re-purposes the 'last mile' which is a communications problem, to make a cultural argument about the role and imagination of technology in India, and the specific ways in which this problem features in talking about Internet Technologies in contemporary India.

In its classical form, the ‘last mile’ is a communications term defining the final stage of providing connectivity from a communications provider to a customer, and has been used as such most commonly by telecommunications and cable television industries. There has however been a a specific Indian variant, seen in its most classical avatar in scientist Vikram Sarabhai’s contention that overcoming the last mile could solve the two major challenges India has faced, of linguistic diversity and geographical distance, and mounted as the primary argument for terrestrial television in the early 1980s. (I will try and attach the Sarabhai paper a little later to this posting).

This specifically Indian variation, where technology was mapped onto developmentalist-democratic priorities, has been the dominant characteristic of communications technology since at least the invention of the radio in the 1940s. For at least 50 years now, that means, the last mile has become a mode of a techno-democracy, where connectivity has been directly translated into democratic citizenship. It has continuously provided the major rationale for successive technological developments, from the 1960s wave of portable transistors, the terrestrial transponders of the first televisual revolution it the early 1980s (the Special Plan for the Expansion of Television), the capacity of satellite since SITE and the INSAT series, and from the 1990s the arrival of wired networks (LANs, Cable, fibre-optic) followed by wireless (WLAN, WiMAX, W-CDMA). At each point the assumption has been consistently made that the final frontier was just around the corner; that the next technology in the chain would breach a major barrier, once and for all.

What I hope to do is to provide a historical account to argue that the theory of the ‘last mile’ has been founded on fundamental (mis)apprehensions around just what this bridge constitutes. Further, that these apprehensions may have been derived from a misconstruction of democractic theory, to assume, first, an evolutionary rather than distributive model for connectivity, and second, to introduce a major bias for broadcast (or one-to-many) modes as against many-to-many peer-to-peer formats. The book, whenever I succeed in writing it, will hope to argue the following:

1. It has been difficult to include human resource as an integral component to the last mile. Contrary to the relentlessly technologized definition of the last mile, it may perhaps be best seen historically as also, and even perhaps primarily, a human resource issue. This is not a new realization, but it is one that keeps reproducing itself with every new technological generation[1], with ever newer difficulties. The endemic assumption, derived from the broadcasting origins of the definition is that it is primarily the sender’s responsibility to bridge the divide, that technology can aid him to do so on its own, and that such technology can negate the need to define connectivity as a multiple-way partnership as it reduces the recipient into no more than an intelligent recipient of what is sent (the citizen model). On the other hand, it is possible to show how previous successful experiments bridging the last mile have been ones where recipients have been successfully integrated into the communications model both as peers and, even more significantly, as originators as well as enhancers of data. Importantly, this paper will show, this has been evidenced even in one-way ‘broadcast’ modes such as film, television and radio (in the movie fan, community radio and the television citizen-journalist).

2. The one-way broadcast versus peer-to-peer versus two/multiple-way debate needs to he historically revisited. The need to redefine the beneficiary of a connectivity cycle as a full-fledged partner tends to come up against a bias written into standard communications models – and therefore several standard revenue models – that consistently tend to underplay what this paper will call the significant sender/recipient. While both terrestrial and satellite systems require some level of peer-to-peer transmission systems to facilitate last-mile communications, it has been a common problem that unless either a clear focus exists on geographic areas or significant peer-to-peer participation exists, broadcast models inevitably find themselves delivering large amounts of S/N at low frequencies without sufficient spectrum to support large information capacity. While it is technically possible to ‘flood’ a region in broadcasting terms, this inevitably leads to extremely high wastage as much of the radiated ICE never reaches any user at all. As information requirements increase, broadcast ‘wireless mesh’ systems small enough to provide adequate information distribution to and from a relatively small number of local users, require a prohibitively large number of broadcast locations along with a large amount of excess capacity to make up for the wasted energy.

This problem, importantly, springs as much from a built-in ideological commitment to one-way broadcasting formats, as from technological limitations. The technology itself poses further problems given the bias of different systems to different kinds of connectivity, and with it different types of peer-to-peer possibilities. Rather than attempting a one-size-fits-all model for all models to follow, we need to work out different synergies between broadcast-dependent and peer-to-peer-enabled platforms.

This book will eventually hope to study the history of peer-to-peer and multiple-way structures as systems where sending has become a component part of receiving. Key technological precedents to the present definition of the sender-communication ‘partner’ would be community radio, low-power transmission-reception systems (most famously the Pij experiment in Gujarat conducted by ISRO), and various internet-based networking models.

3. The need to revisit the technological community is therefore critical. The key question is one of how technological communities have been produced, and how they may be sustained. In January 2007, the attack by V.S. Ailawadi, former Chairman, Haryana Electricty Regulatory Commission, on India’s public sector telecom giants BSNL and MTNL for keeping their ‘huge infrastructure’ of ‘copper wire and optic fibre’ to themselves, when these could be used by private operators as cheaper alternatives to WiMAX, W-CDMA and broadband over power lines, shows the uneasy relationship between new players and state agencies. Mr. Ailawadi’s contention that the ‘unbundling’ of the last mile would bring in competition for various types of wireless applications and broadband services not just for 45 million landlines but also for 135 million mobile users of various service providers, also therefore needs to be revisited from the perspective of community formation. How would the new 135 million mobile users be effectively tapped for their capacity to become what we are calling significant senders?

In defining the last mile as to do with the recipient-as-sender, and thus the community, this paper will focus on a history of community action along specific models of connectivity. These are: cinema’s movie fan, internet’s blogger and networker, solar energy’s barefoot engineer, software’s media pusher and television’s citizen-journalist. A specific focus for study will be the models of participatory learning in the classroom, using film, the vinyl disc, the audio cassette, the radio, the television, the web and now the mobile phone.

For more details visit https://cis-india.org/raw/histories-of-the-internet/blogs/the-last-cultural-mile/definiton