JUNE 2005
Index
I. Introduction 1
II. Working definition of Internet governance 6
The need for a shared understanding of Internet governance 6
Box 1: Some proposed definitions of Internet governance 9
Assessing the terms of the definition 11
III. Identifying public policy Issues that are relevant to Internet governance and assessing the adequacy of existing governance arrangements 14
A. Issues relating to infrastructure and management of critical Internet resources 14
Box 2: Governance issues related to international Internet connection costs 15
B. Issues relating to the use of the Internet, including spam, network security, and cybercrime 27
C. Issues which are relevant to the Internet, but with impact much wider than the Internet 35
D. Issues relating to developmental aspects of Internet governance, in particular capacity building in developing countries 44
Box 3: Free and open source software (FLOSS) 48
IV. Developing a common understanding of the respective roles and responsibilities of all stakeholders from both developed and developing countries 54
A. Introduction 54
B. Types of governance mechanisms 54
C. Applying the WSIS criteria: a general assessment 56
D. Assessing the actors against the WSIS criteria 58
E. Special considerations 60
F. Impact of the private sector on governance 62
G. Civil society involvement in global governance arrangements 64
H. A way forward 65
Annex 67
Glossary 71
1. This Background Report accompanies and is complementary to the Report of the Working Group on Internet Governance (The WGIG Report). It includes much of the work produced in the course of the Working Group process and reflects the wide variety of opinions held within the group as well as many comments made by stakeholders during the consultation process. The Background Report does not have the same status as the WGIG Report, which is a short consensus document for policy makers. However, the Background Report can be used as a reference in that it provides a summary of the process and various issue papers, with some additional thoughts and considerations about potential solutions for issues not covered in detail in the WGIG Report. If not every member of the group agrees with every word, they all agree with this approach and the Background Report makes clear whether an argument or opinion is shared by the entire group or only by some of its members.
The Working Group on Internet Governance (WGIG) was set up by the Secretary-General of the United Nations in accordance with the mandate given to him by the first phase of the World Summit on the Information Society (WSIS), held in Geneva, on 10 – 12 December 2003. The WGIG comprised 40 members from governments, private sector and civil society who all participated on an equal footing and in their personal capacity. It was chaired by Mr Nitin Desai, Special Advisor to the Secretary-General for the WSIS. The list of the members of the WGIG is attached as an annex to the report.
The WGIG held four meetings in Geneva: 23 – 25 November 2004; 14 – 18 February 2005; 18 – 20 April 2005; and 14 – 17 June 2005.
4. The WGIG mandate goes back to the Geneva phase of the WSIS, when Heads of State and government recognized the importance of the Internet. They noted that the Internet is a central element of the infrastructure of the emerging information society and established principles to guide the management of the Internet including as definition of the roles and responsibilities of the various stakeholders1. They recognized that there are differing views on the suitability of current institutions and mechanisms for managing processes and developing policies for the global Internet. For this reason they requested the Secretary-General to set up a Working Group on Internet Governance with a view to preparing the ground for negotiations during the second phase of the WSIS, culminating in the Summit to be held in Tunis in November 20052.
5. The WSIS Declaration of Principles and Plan of Action3 adopted in Geneva set the parameters for the WGIG and contain its Terms of Reference and work programme. The WGIG has been asked, inter alia, to “investigate and make proposals for action, as appropriate, on the governance of the Internet by 2005”4, dealing with the following issues5:
Develop a working definition of Internet governance;
Identify the public policy issues that are relevant to Internet governance;
Develop a common understanding of the respective roles and responsibilities of governments, existing international organizations and other forums as well as the private sector and civil society from both developing and developed countries.
6. It was asked to present its findings in a report to be presented “for consideration and appropriate action” for the second phase of WSIS in Tunis in 2005.
7. Discussions leading to the establishment of the WGIG took place from early in 2004 and included workshops and consultations at a wide range of meetings of intergovernmental and other organizations. Among the events that took up this issue were the International Telecommunications Union (ITU) Workshop on Internet Governance (26 – 27 February 2004); the UN ICT Task Force Global Forum on Internet Governance (24 – 25 March 2004); ITU Telecom Africa in Cairo (4 – 8 May 2004), INET2004 in Barcelona (10 – 14 May 2004); and WSIS PrepCom-1, held at Hammamet in Tunis (24 – 26 June 2004). Discussions continued at the meeting in Kuala Lumpur of the Internet Corporation for Assigned Names and Numbers (ICANN) (19 – 23 July 2004), at the ITU Telecom Asia in Busan, Korea (7 – 11 September 2004) and other ad hoc conferences.
8. The WGIG Secretariat was established in July 2004. Consultations on the setting up of the Working Group were held at the United Nations in Geneva on 20 and 21 September 2004, chaired by Mr Desai in his capacity as Special Advisor to the Secretary-General. The consultations were held in an open mode, allowing all actors involved in Internet issues to participate on an equal footing. Over 250 participants, representing governments, civil society organizations and private sector entities, attended the consultations. Subsequently, after further informal consultations with all stakeholders, on 11 November 2004 the Secretary-General announced the establishment of the WGIG.
9. The WGIG conceived itself not as a negotiating body but as a working group with the task of preparing the ground for the negotiations of the Tunis Phase of the WSIS in November 2005.
10. The guiding principles for the WGIG’s working methods are set out in the WSIS Declaration of Principles and Plan of Action. The WGIG is called upon to be “open and inclusive” in its work and design a “process that ensures a mechanism for the full and active participation of governments, the private sector and civil society from both developing and developed countries, involving relevant intergovernmental and international organizations and forums”6. The WGIG agreed that transparency was another key ingredient to ensure ownership of the process among all stakeholders. In order to maximise transparency and open communication and to facilitate its own work, the WGIG resolved to use the Internet to the maximum extent possible. Thus members worked extensively through e-mail, Internet Protocol (IP) based streaming video, bulletin boards and a discussion forum, and used the WGIG website to communicate with the public. The public sessions, which were held in conjunction with the WGIG meetings, were webcast from February onwards. Real time captioning was introduced for the public meetings held in April and June and made available on the WGIG website.
11. The WGIG decided that all its formal face-to-face meetings would be accompanied by consultations open to all stakeholders and online consultation processes. Observers from intergovernmental organizations (IGOs) could attend all meetings and participate in online discussion lists. The WGIG relied on external comments and contributions as an integral part of its work. Open ended consultations have been conducted both online and offline. A total of 105 contributors submitted 182 written comments.
12. In addition, the following regional and sub-regional meetings have provided input into the work of WGIG: the South-East and East Asia Conference on Preparations for WSIS II in Bali, Indonesia, 1 – 3 February 2005; the African WSIS Regional Conference in Accra, Ghana, 2 – 4 February 2005; the Pan Arab Conference on WSIS in Cairo, Egypt, 8 – 10 May 2005, for the Arab Region; the High-Level Asia-Pacific Conference for the WSIS in Tehran, Islamic Republic of Iran, 31 May – 2 June; and the Regional Meeting for Latin America and the Caribbean in Rio de Janeiro, Brazil, 8 – 10 June 2005.
13. The WGIG also took note of the Marrakesh Declaration and the Action Plan on The Role and the Place of the Media in the Information Society in Africa and the Arab Region, adopted at Marrakesh, Morocco, on 24 November 2004.
14. The WGIG chose as its point of entry into the substantive work the identification of public policy issues that are potentially relevant to Internet governance, as called for in Paragraph 13 (b) of the WSIS Plan of Action and started work by gathering facts and mapping out the terrain. The WGIG worked simultaneously on developing a practical definition of the Internet itself and defining Internet governance public policy issues. It was felt that an iterative method would be the best way of moving toward an implicit working definition of Internet governance.
The WGIG agreed to take a broad approach and not exclude any potentially relevant issue. This first, fact finding phase was intended to lead to the identification of public policy issues that are relevant to Internet governance. As a first step, the WGIG developed a series of draft working papers. The purpose of these papers was to act as a brief summary document setting out some of the basic issues and to identify aspects of relevance to the task of the WGIG. The production of these papers also secured the effective working of the group via the Internet and thereby laid the foundations for future collaborative work within WGIG. Each paper was written by a small group of contributors and then discussed by the group as a whole. However, the papers do not necessarily present a consensus position, nor do they contain agreed language accepted by every member. This whole process was conducted using Internet tools. In all, 21 draft papers were produced and made available on the WGIG website for public comment.
16. On the whole, comments received commended the WGIG for the openness of its process and added many factual elements and corrections. Different opinions were voiced as regards the content of the draft papers. The WGIG agreed that all comments received would be part of the background material it would use. It was understood that these papers should be read with the comments as part of a package.
17. Based on this fact finding phase, the WGIG established the key public policy areas for further investigation and discussion. Following the group’s second meeting, and taking into account the views of the public, the following four clusters of issues were identified:
Issues relating to infrastructural issues and the management of critical Internet resources, including administration of the domain name system and IP addresses, administration of the root server system, technical standards, peering and interconnection, telecommunications infrastructure including innovative and converged technologies, as well as multilingualization. These issues are matters of direct relevance to Internet governance falling within the ambit of existing organizations with responsibility for these matters;
Issues relating to the use of the Internet, including spam, network security, and cybercrime. While these issues are directly related to Internet governance, the nature of global cooperation required is not well defined;
Issues which are relevant to the Internet, but with impact much wider than the Internet, where there are existing organizations responsible for these issues, such as intellectual property rights (IPRs) or international trade. The WGIG started examining the extent to which these matters are being handled consistently with the WSIS Declaration of Principles; and
Issues relating to developmental aspects of Internet governance, in particular capacity building in developing countries.
18. In the second phase of its work the WGIG assessed the adequacy of current Internet governance arrangements and looked into the respective roles and responsibilities of all stakeholders, in accordance with the mandate set out by the Geneva phase of the WSIS. Again, it produced 16 papers that were made available on the WGIG website for public comment.
In carrying out this assessment, the WGIG took the key WSIS principles as its main point of orientation. The group devoted much attention to the issue of coordination and assessed the capacity of existing Internet governance arrangements to address governance issues in a coordinated manner. It clearly felt there was room for improvement in this area and saw a need for existing institutions to have a closer cooperation on Internet governance issues.
The WSIS principles also include the terms “multilateral”, “transparent”, and “democratic” as well as the notion of the “full involvement of governments, stakeholders and international organizations”. While not questioning these principles, the WGIG spent some time clarifying their meaning and developed a common understanding on the extent to which existing arrangements ensure the full involvement of governments, the private sector, civil society and international organizations in a process which is multilateral, transparent and democratic. It came to the conclusion that from an operational point of view, the WSIS criteria of multilateralism, transparency, democracy and full involvement of all stakeholder groups have somewhat different meanings, possibilities, and limits in relation to different types of governance mechanisms. They may therefore be regarded as having different shades of meaning in different contexts. For example, the WGIG recognized that “full involvement of all stakeholders” would not necessarily mean that every stakeholder group should have the same role in the development of policies, the preparation of decisions, the actual decisions and then the implementation of decisions.
21. In the third phase, leading up to its last meeting, the WGIG focused on developing “proposals for action, as appropriate, on the governance of the Internet” and on the drafting of the Background Report. It developed recommendations and options on how to improve on current Internet governance arrangements. Not all options have the support of all members, but all members agree with the approach to present various options.
22. The WGIG consciously decided not to include a separate chapter on the evolution of the Internet. It held the view that by now the Internet was so widely used and its history so well documented that there would be little value added by including a separate chapter on this subject. For readers interested to learn more about the history of the Internet the WGIG recommends the excellent document on this subject available at the website of the Internet Society (ISOC), entitled A Brief History of the Internet with contributions by many leading proponents in the invention and development of the Internet, such as Vinton G. Cerf and Robert E. Kahn7.
23. Nevertheless, the WGIG found it useful for developing an understanding of governance issues to look at the different phases of the Internet’s development from a research project in the 1960s to a widespread commercial infrastructure with close to one billion Internet users connected in 2004. Looking back, it detected some guiding principles and factors that had enabled or contributed to this development8. The WGIG viewed the WSIS principle relating to the stable and secure functioning of the Internet as of paramount importance. At the outset the WGIG agreed that all recommendations aiming to improve current governance arrangements should be fully assessed in function of their capacity to enhance or preserve the stability and security of the Internet. Thus, any recommendation to improve present arrangements should take care of that concern.
Furthermore, the WGIG noted the importance of the following factors:
The decentralized and collaborative process of underlying technological development and core resource management: the technological development and administration of the Internet, allowing participation by all interested parties and rejecting centralized advance validation of content, services and technologies, helps ensure that the network is interoperable, functional, stable, secure, efficient as well as scalable in the long run.
The distributed/decentralized open architecture: the Internet, a ‘network of networks’, is made up of tens of thousands of interconnected networks run by Internet service providers, individual companies, universities, governments and others, which can communicate together, based on the key underlying technical idea of open architecture networking in which any type of network anywhere can be included and be made publicly available.
The open, non-proprietary nature of the core Internet standards: most of the protocols at the core of the Internet are protocols based on open standards that are efficient, trusted, and open to global implementation with little or no licensing restrictions. The protocol specifications are available to anyone, at no cost, thus considerably reducing barriers to entry and enabling interoperability.
Private sector competition and innovation: market mechanisms have by and large enabled the development of the Internet, supported by liberalizing markets.
The end-to-end principle: the neutrality of the Internet, chiefly concerned with the effective transportation of packets, enables its intelligence to reside largely at the networks’ ends through applications in computers, servers, mobile and other devices. This has enabled the development of a wide range of new ICT activities, industries and services ‘at the ends’ and turns the Internet into an important tool within the wider context of economic and societal development.
All these factors are important elements in any consideration of Internet governance arrangements. Any proposal for change would have to assess whether any of these elements, which are important for the functioning of the Internet, would be affected in one way or another.
27. Discussion and debate about Internet governance began in earnest about ten years ago, and has grown and intensified in step with the progressive transformation of the Internet from a research and academic facility into a general purpose communications medium widely available to the public and used for an expanding range of private and public purposes.
28. The founders of the Internet and many of those who participated in its early development did not think of the Internet as being governed, or even as requiring governance. Instead, they saw it as being coordinated in a bottom-up, collaborative, largely voluntary fashion as exemplified in the work of the Internet Engineering Task Force (IETF) and the Internet Assigned Numbers Authority (IANA). However, when the invention of the World Wide Web helped make the Internet accessible to a wide range of users from the mid-1990s on, the increasing commercial value of the Internet, its growing importance to business and government users in all countries, and the emergence of threats to the security of individual users and the Internet as a whole raised new issues and drew new actors into a growing international discussion about how the Internet should be governed.
29. In this new light, different views began to emerge about the appropriate scope and mechanisms of Internet governance. At one end of the spectrum, some actors took a relatively narrow view of these questions in which only the management of the Internet’s core resources (e.g. IP addresses, domain names, the root zone) needed special governance arrangements, and in which contracts were the principal governance mechanism. At the other end of the spectrum, some other actors took a much broader view of the scope of Internet governance that included its social, economic, cultural and political impacts in addition to its technical and logical infrastructure, and envisaged the use of a correspondingly wider range of governance mechanisms, including treaty instruments. Many other actors had views on the appropriate scope and mechanisms of Internet governance that fell somewhere in between these two ends of the spectrum.
30. The differences of opinion about the appropriate scope and mechanisms of Internet governance that emerged in the negotiations preceding WSIS-I were accentuated by the absence of a common understanding about the meaning of the term ‘governance’ and its relationship to government. Although the two words have the same roots, they have very different meanings
31. Governance, in the sense we use it, influences political processes and public institutions by shaping the way people interact with government and how government interacts with them. However, governance also occurs in other areas of social life. In the private sector, it has long been recognized that there is an important dimension of ‘corporate governance’ underlying the management and operation of private companies, comprised of ongoing interactions between shareholders, directors and managers. In recent years, the increasingly visible role of civil society organizations in shaping the agendas and programmes of public and private organizations nationally and internationally has highlighted the fact that there is a governance dimension to every area of economic and social life. Seen in this light, governance is a part of many different processes related to the Internet, including the development of technical standards and the management of core resources, as well as regulation of the misuse and abuse of the Internet.
32. Even if it is not equated with government, there is sometimes a tendency to think of governance either as a top-down process modelled on government policy making and administration, or as a bottom-up process modelled on electoral mechanisms. As we see it, governance is primarily a horizontal process of interaction that influences the way in which hierarchical processes operate, both top-down and bottom-up. In this sense, governance is closely associated with the notion that all stakeholders should play a role in all Internet related decision making processes, but that the precise way in which they play these roles should vary according to the nature of the issue and the character of the process.
33. For WSIS participants, a lot is at stake in the way the Internet is governed in the future, since all recognize that the Internet is one of the key tools for achieving the development goals that are at the heart of the summit process. Given the relative novelty of the subject, the different points of view that exist, and the lack of previous opportunities to discuss Internet governance in international fora, it is perhaps not surprising that it was not possible to get complete agreement on Internet governance in the negotiations that preceded WSIS-I. In order to break the logjam in the months remaining before the second phase, it is essential for participants to develop a shared view and common understanding of Internet governance, beginning with its definition.
Descriptiveness
37. To be accurate, the definition must capture what Internet governance is, rather than what we wish it to be. This is standard practice with respect to both social and physical phenomena; one does not define ‘politics’ as the pursuit of truth and justice or ‘temperature’ as pleasant weather, if even both are desirable. Accordingly, while the WGIG gave significant attention to the WSIS Declaration of Principles’ overarching prescriptions9 – that Internet governance should be multilateral, transparent and democratic, with the full involvement of governments, the private sector, civil society and international organizations, and should ensure an equitable distribution of resources, facilitate access for all and ensure a stable and secure functioning of the Internet, taking into account multilingualism – and used these to guide its evaluation of existing governance mechanisms10, it did not incorporate them into the definition itself.
|
Box 1: Some proposed definitions of Internet governance
Note: Source for 1 to 7: ITU, Compilation of Proposed Definitions, informal input paper to the WGIG, November 2004. http://www.wgig.org/docs/IG-definitions.doc
|
|
(Don MacLean, February 2005 WGIG e-mail)
|
42. There are at least five noteworthy commonalities among the formulations in Box 1 that merit consideration when developing a working definition. First, almost all of them position as the core components prescriptive and proscriptive injunctions that constrain and enable participants in the Internet environment. The most commonly used terms were rules and procedures, with related terms like principles, norms, standards, processes, programmes, policies, institutions, and dispute resolution mechanisms varyingly receiving mention. Second, many of the proposed definitions emphasize that the key feature of such injunctions is that they are shared or collectively recognized and applied. Third, many of the proposals state what these injunctions do or how they function in relation to social actors. That is, they shape social actors’ expectations, practices, and interactions; steer, coordinate, and administrate; or accommodate conflicting or diverse interests. Fourth, several inclusively identify the parties that develop the injunctions, i.e. governments, the private sector and civil society. And fifth, all the definitions are implicitly or explicitly inclusive with respect to the scope of Internet governance, or the range of issues and functions it encompasses. Most notable in this regard, and in contrast to the common pre-WSIS practice mentioned above, none of the definitions restricts the definition to only the governance of logical infrastructure issues – that is, naming, numbering, and root server and the zone file.
44. The terms employed in this working definition merit some elaboration.
Development and application by governments, the private sector, and/or civil society
45. This phrase establishes at the outset that governance is fundamentally a process that involves supplying and applying mechanisms that steer or give order to courses of action. It is neutral with respect to how and where these two functions are performed, as this varies widely across different issue areas and institutional contexts. From a logical standpoint, listing the stakeholders who engage in this process is not strictly necessary. But from a political standpoint, it seems useful to underscore that any of the three societal sectors may play a role, depending on the particular case at hand.
(b) Shared
46. A core feature of any governance mechanism is that the parties involved collectively recognize that it exists and applies to them. Internet governance being a highly distributed domain involving many different organizational models, the term is neutral with respect to the number of parties and modalities involved in the processes of creating and applying governance. One party, a few, or many, may supply or implement governance, and this can be done through a wide variety of techniques that vary along multiple dimensions, e.g. formal/informal, explicit/implicit, top-down/bottom-up, imposed/negotiated/spontaneously emergent, etc. As the definition is descriptive rather than normative, no judgment can be made within its terms as to whether one or the other approach is more appropriate than others. What matters here is simply that the stakeholders involved in a particular arena acknowledge and are guided by the same mechanisms as their counterparts.
(c) Principles
47. In the global policy environment, as elsewhere, this term is often used in two different ways. One is to refer to statements of fundamental facts or causation about the subject matter at hand. Examples in the Internet environment would include the principle of open, non-proprietary technical standardization, or the ‘end-to-end’ principle according to which the network simply provides data transport, with applications and processing left to the users at the ends. The other is to refer to the overarching objectives that define an activity, such as global governance. For example, the interconnection of networks is a guiding principle of the international telecommunications regime, most favoured nation treatment is a guiding principle of the international trade regime, and competition among registrars is a guiding principle of the international regime for Internet naming and numbering. Both of these uses, which can blend into one another at times, are entailed by the term’s inclusion in the definition. In short, principles define what a given governance mechanism is about and, at the highest level, is intended to promote.
(d) Norms
48. This term refers to broadly framed behavioural standards that the parties adopt in light of or to give effect to the relevant guiding principles. Norms set out obligations and rights, and are both proscriptive and prescriptive – i.e. they constrain actors from following some courses of action while empowering them to pursue others with community assent.
Rules
49. This term is often used generically for all types of prescriptions and proscriptions of varying degrees of generality or specificity, as in the popular formulation, ‘the rules of the game’. Many of the proposed definitions cited above use the word in this inclusive manner, i.e. as a more economical way of saying ‘rule systems’ or ‘principles, norms, and rules’. However, in the course of the WGIG’s deliberations and consultations with external stakeholders, it has become clear that using the term as a catch-all for any sort of injunction may invite misunderstanding, in that ‘rules’ is sometimes interpreted to mean mandatory rules, or regulatory rules. Hence, it seems advisable to separately mention the terms principles and norms, which seem to be understood in a broader and more open ended manner, and to use rules to refer to more narrowly bounded injunctions to do this or that in a specific circumstance.
(f) Decision making procedures
50. Whereas principles, norms and rules refer to substantive injunctions, this term pertains to the operational processes followed in making decisions.
(g) Programmes
51. This term refers to collective, purposive activities that impact social actors’ expectations and practices but do not consist of developing and applying rules regulating their daily conduct in a given arena. For example, many intergovernmental and nongovernmental organizations develop and administer programmes concerned with producing and disseminating information, monitoring events, managing shared resources and facilities, providing technical and financial assistance to developing countries, and so on. When these activities are significant enough to impact stakeholders’ capabilities and practices and the technical operation of the Internet, they become an important element of Internet governance.
(h) Development and utilization of the Internet
52. Finally, and in keeping with the observations made in Section II.A, this phrase reflects the ‘broad understanding’ of Internet governance that has been collectively embraced in the course of the WSIS process. That is, the objects of governance are expectations and behaviour concerning all arenas in which shared principles, norms, rules, decision making procedures and programmes exist. That includes many (but by no means all) aspects of both the ‘supply side’, i.e. the design and provision of physical and logical infrastructures, services, and applications, and the ‘demand side’, i.e. use of the Internet to communicate, access information resources and processes, and engage in commercial or non-commercial transactions. As this report demonstrates in subsequent chapters, there is a highly distributed and heterogeneous array of governance mechanisms pertaining to different issues and functions, and this definition is intended to cover these and any additional mechanisms yet to be defined.
53. Our working definition seems to fit the five criteria outlined earlier. It is arguably adequate and clearly descriptive, concise, and process-oriented. It also appears to be fully generalizable, in that it applies equally to all extant governance mechanisms that have been formally agreed by governments, the private sector, civil society, or multi-stakeholder partnerships; or that rest on informal social conventions, including those developed spontaneously, from the bottom up, through an evolutionary process of mutual adjustment among diverse participants. Further, it applies equally irrespective of a particular governance mechanism:
Substantive focus and purpose (core resources and names and numbers, information or network security, technical standardization, intellectual property, trade and electronic commerce, privacy and consumer protection, communication and content, development, etc.);
Functions (managing scarce resources, ensuring interconnection and interoperability, establishing terms and conditions for economic transactions and communications, reducing transaction and information costs, establishing liability rules, facilitating knowledge transfer and collective learning, and so on);
Organizational context (whether negotiated under the aegis of formal organizations or free standing);
Institutional form, e.g. intergovernmental (e.g. treaties, recommendations, guidelines, declarations, memorandums of understanding, custom) or private sector (contracts, memorandums of understanding, codes, custom) agreements;
Strength (formal or informal, binding or voluntary of prescriptions and proscriptions);
Decision making procedures (voting or consensus, dispute resolution and the like);
Issues covered;
Participants involved (public sector/private sector/civil society, or, in the case of governmental arrangements, unilateral, bilateral, plurilateral, regional, or multilateral);
Compliance mechanisms (centralized or decentralized monitoring and enforcement systems, or the absence thereof);
Distributional biases (equitable or inequitable allocations of rights, responsibilities, and benefits); and so on.
III. Identifying public policy issues that are relevant to Internet governance and assessing the adequacy of existing governance arrangements
Introduction
Issues relating to infrastructure and the management of critical Internet resources are of direct relevance to Internet governance and fall within the ambit of existing organizations with responsibility for these matters. The issues divide fairly naturally into two subgroups:
Issues relating to physical infrastructure including related technical standards, and telecommunications infrastructure including innovative and converged technologies; and
Issues relating to the management of critical Internet resources, including administration of the domain name system and IP addresses, administration of the root server system, as well as multilingualization of the domain name system.
55. The Internet can be envisaged as a large number of interconnected networks that use a common set of protocols, especially the Internet Protocol (IP). The Internet and the applications layered over it are directly dependent on the underlying physical infrastructure being available, secure, well managed, of adequate capacity and service quality, and interconnected. At the same time, Internet protocols are being deployed as a key element in the core telecommunications networks and are the basis of moves towards the convergence of technologies, applications and services that arise from digitization and packetization of telecommunications communications.
2. Physical and secured infrastructure
This section summarizes issues relating to the management of the physical infrastructure, including:
Telecommunications infrastructure, broadband access;
Peering and interconnection;
Radio spectrum policy;
Technical standards.
(a) Governance mechanisms
57. Governance mechanisms operate at four levels: global, regional, national and local, and include intergovernmental agencies with private sector participation, as well as private sector and industrial organizations.
58. The primary regulatory functions, relating to investment and ownership, for the telecommunications networks are exercised through national regulatory bodies, mediated by international agreements through regional institutions and those of the European Union (EU) or treaty organizations such as the World Trade Organization (WTO). The ITU is responsible for the International Telecommunications Regulations (ITRs) which constitute an international treaty covering the provision and operation of international telecommunications services offered to the public and the underlying international telecommunications transport means used to provide such services. Responsibility for broadband access in the ‘last mile’ is a matter of national development and regulation, with some moderation through international investment and trade rules. In some jurisdictions there is heavy reliance on industry self-regulation and, internationally, a de facto form of governance is arising from arrangements being increasingly negotiated by the private telecommunications service providers who own the infrastructure.
59. Interconnection standards and agreements, including peering arrangements, are critical to the successful functioning of the Internet and for maintaining its end-to-end and cost effective availability, and reliability. The primary mechanism for interconnection and peering are private negotiated arrangements or contracts between the owners of the physical infrastructure and do not generally fall under the rubric of international governance mechanisms. International interconnection arrangements and costs have been a matter of study and debate, especially within the ITU, the Asia-Pacific Economic Co-operation Telecommunications and Information Working Group (APEC TEL) and the Organisation for Economic Co-operation and Development (OECD). The ITU Telecommunications Sector (ITU-T) has negotiated a set of principles for international Internet connection, published as Recommendation D.50, as amended. It is not clear to what extent this recommendation has had any effect on interconnection arrangements to date. It has been suggested that peering and the costs of interconnection should be considered in relation to market/trade issues or as a development issue and is, in any case, a highly complex matter of sufficient importance as to require a separate discussion12.
|
Box 2: Governance issues related to international Internet connection costs
1. What is the issue? Internet service providers based in countries remote from Internet backbones, particularly in the developing countries, must pay the full cost of the international circuits.
2. What are the implications for developing countries? Traditionally, the settlement mechanisms that applied to international telephony have been a source of revenue for developing countries that has helped to subsidize universal service and/or to finance investment in telecommunications infrastructure. As more telecommunication traffic is shifting to the Internet, this revenue is disappearing. The high costs of connecting to the global Internet result in the diversion of many millions of dollars that could otherwise be used for network development. Moreover, dominant public operators typically pass these costs along to independent Internet service providers and customers, thereby helping to suppress the Internet’s growth and utilization in government, business, and society within developing countries.
3. How much are we talking about? The ITU estimates that, between 1993-98, net flows of telecommunications settlement payments from developed counties to developing ones amounted to some $40 billion.
4. Could a solution be achieved through free markets and competition? Some think that liberalization and privatization of developing countries’ telecommunications industries will provide sufficient incentives for the deployment of backbones in the global South that would reduce the reliance on costly connections to the industrialized world. However, others believe that these measures alone have already proved insufficient to tackle the problem, inter alia, due to lack of incentives for the requisite backbone deployments.
The costs of Internet based telecommunication services are indeed much lower where the infrastructure is already in place. This is not the case for developing countries, where in many cases new and lower cost technologies, like Voice over Internet Protocol (VoIP), are seen as threats because they deprive the country of revenue needed to modernize their infrastructure and to deploy widely new technologies such as Internet. This is regardless of whether the country has a liberalized competitive regime or the traditional monopoly regime. On the other hand, from the consumers’ perspective, there is the advantage of reduced cost for services.
6. So is this a ‘Digital Divide’ issue? Yes indeed, and it also cuts across other issues such as: ‘equitable distribution of resources’ and ‘access for all’. It is also related to other political and developmental aspects, such as: ‘investment and funds for development’, ‘international cooperation’, etc. Finally this could become critical to a ‘stable and secure functioning of the Internet’.
7. Was this issue raised on the first phase of WSIS? This issue is directly mentioned in WSIS Plan of Action C2. 9. k): “Internet transit and interconnection costs should be oriented towards objective, transparent and non-discriminatory parameters, taking into account ongoing work on this subject.” Also, a possible way to find a solution to this problem is mentioned in the Plan of Action C2. 9. j): “The creation and development of regional ICT backbones and Internet exchange points, to reduce interconnection costs and broaden network access.”
8. What is the ongoing work in this subject? ITU Study Group 3 started discussing the issue of international Internet connectivity in 1998. In October 2000, the ITU World Telecommunications Standards Assembly approved the ITU-T Recommendation D.50 regarding ‘peering’ or ‘transit’ arrangements between ISPs and Internet backbone providers. The Recommendation is voluntary and suggests that parties involved take into account the possible need for compensation for elements such as traffic flow, number of routes, geographical coverage and the cost of international transmission among others when negotiating commercial arrangements. The Recommendation was hotly contested and is not being implemented, most notably by key industrialized countries and elements of the global private sector. The issue has also been debated at length within APEC TEL and has been raised in the WTO.
Some countries have suggested that provisions from these Recommendations should be included in the International Telecommunication Regulations (ITRs) so as to give them binding force. They have also suggested that there should be a binding international dispute resolution mechanism (similar perhaps to that which exists in the WTO) in order to resolve disputes concerning these matters.
9. What are the results so far? Study Group 3 agreed in June 2001 to pursue further studies on international Internet connectivity (IIC), and established two Rapporteur Groups, one for developing further guidelines to facilitate the implementation of Recommendation D.50, and the other for examining the possibility of using traffic flow as a main factor of negotiation for IIC. In June 2004, Study Group 3 adopted Amendment 1, on “General considerations for charging criteria and options for international Internet connectivity”, which complements Recommendation D.50. However, the study on the traffic flow methodology was not concluded and work continues during a new study period 2005-2008. ITU’s Council Working Group on ITRs is studying the matter and will report to the ITU Council in 2005.
|
60. The ITU is responsible for global organization of radio spectrum policies and agreements, with regional organizations in each region of ITU playing an important role. National governments are responsible for radio spectrum policies and management within their jurisdiction.
61. Physical infrastructure standards are developed by a wide range of technical organizations. Each has its own method of determining the status of standards. Several examples include:
National voting on recommendations: e.g., the International Organization for Standardization (ISO), International Electrotechnical Commission (IEC), European Technical Standards Institute (ETSI));
One company one vote (many industry fora);
Full consensus of the membership (ITU);
‘Rough consensus and running code’: demonstration of interoperability and proof that things actually work, with review/approval by technical management which has been put in place by a nominating committee drawn from individual participants (IETF).
Generally, however, market forces are more dominant than governance mechanisms in determining if or when standards are adopted.
Coordination
63. Coordination of activities relating to telecommunications infrastructure at the international level is through the various Sectors and Study Groups of the ITU on the one hand, and internationally negotiated private sector arrangements on the other, and generally functions well. International coordination of radio spectrum through the ITU Radio Sector (ITU-R) is generally said to be excellent.
64. Despite the multiplicity of organizations involved in developing standards, the limited and overlapping memberships of many of these mean that coordination is generally good. Nevertheless, problems may arise when technical bodies are not well informed of critical policy issues, and vice versa13.
Overall assessment
International level governance of the telecommunications infrastructure is mainly under auspices of intergovernmental treaty organizations on the one hand and private sector owners of infrastructure on the other.
Concerns expressed, especially by developing countries, about international Internet charging arrangements have resulted in ongoing discussions in ITU Study Group 3 in relation to a mechanism for implementing ITU Recommendation D.50. However, some feel that it is not clear that any such mechanism would overcome the lack of transparency inherent with commercial arrangements between the contracting parties, nor increase opportunities for multi-stakeholder participation. On the other hand, recent OECD work indicates that a competitive transit market does exist. Some have suggested that, to the extent that WTO Member States include the Reference Paper as part of their specific commitments in telecommunications, the application of the regulatory principles contained in the Reference Paper to Tier One telecommunications services provided by operators in that Member State could be clarified, and that the WTO agenda for market liberalization and increased competition might well provide a different basis for dealing with the issue and, hence, approach to solutions. Others have suggested that this is primarily a development issue.
67. Some believe that current mechanisms for allocating spectrum work well and are satisfactory including, for example, spectrum auctions, which are believed to be more transparent and efficient under specific conditions than beauty contests. Others are disturbed at what has happened in the past, including with domestic auctions of mobile telephony spectrum. In addition, some note the upcoming spectrum management and interconnection challenges that will be introduced by new technologies such as WiMax and similar technologies and note that careful regulation will be required so as to enable adoption and create the maximum benefits possible.
68. Some new technologies (both wireless and terrestrial), because they are inexpensive and effective, allow users to create and share their own community networks. However, existing operators, and some governments, are wary of such developments, fearing that they may upset existing business models, which might result in unfavorable impacts on universal access, competition, and/or research and development. Some perceive the actions or inactions of governments and operators as attempts to outlaw or reduce the phenomena of radical new developments, as they are afraid of the loss of control and revenues potentially associated with them. On the other hand, some stakeholders see such developments as being effective and innovative means of extending access to online networks and are encouraging these entrepreneurial activities.
69. The highly technical nature of standards development and promulgation means that many organizations and countries that might wish to have a voice, or to closely track and anticipate developments, lack the capacity to become involved in organizations engaged in standards development work.
70. Generally, direct participation by those parts of the private sector that have greatest interest in these issues is high, as sector members in the case of the ITU, or through membership of industry organizations such as the Institute for Electrical and Electronics Engineers (IEEE).
71. On the other hand, participation by civil society in governance mechanisms is low. Civil society organizations have said that they are not able, for financial reasons, or because they lack the resources or expertise, to become sector members of the ITU, or to join other organizations. It has been suggested that the ITU should extend its outreach activities to enable and encourage civil society organizations to engage with its work and that free access to ITU publications, which is presently available only to members, should be made more readily available.
72. There is broad agreement on the need to encourage, and provide opportunities for, full participation by actors from developing countries, including governments, the private sector and civil society organizations, in international fora such as ITU-T, and in industry forums such as IEEE and IETF where standards are agreed. Increasing participation from the developing world can be seen as a key development and capacity building need.
3. Critical Internet resources
73. This section summarizes issues relating to the management of critical Internet resources, including:
Administration of the domain name system;
Administration of IP addresses;
Administration of the root zone file and root server system;
Multilingualism of the domain name system;
Voice over IP (VoIP).
(a) Governance mechanisms
74. The governance of critical resources has been motivated by a combination of bottom-up operational development, standards and process documents, and a variety of contracts and agreements that originated during the formative period of the Internet and that have evolved with the Internet. Consequently, a wide variety of governance mechanisms and actors have formed or adapted to address the many emerging Internet governance issues.
(i) Domain name system (DNS) management
75. In the management/governance of domain names, the main actors are ICANN and organizations under the ICANN umbrella: Governmental Advisory Committee (GAC), Generic Names Supporting Organization (GNSO) and the Country Code Names Supporting Organization (ccNSO), the IANA functions, and the related international and regional country code top-level domain (ccTLD) organizations, including the Council of European National Top-Level Domain Registries (CENTR), the Asia Pacific Top-Level Domain Association (APTLD), the Latin American and Caribbean Top-Level Domain Association (LACTLD), the North America Top-Level Domain Organization (NATLD) and the African Top-Level Domain Association (AFTLD). The participation of governments in ICANN decision making occurs through the GAC; that of civil society occurs through the Non-Commercial Users Constituency of the GNSO, in respect to generic domain name issues, and through the At-Large Advisory Committee (ALAC) for all issues within ICANN’s mandate. While the policy functions in each of the top-level domain (TLD) types, country code (ccTLD), generic (gTLD) and sponsored (sTLD) differ, the operational aspects are similar. There are concerns over TLD policy management, some of which have figured prominently in the discussions of the Geneva WSIS meeting. The ITU has a mandate from Member States to work in this area14.
76. With the ccTLDs, issues include:
Some have expressed concern that the formal arrangements15 provided in existing ICANN documentation requiring memoranda of understanding (MoUs) to be executed between ICANN/IANA and ccTLD registrars are inadequate. Others feel that generally arrangements between ICANN/IANA and ccTLDs have been stable and satisfactory;
The application of national sovereignty decisions to ccTLD management is evolving and is considered by some to be problematic. The GAC Principles and Guidelines for the Delegation and Administration of Country Code Top-Level Domains (the GAC Principles) “strongly encourage” that “ccTLDs (be) administered in the public interest, within the framework of … national public policy and relevant laws and regulations”. The GAC Principles further specify that the ccTLD manager “has a duty to serve the local Internet community as well as the global Internet community”. Some governments consider this to be an adequate and clear direction, while others believe that the directions are unclear and inadequate. However, the GAC has no decision making powers, and cannot determine policies. Despite recognition of regional registries (most recently in Africa), holding formal meetings around the world and attempts to introduce more diversity to its board membership, ICANN is still felt by some to have met with limited success in extending its outreach to all regions of the world;
There is disagreement about whether there is clear direction regarding the limits of national sovereignty in relation to delegation and redelegation. Some are of the view that formal international arrangements for the delegation and redelegation of ccTLDs are needed, while others are of the opinion that workable guidelines are constituted by the GAC Principles and the Best Practice document of the ccTLD Community on delegation and redelegation. The GAC Principles specify that redelegations of ccTLD management should be resolved nationally and in accordance with national laws, taking into account the views of all local stakeholders and the rights of the existing ccTLD Registry. Such redelegations are performed as an IANA function and require confirmation of the agreement of all parties involved in a redelegation.
With the gTLDs, areas of concern include:
The need for further development of policies for the management and further development of the domain name space which, due to the inherent complexity of the matter, impacts strongly on key issues such as the equitable distribution of resources, access for all, trademark maintenance costs, multilingualism, and others;
Whether ICANN makes use of the best method for allocating or reallocating gTLDs;
The level of competition in the domain name registry market which is still dominated by one big player, who controls more than 50% of the market, and whose players are still based in a few countries only;
The lack of clarity about the status of the .int TLD, where the planned transfer from IANA to an organization under the authority of the UN has been delayed several times;
The zone files of the three gTLDs (.gov, .mil and .edu), used predominately by US institutions, are linked to the authoritative root and part of the root zone file, but operate outside the ICANN system and do not currently have any contractual relationship with ICANN.
78. Some consider that the reason ICANN (or, more precisely, ICANN’s Generic Names Supporting Organization (GNSO), which is the entity that deals with gTLD policy) has not addressed these concerns is that the decision making power in the GNSO is primarily in the hands of the private sector, with a tiny minority role for users and no direct role for governments. Moreover, many of the industry constituencies in the GNSO are dominated by a small number of representatives of big players from a few countries.
79. Some have expressed concern about the nature of the exclusive relationship between the US Government and ICANN, which is expressed through three agreements:
The MoU between the US Department of Commerce (US DoC) and ICANN, which terminates at the end of September 2006, the terms of which require that ICANN take the necessary steps by that time to assure the US DoC and the Internet community that ICANN is able to effectively carry out its important core technical missions in a stable and sustainable manner into the future. However, it has been noted that this also provides an opportunity for constructive developments and underlines the importance of establishing, in a timely manner, the appropriate governance structure and mechanisms consistent with the WSIS principles, including those relating to the roles and responsibilities of all the stakeholders, to ensure stable and secure functioning of the Internet into the future;
The contract between the US DoC and ICANN for performance of the IANA function, which regulates the special elements of the IANA service;
The Cooperative Research and Development Agreement (CRADA) between the US DoC and ICANN, which regulates the special treatment of proprietary information, which has no fixed date for a termination and specifies that the US DoC has the right to terminate should control of ICANN be transferred to a foreign company or government.
80. Some are concerned about the nature of the contract between the US DoC and VeriSign Inc. relating to the operation and management of the root servers.
81. There are concerns about the absence of privacy protection with respect to access to information on the WHOIS database16. On one hand the registrars require accurate and detailed identifying information of applicants for operational purposes. On the other there is a need to protect the privacy of individuals and to prevent crimes such as identity theft.
82. The composition and mechanism for selection of ICANN Board members is of concern to some. Currently, the Board of ICANN includes six members selected by the industry through the Supporting Organizations, eight members selected by a Nominating Committee formed by ten voting members appointed by private sector constituencies, and seven voting members appointed by civil society constituencies. Given the nature of this type of process, where confidentiality and full consensus are required, there is no transparency to the public of the nominations considered and of the evaluations made.
83. While the general public interest and the consumers’ side is intended to be represented through the eight Board members selected by the Nominating Committee, civil society groups think that this mechanism is not sufficiently effective to serve that purpose, and that the present structure of power is unduly imbalanced in favour of the private sector. According to these groups, the principles embodied in the original ICANN structure – half of the Board selected by industry and the other half selected by consumers through an open process involving a vote by the general public – should be used as a basis to adjust the current structure and to create a better balance between the representation of private sector interests and general public interest.
(ii) IP address management
The management/governance of IP addresses is primarily in the hands of not-for-profit private sector organizations, that is ICANN/IANA and the five regional Internet registries (RIRs). The ITU has a mandate from Member States to work in this area17. The RIRs have demonstrated their capability to fulfil their tasks and to make their contribution to the functioning of the Internet. They have been able to manage the transfer of functions from the American Registry for Internet Numbers (ARIN) to the Latin American and Caribbean Internet Addresses Registry (LACNIC) and from ARIN, the Réseaux IP Européens Network Coordination Centre (RIPE NCC) and the Asia Pacific Network Information Centre (APNIC) to the African Network Information Centre (AfrINIC) in a way which did not only not interrupt Internet services and connectivity for end users but also developed broader opportunities for the regional Internet community in regions served by the new RIRs. This demonstrates the ability for change and improvement in the Internet management structure without causing disruption to the functioning of Internet services. IP numbers are considered a valuable international resource to be shared by all users in all nations.
Concerns have been expressed in the following areas:
Some have expressed the view that the current management system and distribution of responsibilities for number allocation has worked well, and that realistic projections of requirements indicate that there will not be any problems for the foreseeable future. Others have argued that, because of the rapid increase of demand and utilization of the Internet, a review of the current numbering management is required to ensure equitable distribution of resources and access for all into the future;
There is currently limited involvement of either governments or civil society in the policy making or practical management of IP addresses, although generally RIRs encourage such groups to participate in RIR policy development. Some governments have the position that the allocation of IP numbers, or some subset of these numbers, should be under the sovereignty of national governments and should be managed via a national Internet registry (NIR). Note that in some regions there are already NIRs, which respond to regional needs; the allocation of IP addresses from RIRs to NIRs is made on the normal basis of documented need. Some other governments and stakeholders do not see the need for any change to the existing system. Some consider that making a change in the address allocation mechanisms would result in operational risks, for example, with routing aggregation.
(iii) Root server system management
The root zone file contains records for all TLDs and is managed by the distribution master root server.
There are many steps involved in the root server system: standard setting, initiation, selection, editing, and the IANA functions of allocation, authorization, publication and mirroring. Some of the activities are purely technical or operational while other activities include a public policy dimension. The system is managed on the basis of numerous bilateral and multilateral cooperative agreements, MoUs, sponsorship agreements, contracts, statements of work, and voluntary arrangements. The main actors in management/governance of the root zone file and root name servers are ICANN, the IANA functions, the US Department of Commerce (US DoC), VeriSign Inc. and the root server operators themselves.
88. The operators of root servers restrict themselves to operational matters and are not involved in policy making and data modifications. Some have expressed concerns about the current situation and consider that the following issues should be addressed:
They have no clearly defined responsibilities and accountability, especially in relation to the stability and secure functioning of the Internet;
The decision making procedure for the authorization of the publication of modifications, additions or deletions to the root zone file or associated information that constitute delegation or redelegation of top-level domains (‘approval’ of IANA function recommendations by the US DoC) is neither multilateral nor democratic and does not involve other governments, private sector, civil society or international organizations;
The existing system is mainly based on trust, not on a treaty. The system reduces the governmental participation in the authorization of modifications, additions or deletions to one single government, which has no contractual relationship with other governments with regard to the execution of this function.
(iv) Multilingualism
Governance issues associated with multilingualism are closely linked with DNS governance issues. The IETF is responsible for technical standardization and maintains its role as a technical standard discussion forum, generator, and publisher. ICANN is responsible for policies including confirmation of language code tables, decision to support multilingual TLDs, registration policy for script variants, etc. Currently, countries in Asia and Africa (generally the actors are the national network information centres (NICs) under the support of the their governments) are actively participating in such activities. The ITU liases and cooperates with relevant entities concerning internationalized domain names (IDNs), and provides information to its membership in order to promote effectively the role of Member States in the internationalization of domain names and addresses in their respective languages. Other actors involved in this issue include UNESCO, the Multilingual Internet Names Consortium (MINC), ISO, and national governments.
Concerns have been expressed that:
Despite the efforts to date, insufficient progress has been made towards providing solutions to IDN issues. Unsolved issues include stable versions of alphabets for digitization, rendering of characters which include versions and variants of ‘.’, agreements on the use of alphabets and scripts that are shared by many countries, languages, and cultures, security matters such as homographic attacks, e-mail addresses, and keyword lookup;
For global deployment of IDNs, international coordination is key. The current mechanism of governance has an evident weak point in the capacity for international coordination, particularly among sovereign state governments and intergovernmental organizations, which must continue their efforts in an environment that is particularly sensitive to WSIS criteria;
An orderly and well-studied process of global deployment of IDN is necessary to prevent cybersquatting, phishing and user confusion18. For example, no clear attribution of responsibilities has been made in regards to the definition of variant tables and equivalences; no procedure has been considered to protect existing ‘romanized’ registrations. In the absence of clear rules, processes and responsibilities, the risk is that the adoption of IDN will be stopped (some browser makers have already disabled IDN from their products);
The current market led approach to IDN only tends to maximize the number of domain names that are sold. However, there might be cases in which global public service issues should be considered – for example, whether gTLDs should be required to support all scripts, including minority scripts that might not be commercially viable. Without these considerations, IDN might become available only for scripts used by big countries and communities, thus contributing to the loss of linguistic diversity.
(v) Voice over IP (VoIP)
Overall, for Voice over IP (VoIP), any applicable regulation is mostly dealt with at the national level. National regulators are dealing with the question of how, or indeed if, VoIP services are regulated alongside conventional telephony services. Of particular concern are univers