The Global Multistakeholder Meeting on the Future of Internet Governance (NETmundial)

This case study by the Center for Technology & Society at the FGV School of Law, Rio de Janeiro examines NetMundial, the 2014 Global Multistakeholder Meeting on the Future of Internet Governance. It sought to develop a set of universally acceptable Internet governance principles, which serves as an innovative example of the use of technological tools for both planning and participation.

mikiwoz: CC BY-SA 2.0Photo: mikiwoz (CC BY-SA 2.0)


The Global Multistakeholder Meeting on the Future of Internet Governance (NETmundial)

Authors: Marilia Maciel,[1] Nicolo Zingales, and Daniel Fink [2]
Center for Technology and Society of the Getulio Vargas Foundation

Abstract: This case study examines NETmundial, the Global Multistakeholder Meeting on the Future of Internet Governance, which was held in Sao Paulo, Brazil on April 23 and 24 of 2014. The meeting was convened by 1net, a coalition of stakeholder groups involved in Internet governance discussions, in partnership with the Brazilian Internet Steering Committee (CGI.br), in response to revelations of mass surveillance of communications by the United States. It sought to develop a set of universally acceptable Internet governance principles as well as a way forward for the evolution of the Internet governance system, which together could serve as a framework for the governance and use of the Internet. It convened 930 participants from 110 different countries, representing civil society, the private sector, academia, the technical community, governments and intergovernmental organizations, as well as over 1000 remote participants from 23 countries around the globe. It also employed a content contribution platform that sought to crowd source inputs from stakeholders for the production of the outcome document. The meeting served as a demonstration of the multistakeholder process in action: in the production of the outcome document, stakeholders with a diverse range of backgrounds and interests collectively negotiated the inclusion or exclusion of highly sensitive and complex issues. While the process of achieving rough consensus involved sometimes messy debates and there were procedural imperfections, the case is informative for its structured production of bottom-up multistakeholder outcomes.

[1] Marília was a member of the Executive Multistakeholder Committee, which played a central role in the organization of NETmundial.
[2] Daniel was the Executive Director of the NETmundial Secretariat.

Table of Contents

I. Introduction
II. Values and Function
III. Organizational Model and Structure
IV. Participation and Decision-Making Structures
V. Outcomes
A. Political Outcomes of the Meeting
B. Outcome Document
VI. Best Practices and Templates

I. Introduction

The Global Multistakeholder Meeting on the Future of Internet Governance (NETmundial) was held on April 23rd and 24th, 2014 in São Paulo, Brazil. The event was organized by means of a partnership between the Brazilian Internet Steering Committee (CGI.br)[3] and 1net,[4] a coalition that gathers actors from the various stakeholder groups involved in Internet governance discussions.

Revelations of mass surveillance of digital communications were a catalyst for the decision to convene NETmundial, although the issue of surveillance was not initially included in the agenda of the meeting. In September 2013, Brazilian President Dilma Rousseff gave a speech at the opening of the 68th Session of the United Nations General Assembly, in which she criticized policies of mass surveillance.[5] President Rousseff also mentioned the need to develop a framework for the governance and use of the Internet, and to create mechanisms to ensure basic principles are guaranteed, such as privacy, freedom of speech, and net neutrality. The following month she received a visit from Fadi Chehadé, Chief Executive Officer of the Internet Corporation for Assigned Names and Numbers (ICANN). After this conversation, the global meeting was jointly announced.[6] This partnership helped both actors advance some of their objectives.

Brazil is one of the most politically active developing countries in Internet governance discussions. It has consistently stressed the need for a “multilateral, transparent and democratic governance of the Internet”,[7] advocating for changes, especially with regards to Internet governance institutions and the United States’ perceived unilateral oversight thereof.[8] In some interviews with the press, the chair of the NETmundial meeting, Virgilio Almeida, explicitly mentioned the goal of strengthening the multistakeholder model[9] and introducing changes to the contractual relationship between ICANN and the United States Department of Commerce.[10] Both topics—the need for democratic and multistakeholder governance and the transition of the stewardship of the IANA functions—were intensively discussed at NETmundial.

For ICANN and the I* organizations (read “I star”),[11] the meeting was viewed as an opportunity to counterbalance government-centric multilateral proposals for improving the Internet governance ecosystem. Those proposals emerged in the World Conference on International Telecommunications (WCIT), held by the International Telecommunications Union (ITU) in 2012, and in the ongoing World Summit on the Information Society (WSIS) +10 process.[12] During WCIT, some participants attempted to include the Internet within the ITU’s scope over telecommunications and introduced a non-binding resolution that blurred the lines between telecommunications and the Internet. There has been a similar attempt to strengthen government-centric multilateral governance over the Internet. For example, in July 2014, a UN resolution about the goals and modalities for participation at WSIS +10 was approved,[13] calling for the establishment of an intergovernmental negotiation process that should include informal consultation with other relevant stakeholders.

A lack of alternative proposals raised concerns that the pressure to move towards more traditional, top-down inter-governmental arrangements would increase.[14] NETmundial was seen as an opportunity to start the discussion about other options for the evolution of the Internet governance ecosystem and to gain a better understanding of what a new international consensus might look like.[15]

This case study will examine how NETmundial was organized and operated, with a focus on the process of drafting the outcome document, named the NETmundial Multistakeholder Statement. The process of drafting the document in a bottom-up and multistakeholder way is a key legacy of the event. It is important, therefore, to document and analyze the methodology of NETmundial, not only to capture what worked well, but also to identify aspects for improvement of this unique experiment.

[3] Brazil has extensive experience in multistakeholder discussions of Internet-related issues. The primary example is the Brazilian Internet Steering Committee (CGI.br), which currently encompasses 21 members. Ten are governmental representatives and 11 are non-governmental actors. Among other activities, CGI.br is responsible for coordinating joint actions for the proposal of policies and procedures related to the regulation of Internet activities, deliberating on any matters referred to it that relate to Internet services in Brazil, and adopting the necessary administrative and operational procedures for the management of the Internet in Brazil in accordance with accepted international standards. CGI.br. “About CGI.br.” Accessed January 6, 2015. http://cgi.br/about/.
[4] 1net was created at the end of 2013 with the purpose of providing an inclusive and open platform for discussion of Internet governance matters among interested actors. The platform was launched quickly so it could facilitate the involvement of stakeholders in the preparatory process of NETmundial. NETmundial was the first and most important topic in the agenda of 1net so far and participation in the mailing list has declined following the event. 1net. “1net Homepage.” Accessed January 6, 2015. http://1net.org/.
[5] Dilma Rousseff, “Statement By H. E. Dilma Rousseff, President Of The Federative Republic Of Brazil, at The Opening Of The General Debate Of The 68th Session Of The United Nations General Assembly, 2013,” September 24, 2013, http://gadebate.un.org/sites/default/files/gastatements/68/BR_en.pdf.
[6] “Brasil vai sediar encontro mundial sobre governança da internet em 2014,” Blog do Planalto, October 9, 2013, http://blog.planalto.gov.br/brasil-vai-sediar-encontro-mundial-sobre-governanca-da-internet-em-2014/.
[7] This expression emerged for the first time during the Regional Ministerial Preparatory Conference of Latin America and the Caribbean for the World Summit on the Information Society (WSIS), as part of the Bávaro Declaration. The Tunis Agenda, one of the documents approved by the end of WSIS, mentions that Internet governance should be “multilateral, transparent and democratic, with the full involvement of governments, the private sector, civil society and international organizations.” In the 68th UNGA president Dilma Rousseff called for “Open, multilateral and democratic governance, carried out with transparency by stimulating collective creativity and the participation of society, Governments and the private sector.”
[8] Everton Lucero, Governança da Internet: Aspectos da formação de um Regime Global e Oportunidades para a Ação Diplomática, (Barsília: Fundação Alexandre de Gusmão, 2011), accessed January 6, 2015, http://www.funag.gov.br/biblioteca/dmdocuments/0514.pdf.
[9] “Conferência Multissetorial sobre Internet no Brasil deve ter 150 países participantes,” Teletime, January 28, 2014, http://www.teletime.com.br/28/01/2014/conferencia-multissetorial-no-brasil-deve-ter-150-paises-participantes/tt/366869/news.aspx.
[10] Taíssa Stivanin, “Brasil sedia conferência mundial sobre governança na Internet,” RFI, April 23, 2014, http://www.portugues.rfi.fr/geral/20140423-brasil-sedia-conferencia-mundial-sobre-governanca-na-internet.
[11] The I * is a group of organizations responsible for coordination of the Internet’s technical infrastructure. The organizations include: the Internet Corporation for Assigned Names and Numbers, the Internet Engineering Task Force, the Internet Architecture Board, the World Wide Web Consortium, the Internet Society, and the five regional Internet address registries. These registries are: African Network Information Center (AFRINIC), American Registry for Internet Numbers (ARIN), Asia-Pacific Network Information Centre (APNIC), Latin America and Caribbean Internet Addresses Registry (LACNIC), and Réseaux IP Européens Network Coordination Centre (RIPE-NCC).
[12] The WSIS was convened in two phases. The first meeting took place in Geneva in 2003, with the aim to foster the political will to advance an Information Society for all. The main documents approved were the Geneva Declaration of Principles and the Geneva Plan of Action. The second phase took place in Tunis in 2005, with the aim to put Geneva’s Plan of Action into motion as well as to find solutions and reach agreements in the fields of Internet governance, financing mechanisms, and follow-up on the implementation of the Geneva documents. The main documents approved on that occasion were the Tunis Commitment and the Tunis Agenda for the Information Society.
[13] Un General Assembly, Resolution A/Res/68/302, “Modalities for the overall review by the General Assembly of the implementation of the outcomes of the World Summit on the Information Society,” August 13, 2014, http://www.un.org/en/ga/search/view_doc.asp?symbol=A/RES/68/302.
[14] Stéphane Van Gelder, “ICANN Explains Brazil Meeting Initiative,” CircleID, October, 23, 2013, http://www.circleid.com/posts/20131023_icann_explains_brazil_meeting_initiative/.
[15] Markus Kummer, “The Ever Evolving Landscape of Internet Governance,” in Stakes are High: Essays on Brazil and the Future of the Global Internet (Internet Policy Observatory, April 2014) http://globalnetpolicy.org/wp-content/uploads/2014/04/StakesAreHigh_BrazilNETmundial_final.pdf#markuskummer.

II. Values and Function

It is impossible to extricate NETmundial from the political context in which it was convened, marked by revelations of mass surveillance. During the announcement of NETmundial, Chehadé affirmed that “the trust in the global Internet has been punctured and now it's time to restore this trust through leadership and institutions that can make that happen.”[16] The perception that the chain of trust had been broken “created a sense of urgency to review current Internet governance arrangements.”[17] It is possible to affirm, therefore, that one of the goals of NETmundial was to restore a positive governance environment that would foster trust, cooperation and dialogue about common challenges.

NETmundial was explicitly aimed at addressing two problems that directly inspired its agenda items: the need to identify a set of universally acceptable Internet governance principles and the need to propose a way forward for the evolution of the Internet governance ecosystem. These concrete goals were defined in a meeting between the Brazilian government and representatives from the I* organizations.[18] They were further detailed by the members of the Executive Multistakeholder Committee of NETmundial (EMC) in two short paragraphs published on the NETmundial website.

Principles have been historically relevant for the development of Internet governance. The WSIS Declaration of Principles, for example, has been called a constitution for the information age.[19] In recent years, sets of principles aiming to guide Internet governance and policy development have been put forward by several organizations, leading to some to call this a “constitutional moment” for global Internet governance.[20] This bevy of principles includes those developed by the Council of Europe, OECD, European Union, the G-8, by the government of the United States, and by CGI.br. Efforts to map, compare, and identify commonalities among sets of principles have recently been undertaken, showing that there is considerable degree of convergence among them.[21]

Despite the similarity between documents, these sets of principles have yet to combine both global scope and multistakeholder support: most of them have been recognized only by a particular country or region and have not been developed with the involvement of all stakeholders. One of the aims of NETmundial was to “identify a set of universal Internet governance principles to be promoted as a global inspiration for the evolution of the Internet worldwide. Those principles should support the perspective of the Internet as a global platform for social, economic and human development and a catalyst to exercise human rights.”[22] In order to accomplish that, NETmundial participants were encouraged to build upon existing sets of principles, which were listed and made available on the website.[23]

The second agenda item of NETmundial, setting out a roadmap for the future of Internet governance, was less neatly defined, probably reflecting the diversity of topics it encompassed. The improvement of the ecosystem has been debated in several UN forums, such as the Internet Governance Forum (IGF),[24] in the International Telecommunications Union (ITU), in the WSIS process, and in the UN General Assembly. Discussions have also taken place in ad-hoc groups, such as the Panel on Global Internet Cooperation and Governance Mechanisms and in the Global Commission on Internet Governance, created by CIGI and Chatham House.

Given the multitude of parallel forums and the relative paralysis on the issue in the UN, a goal of NETmundial was “to energize discussion and pursue agreements for the way forward (...) to evolve and extend Internet governance institutions and networks to address current and emerging issues.”[25] Given the broad scope of topics, the EMC suggested a set of questions to facilitate contributions from participants.[26]

In addition to these explicit goals, the convening of NETmundial—a new event in an already crowded agenda of Internet governance meetings and forums—revealed the existence of a gap. There was no global multistakeholder space for actually drafting outcome documents. NETmundial was an alternative way to address this problem. The meeting was convened under the expectation that it should present conclusions and a concrete outcome. According to a report presented by the chair of 1net, a critical element for the success of the meeting would be “a final joint declaration of internet principles and an institutional framework for multistakeholder Internet governance. The declaration should aim to be concrete/practical, linked to prior/current Internet governance initiatives, and hopefully include some next steps.”[27]

It was a key objective of NETmundial that this outcome document would be developed through a bottom-up multistakeholder process. However, the methodology for drafting the document was not clear from the outset, with the exception of the fact that inputs should be channeled through online consultation in addition to face-to-face discussions during the two days of NETmundial.

[16] Van Gelder, “ICANN Explains.”
[17] Kummer, “Ever Evolving Landscape.”
[18] Adiel Akplogan, e-mail to I-coordination list, November, 18, 2014, https://www.nro.net/pipermail/i-coordination/2013-November/000077.html.
[19] Daniel Stauffacher and Wolfgang Kleinwächter, eds., The World Summit On The Information Society: Moving From The Past Into The Future (New York: The United Nations ICT Task Force, 2005), http://www.icvolunteers.org/files/wsis_past2future_ebook1.pdf.
[20] Wolfgang Kleinwächter, “Internet principle hype: how soft law is used to regulate the Internet,” .NXT: Internet Policy and Governance Dissected, July 25, 2011, http://news.dot-nxt.com/2011/07/27/internet-principle-hype-anon.
[21] See Jeonghyun Baak and Carolina Rossini, “Issue Comparison of Major Declarations on Internet Freedom,” Summer 2013, http://bestbits.net/issue-comparison-of-major-declarations-on-internet-freedom/.
[22] Executive Multistakeholder Committee (EMC), NETmundial, “Executive report Executive Multistakeholder Committee (EMC), 1st meeting,” January, 27, 2014, http://netmundial.br/wp-content/uploads/2014/01/EMC-Jan-27_Executive-report.pdf. [23] “Links to Internet Governance Principles,” NETmundial, http://content.netmundial.br/internet-governance-principles/.
[24] The IGF work took place through working groups created under the auspices of the UN Commission on Science and Technology for Development, such as the Working Group on IGF improvements and the Working Group on Enhanced Cooperation. [25] EMC, “1st meeting.”
[26] The guiding questions were: a) Is there a forum or Internet governance body that develops policy or technical outcomes involved in these issues?; b) If there is, how and why are these issues not being adequately dealt with by that forum or organization?; c) What are the possible responses to the challenges posed by these issues?; d) How will the possible responses proposed ensure the stability, resilience and efficiency and also comply with principles of equitable multistakeholder participation, accountability, transparency and predictability?
[27] Akplogan, e-mail.

III. Organizational Model and Structure

The organization of NETmundial was carried out by a multistakeholder board of chairs, appointed by the Chairman of the meeting to represent the four key communities (civil society, the technical community, academia, and the private sector),[28] and four organizational committees. The structure and the methodology for the composition of the committees was jointly determined by Brazilian government representatives and the I organizations.[29] It is worth noting the prominent role that the I organizations had on the shape of the meeting. 1net co-organized the meeting and its chair, Adiel Akplogan, represents AFRINIC among the I organizations. ICANN, which is also part of the I organizations, had a prominent role from the outset, financing part of the logistical expenses and working in close connection with the Brazilian government and CGI.br on the organization of the meeting. It is possible to say, therefore, that this subset of the technical community was the most influential among non-governmental stakeholders when it came to the initial formation of NETmundial.

The mandates of each committee were the following:[30]

  • High-Level Multistakeholder Committee (HLMC): Responsible for conducting the political articulation and fostering the involvement of the international community.
  • Executive Multistakeholder Committee (EMC): Responsible for organizing the event, including the agenda and execution, and for the review of the proposals from participants;
  • Logistics and Organizational Committee (LOC): Responsible for overseeing every logistical aspect of the meeting;
  • Council of Governmental Advisors (CGA): Open to all governments willing to contribute to the meeting.

Figure1Figure 1. The organization of NETmundial.

[28] The chair, Virgílio Almeida, was supported by Jeanette Hofmann (academic community), Subi Chaturvedi (civil society), Fadi Chehadé (technical community), and Andile Ngcaba (private sector).
[29] Akplogan, e-mail.
[30] “Committees,” NETmundial, http://netmundial.br/#committees.

The HLMC and the EMC were multistakeholder bodies. By contrast, the CGA was composed only of governments, and the LOC had representatives from the Brazilian government, CGI.br, ICANN, and 1net. In addition, an Executive Secretariat was established to provide support to all committee activities. It was composed of a team of five people, who were the only personnel with full-time dedication to the event and who were financially supported by the event. This group worked on site at the Brazilian Internet Steering Committee (CGI.br) in São Paulo.

The HLMC included 27 members: one chair (the Brazilian Minister for Communications), 12 representatives of States as co-hosts (invited by the Brazilian Ministry of Communications),[31] two from international organizations (appointed by the UN Secretary General), and 12 from the multistakeholder community (three each from civil society, the private sector, academia, and the technical community) nominated through 1net.

Figure2Figure 2. HLMC Composition.

The EMC was composed of: a) eight Brazilian members appointed by CGI.br (one co-chair from the technical community, one representative from academia, and two each from government, civil society and private sector); b) nine representatives from the global multistakeholder community selected through 1net (including one co-chair and one member from the technical community and two each from the private sector, civil society, and the academic community); c) finally, one from the United National Department of Economic and Social Affairs (UN DESA) (appointed by the UN from the staff of the international organizations). After the EMC was formed, a representative from the government of Egypt and a representative of the European Commission were added.

1net was responsible for selecting the global non-governmental members of the EMC and HLMC. This process was coordinated by the 1net steering committee. Each steering committee member was responsible for coordinating discussions within their constituencies to select individuals for the HLMC and EMC.

Figure3Figure 3. EMC composition.

[31] The twelve countries were: Argentina, Brazil, France, Ghana, Germany, India, Indonesia, South Africa, South Korea, Tunisia, Turkey and the United States of America. Each continent was granted three slots with the recommendation of involving other governments from their region.

The Logistics and Organizational Committee was composed of two members from CGI.br (one of them was a co-chair), a member of ICANN (also co-chair), a member of 1net, and three members from the Brazilian government (one from the Ministry of Justice, one from the Ministry of External Relations, and one from the Cabinet of the Presidency).

The Council of Governmental Advisors, which was formed by participating government representatives, was coordinated by the Brazilian Ministry of External Relations and co-chaired by the governments of China, Portugal, and Australia. The mandate of this council was to channel the content proposals and comments coming from government participants. The council had a preparatory meeting on April 22nd in Sao Paulo, one day before NETmundial. During this meeting, government representatives exchanged their impressions about the preparatory process for NETmundial, and also discussed their positions regarding Internet governance principles and the roadmap for the further evolution of the Internet governance ecosystem as proposed in the draft outcome document.

Due to the different mandates and characteristics of each committee, the working procedures varied and were adjusted during the organizational phase. A brief description of the dynamics developed by each group can provide a sense of the decision-making processes. The Executive Multistakeholder Committee (EMC), which had the mandate to determine the overall processes, had the highest number of meetings (eight in total) to prepare proposals related to participation, meeting format, and the production of the first draft of the outcome document. Most meetings occurred in CGI.br’s headquarters, but given the composition of the EMC, which included members from all over the world, all meetings included remote participation using Adobe Connect.

EMC decisions were constantly reported to the NETmundial chair by the Secretariat. Board meetings were conducted to review the EMC decisions and provide further input. Such meetings were the ultimate decision-making point and included the chair and co-chairs of NETmundial and the chairs from the EMC and HLMC. Their discussions mostly served to review and endorse the approaches adopted by the EMC, and any substantive suggestion from the board was taken to the EMC through its chairs.

The HLMC working process differed from the EMC. One reason for this was the high-level composition of the group and the resulting difficulty it had in gathering its members on a regular basis. A face-to-face meeting occurred during the GSMA forum in Barcelona on February 14th, 2014, since the event counted on the participation of several high-level representatives from the committee.

The second and final face-to-face meeting of the HLMC happened during NETmundial on April 25th, when the outcome document was presented by the EMC for final review before the closing ceremony. Besides those meetings, HLMC members reported having had little guidance from the chairs and little inter-committee communications, which led to a lack of clarity about the role they should play. One HLMC member reported “we have been left without guidance on the process and without work to do, till [sic] the EMC draft outcome arrived.”[32] This helps to explain the limited involvement of this committee during most of the organization process.

The creation of NETmundial committees was important for several reasons. First of all, it strengthened the bottom-up multistakeholder characteristics of the meeting. Secondly, it increased the transparency of the event, by giving stakeholders individuals that they could communicate with in case they had questions or suggestions to make. It also helped to maintain trust, even when procedural aspects of the meeting were not always clear or made known in advance. Finally, the committees became official instances of decision-making authority, which helped to minimize informal and power-based influence on the organization of the meeting.

It is interesting to notice that the composition of the committees endorsed the interpretation that the technical and the academic communities have the same status as other stakeholders—something that was not well established in the Tunis Agenda. In addition, NETmundial also recognized the academic community as a distinct stakeholder group. This is a positive trend, since academics have been historically identified with the technical community or civil society. These groups have their own common language and strategies to intervene in discussions, which are not always the same ones normally followed by academics. In order to take advantage of academics’ expertise, it is important to fully recognize them as a stakeholder group.

In spite of the positive aspects that the committees brought about, the way that they were populated presented some issues. First of all, the procedure that each constituency followed to appoint its representatives has not been made public, although civil society has discussed the selection process in its mailing list and the archives can be retrieved online.[33] Names were forwarded to the organizers of NETmundial through the 1net platform without procedural information about the selection. More transparency about the selection process could have increased the legitimacy of the committees and strengthened the connection between representatives and their stakeholder groups.

The composition of the EMC presented two main issues. Firstly, the group was unbalanced in terms of diversity. Nine of its members were from Brazil and ten were from Latin America. Considering the fact that the importance of the group grew throughout the preparatory process, including drafting the initial outcome document, this unbalance became increasingly relevant. The participation of other governments—only Brazil was an official member of the EMC—could have helped to minimize some of the criticism of the process. In particular, some governmental representatives expressed concern after the final plenary of NETmundial regarding the lack of transparency of the drafting process. According to the representative of the Russian Federation, “all the decisions were made by a specially formed committee and we were unable to understand the principles underlying the shaping of such a committee.”[34]

Secondly, the inclusion of special members was not based on any clear justification; they were just added to the mailing list of the EMC. This addition was well accepted by EMC members and did not stir controversy. However, despite the increased regional diversity the special members provided and the high quality contributions they offered, this late inclusion lacked transparency.

Finally, there was lack of clarity and adequate communication between the committees, especially between the HLMC and the EMC while drafting the outcome document. There was also some overlap and duplication among committees, particularly regarding the role of governmental representatives in the HLMC and in the CGA when it came to the tasks of providing political articulation and fostering the involvement of the international community.

[32] HLMC member, private communication with author, April 11, 2014.
[33] Civil society organized itself into a coordination group, a loose body with the aim to facilitate joint civil society participation in several nominating processes. At the time of NETmundial, it was comprised of representatives from the following civil society coalitions or networks: Internet Governance Caucus, Diplo Foundation, Best Bits, the Non Commercial Stakeholder Group of ICANN, and the Association for Progressive Communications. Among the criteria the coordinating group used to select its representatives, were: 1) Ability to represent civil society as a whole, not just her individual organization; 2) Ability to work collegiately with other stakeholder groups in a multistakeholder setting; 3) Ability to consult widely with civil society groups and to report back as the process progresses; 4) Ability to represent civil society at a senior level in these discussions; 5) Broad knowledge of Internet governance issues and the range of civil society perspectives on these issues; 6) Capacity to participate assertively and creatively. Jeremey Malcom, e-mail to Best Bits mailing, Dec. 22, 2013, http://lists.bestbits.net/arc/bestbits/2013-12/msg00272.html.
[34] Statement of the Russian Federation at the NETmundial closing session, April 24, 2014.

IV. Participation and Decision-Making Structures

In line with the spirit of WSIS and the Tunis Agenda,[35] the NETmundial process followed principles of openness, inclusiveness, and transparency. It aimed to involve governments, the private sector, civil society, intergovernmental and other international organizations, as well as the academic and technical community.[36]

Two different modalities of participation were used: 1) using online tools on www.netmundial.br for collecting contributions; and 2) participating at the NETmundial meeting.

A group of civil society organizations proposed online consultations to the Brazilian government in a meeting that took place during the IGF Bali (September 2013). The idea drew on the successful experience of the process leading up to the passage of the Brazilian Civil Rights Framework for the Internet (Marco Civil), a law that was drafted collaboratively online based on the inputs from a range of stakeholders. As a follow up to the IGF meeting, Best Bits, a network of civil society organizations, sent a letter to the hosts of NETmundial suggesting concrete mechanisms that would “facilitate purposeful deliberation and help to narrow down the meeting’s conclusions.”[37] Additionally, online platforms were used for consultation and facilitating deliberation during NETmundial.

The first phase of consultations (February 14 - March 8) consisted of collecting submissions through a dedicated web-form on the two agenda topics. A total of 188 contributions were received: 65 were related to principles, 69 were dedicated to the roadmap, and 54 combined the two topics. The majority of contributions were from United States (31) and Brazil (16), followed by the United Kingdom (7), India (7), and Switzerland (6).

From an inclusiveness and equality standpoint, it should be noted that submissions were only allowed in the English language, due to the limited time available for the analysis of contributions. Although this may have negatively impacted participation, the organizers felt compelled to sacrifice multilingualism at this stage, despite it being one of the guiding principles established by WSIS[38] and the Tunis Agenda.[39]

All the contributions were made available online, which added to the transparency of the process. Anyone interested could develop an analysis of the inputs. The Diplo Foundation conducted a data mining exercise, for example,[40] and the Center for Internet and Society in Bangalore offered a data visualization platform.[41] Since the participants would have only two days of discussions in São Paulo, meeting organizers wanted a draft outcome document in order to give attendees a starting point for discussion. The EMC was directed to produce this first draft based on the inputs provided in the public consultation.[42] The Secretariat of NETmundial assisted in this process by producing a summary of all the inputs.[43]

To prepare the draft outcome document, the EMC followed a few guidelines. Primarily, they opted to include the issues mentioned in a large number of contributions (relevance) and tried to address them in a way that made consensus possible (consensus-driven approach). In order to be included in the draft, each paragraph needed to achieve consensus among EMC members and also have a reasonable expectation of achieving consensus in the NETmundial plenary. EMC members tried to ensure the document reflected the content of inputs instead of representing the points of view of their own stakeholder groups. They were largely successful in impartially representing the inputs,[44] but there was no clear guideline about that before the drafting work started. It is important to notice, however, that these guidelines were not publicly made available, and were only informally discussed among EMC members. This may have led to a lack of understanding about the criteria for drafting the text and how conflicting views were resolved. According to the representative of the Russian Federation, “we are in the dark as to what principles were used to select which comments to incorporate and which not to incorporate.”[45]

[35] See WSIS, “Geneva Declaration of Principles,” WSIS-03/Geneva/DOC/4-E, Dec. 12, 2003, ¶ 48, http://www.itu.int/wsis/docs/geneva/official/dop.html; WSIS, “Tunis Agenda for the Information Society,” WSIS-05/TuNIS/DOC/6 (Rev. 1)-E, Nov. 18, 2005, ¶ 29, http://www.itu.int/wsis/docs2/tunis/off/6rev1.html.
[36] See Tunis Agenda, at ¶ ¶ 35-36.
[37] Best Bits, “Submission to Brazil meeting committees on deliberative processes for the Brazil meeting,” February4, 2014, http://bestbits.net/brazil-processes/.
[38] See WSIS Declaration, at ¶ ¶ 48, 52.
[39] See Tunis Agenda, at ¶ ¶ 29, 49, 53.
[40] Vladimir Radunovic, “Predicting NETmundial: What does data-mining the contributions tell us?,” Diplo Foundation, April 18, 2014, http://www.diplomacy.edu/blog/predicting-netmundial-what-does-data-mining-contributions-tell-us.
[41] Sumandro Chattapadhyay, “NETmundial- Which Countries Have Not Submitted Contributions to NETmundial?,” The Centre for Internet & Society, April 22, 2014, http://cis-india.org/internet-governance/blog/net-mundial-which-countries-have-not-contributed-to-net-mundial.
[42] All received contributions can be found at http://content.netmundial.br/docs/contribs.
[43] NETmundial, “Content contribution Summary,” March 17, 2014, http://netmundial.br/wp-content/uploads/2014/05/Content-Contribution-Summary_1703_final.pdf.
[44] Raúl Echeberría, “IGF 2014 Pre-Event: NETmundial Looking Back, Learning Lessons and Mapping the Road Ahead,” video recording, https://www.youtube.com/watch?v=Iw8xOJL1BoE&index=188&list=UUk0zf4oI0IsJLh1owvUQSfQ
[45] Statement of the Russian Federation at the NETmundial closing session. April 24, 2014.

In order to develop the draft outcome document, the EMC divided itself in two groups: one focused on the drafting of the principles section and the other focused on the roadmap section. One of the reasons for this was the unexpectedly large amount of contributions received during the consultation that needed to be reviewed by the EMC in a short timeframe to produce the draft outcome. Having two drafting groups working in parallel was the only way to make sure that a draft outcome document could be successfully produced in time. Moreover, specialization allowed EMC members to focus on the contributions related to the agenda item they were working on and to do a careful mapping of the different opinions. This mapping made it easier to identify the areas of potential agreement. The final work was reviewed by the whole EMC. The success of this specialization within the committee suggests that relatively small and issue-oriented groups may be a good way to structure similar drafting endeavors.

The second phase of online consultation followed the publication of the draft outcome document on the NETmundial website on April 11th. From April 15th to April 21st, the public could comment on the draft text. This phase generated 1370 comments, which were then assembled by the NETmundial data team into a spreadsheet.[46]

A website was specially prepared to receive public comments based on a dedicated system (Wordpress plugin) called “Commentpress,”[47] which allows short comments in specific parts of a document. The system requested users to identify themselves and special customizations were made for NETmundial’s purposes, such as asking contributors to identify their affiliated stakeholder group. Although there was no mechanism to validate the real identity of each contributor, each comment showed the name openly for further validation. No issues related to misbehavior or manipulations were registered; on the contrary, the availability of such a system was perceived as a good format to initiate the discussion before the meeting took place. The system remains available for review on the event website.[48]

Subsequently, the Secretariat extracted the most meaningful comments (based on the existence of a “rationale” for the suggested edit)[49] and produced a summary, which was published on April 22. In the document, the Secretariat acknowledged the imperfection of this consolidation, suggesting that the best and most transparent approach would be to go through all the comments one by one - something that the EMC endeavored to do in the single day remaining prior to the start of the meeting. However, given the time constraints under which the EMC operated, it is not clear whether comments were sufficiently analyzed. The EMC did not produce another version of the outcome document: online comments on the original draft and comments made in person by NETmundial attendees were incorporated simultaneously. If time was sufficient, it would have been useful to produce and publish an intermediary version of the draft outcome document incorporating the results of the online comments prior to the in-person comments.

To manage and accept registrations for on-site participation, the EMC was guided by various principles, such as openness, equality, and flexibility.[50] In this light, a decision was made to adopt a pre-registration procedure, inviting anyone with an interest in attending the meeting to sign up via a dedicated web form between February 3rd and February 28th. Expressions of interest allowed the organizers and the EMC to monitor the levels of attendance, keep track of the balance among the stakeholders, and develop early strategies to cope with the possibility of over-registration.[51]

[46] Spreadsheet with all the public inputs is available at: http://netmundial.br/wp-content/uploads/2014/05/NETmundial-Comments.xlsx.
[47] More information about Commentpress is available at: http://futureofthebook.org/commentpress/
[48] The interface for public comments is available at: http://document.netmundial.br/. [49] In the words of the Secretariat, “rationale” includes “even the most primary forms of supporting arguments. Some rationale are elaborated and go deep into the subject. Others are one-liners which will not develop the subject as much as other arguments. Nevertheless, despite the stark difference between those simpler rationale and deep arguments that elevate the level of the debate, they are something to start with and that is why they are included in the report.” See “NETmundial Draft Outcome Document—Public Consultation: final report on comments,” April 22, 2014, p. 8 (emphasis in the original), http://netmundial.br/wp-content/uploads/2014/04/NETmundialPublicConsultation-FinalReport20140421.pdf.
[50] EMC, “First Meeting.”
[51] A total of 869 expressions of interest were received coming from 94 different countries, divided into self-declared categories of academia (105), civil society (245), government (138), private sector (210), technical community (107) and “other” (64). The highest numbers of requests were, much like in the case of submissions, from Brazil (251) and the United States (136).

On March 15th, participants were informed of the outcome of their requests for participation. Due to interest exceeding available capacity, the EMC used criteria to select among participants. These criteria included: a limit of no more than five people from each organization (up to two from the same organization and also from the same country); up to three governmental representatives per country if the country was represented at a ministerial level and two in case of no ministerial representation. The EMC guaranteed that each country would have at least one representative.[52] The Brazilian Ministry of External Relations through diplomatic channels handled the registration of governments.[53]

Other general guidelines adopted by the EMC were intended to foster balance among stakeholder groups, giving preference (in the form of pre-registration) to those that showed previous experience with the theme and prioritizing the registrations of actors from underrepresented countries. Eventually, the total number of registered participants was 933, with 364 governmental representatives, 87 coming from academia, 160 from civil society, 141 from the private sector, 109 from the technical community, 186 from the press, and 51 from other categories (41 of which were classified by the Secretariat as from “international organizations”).[54]

The management of funding for the organizational aspects of NETmundial fell under the mandate of the LOC, who sought resources from CGI.br, ICANN, and the I* community. One fundamental question for the LOC was whether any travel support would be available. From an early stage, it was decided that the meeting would be similar to the Internet Governance Forum, with no funding officially provided by the organizers, but some funding would be offered by organizations that usually support Internet governance-related meetings. However, the LOC sought funding from those various Internet-related institutions with no success. Finally, at the beginning of April, approximately 3 weeks before the meeting, the Secretariat instituted a donor fund to support people who requested it. However, since participants who submitted requests were not informed until very late in the process, this prompted many of them to seek other sources of funding or abandon the idea of attending altogether.

Participation during the meeting was also made possible through online interaction. This was done in two different ways. First, participants could submit chat comments to the online moderator via Adobe Connect. This would then result in the question being read out loud by the moderator. Second, individuals could participate remotely via one of the 33 hubs available worldwide, spread across 30 cities and 23 countries. This distribution was the result of an open “call for remote hubs proposals” issued on March 18.[55] Ten of the 33 remote hubs were given the status of official hub on the basis of geographical coverage, adequacy of infrastructure, personnel provision, and costs. In these official hubs, connection would be possible via video, as opposed to merely by text. Moreover, the crucial difference between accessing the event at an official hub as opposed to Adobe Connect was that participants could benefit from simultaneous translation of the meeting into six languages (Arabic, Chinese, French, Portuguese, Russian and Spanish) in addition to English, and have their comments translated into English from one of these languages. Remote participation was a key element to broaden the scope of participants in NETmundial and to enhance transparency.

One of the most innovative features of the NETmundial meeting concerned the management of microphones for comments from the audience and from remote participants: four different microphones were provided for comments from each stakeholder group, and the opportunity to speak rotated to each microphone. The four microphones were allocated to: 1) civil society; 2) the business community; 3) the technical and academic communities; and 4) government and international organizations. Thus, equality among different stakeholder groups was ensured by having participants from each of these groups queue in different lines in order to make their comments. Each commenter was allotted the same time (a maximum of 2 minutes). In addition, on each rotation inputs would be gathered from remote participants—so there was a fifth microphone in practical terms—applying the same time limit. Thus, government representatives—eveb those on the ministerial level - stood in line waiting for their time to speak.

[52] EMC, “Executive Summary: Teleconference Meeting,” March 10, 2014, http://netmundial.br/wp-content/uploads/2014/01/EMC-meeting-report_1003_final.pdf.
[53] This explains why there were 138 expressions of interest from governments and 364 governmental representatives: not all of the government representatives filled out the web form.
[54] Chairman Virgilio Almeida provided the most recent participation data in the article “The Evolution of Internet Governance: Lessons Learned from NETmundial,” Internet Computing, IEEE, Sept.-Oct. 2014, http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6886132.
[55] “NETmundial opens opportunity for interested to host Official Hubs of participation,” NETmundial, March 18, 2014, http://netmundial.br/blog/2014/03/18/netmundial-opens-opportunity-for-interested-to-host-official-hubs-of-participation/.

Figure4Figure 4. NETmundial meeting structure.

The meeting started with an opening ceremony followed by welcoming remarks. Although these sessions were important for encouraging the participation of governments, they delayed the multistakeholder working sessions. The working sessions of the meeting revolved around the two agenda items. They were conducted by individuals that were invited to be the chairs of the working sessions, which should not be confused with the co-chairs of the meeting.[56]

The sessions of the first day, entitled “Principles Part I” and “Roadmap Part I,” gave rise to a lively debate. After the working day finished, drafting groups met in a separate room. Two separate groups were created, one for principles and another for the roadmap, similar to the structure the EMC used for preparing the draft outcome document. These groups were composed of the chairs of the working sessions and the members of the EMC (also called “advisors” in the document that announced the dynamics for the working sessions).[57] The groups made changes in the outcome document in order to reflect comments made in the online consultation and in the plenary sessions. The same process was followed at the end of the second day for the elaboration of the final text, after incorporating into the outcome document the comments made in working sessions III (entitled “Principles Part II”) and IV (entitled “Roadmap Part II”).

The drafting room was open to observers and the text was displayed on the screen of the conference room along with the modifications as they were made, allowing for transparency in the drafting process. Although the procedures for the drafting session explicitly prohibited any interference from observers, this prohibition was not strictly enforced and as a result several observers engaged in bilateral conversations with members of the drafting committee, influencing the drafting process.

Since it was not clear from the outset whether this form of lobbying would be permitted, some stakeholders were not in the drafting room. This put them at a disadvantage compared to the more organized set of participants (especially from governments and the business constituency) who were present and able to more consistently intervene in the process.[58] Furthermore, the lack of a sufficient number of trusted experts among the advisors to the chairs appears to be a weak point of the drafting process. In the roadmap drafting group, this lack was particularly felt when dealing with mass surveillance and privacy. While the source of the legitimacy for the drafting group was political—their constituencies chose them—the legitimacy of the assistants was based on expertise.

Finally, some controversy arose about the involvement of the HLMC in the drafting of the outcome document. This happened in two moments. When the first draft was produced by the EMC, the HLMC played a role in reviewing it before it went public. Some HLMC members presented comments individually, but there was no joint committee discussion about the draft. This lack of common position made it harder for the EMC to clearly assess if suggestions from individual HLMC members garnered enough support to be incorporated into the draft.

The second moment happened during a meeting to present the outcome document to the HLMC that was arranged right before the final plenary session. The organizers of NETmundial tried to express deference to the committee and probably expected the meeting to be a formality. Nevertheless, the HLMC decided to propose some substantive modifications to the text. Tension in the room rose and the meeting was close to a breaking point. Some HLMC members took advantage of the moment to advocate the inclusion of their own positions in the text (see section E on outcomes) and a few changes were accepted to save the outcome document. This meeting of the HLMC added an element of top-down decision making into the bottom-up process.

[56] The chairs of the working sessions on principles were Markus Kummer and Adiel Akplogan, and the chairs of the working sessions on the roadmap were Janis Karklins and Jeanette Hofmann.
[57] “NETmundial announces dynamics for working sessions,” NETmundial, April 20, 2014, http://netmundial.br/blog/2014/04/20/netmundial-announces-dynamics-for-working-sessions/.
[58] Gabrielle Guillemin, “Netmundial: Success or Failure?,” Article 19, April 29, 2014, http://www.article19.org/join-the-debate.php/143/view/.

V. Outcomes

A. Political Outcomes of the Meeting

According to most analyses “the most remarkable aspect of NETmundial was its process.”[59] NETmundial will be positively remembered for having achieved a bottom-up multistakeholder outcome document. The meeting triggered positive reactions throughout the Internet Governance ecosystem. During the IGF open consultations that took place in Paris (May 1-2 2014), there was overwhelming recognition of the relevance of NETmundial as a learning opportunity. NETmundial was also one of the main topics discussed in the IGF 2014, in Istanbul (September 1-5), with several workshop sessions and a pre-event dedicated to it.

Nevertheless, this positive understanding is not unanimous. During the closing session Russia, India, and Cuba expressed their reservations regarding the process and its outcome document.[60]

This public divergence seems to have driven further apart countries that pursued alliances in the past, such as India and Brazil,[61] and showed some of the difficulties in articulating political alliances among the BRICS nations (comprised of Brazil, Russia, India, China, and South Africa).

[59] Id.
[60] See Transcripts of NETmundial closing session, April 24, 2014, http://netmundial.br/wp-content/uploads/2014/04/NETMundial-23April2014-Closing-Session-en.pdf.
[61] During WSIS, the most promising diplomatic alliance for Brazil eventually seemed to be coordination with India and South Africa, two large democratic developing nations. The three members of the IBSA Forum had been working together on Internet governance, interconnection costs, and Intellectual property since the first phase of WSIS, following a proposal from South Africa. However, the IBSA framework agreement for Cooperation on the Information Society would only be adopted in 2006. In 2011 the Brazilian Ministry of External Relations sought support from CGI.br and civil society organizations to organize a multistakeholder seminar. The event discussed substantive issues deemed relevant for developing countries, such as access to infrastructure, commerce and services and institutional frameworks. At the end of the seminar, the government representatives of the three countries drafted a document focused on Internet Governance institutional arrangements. The document with recommendations was supposed to be improved until the IBSA meeting in Durban, but the Declaration of Tshwane only took of the recommendations; the document became a historical reference, not an official IBSA position. The conversations about this topic in IBSA went cold. In fact, the IBSA Forum in general is going through a period of low diplomatic activity.

B. Outcome Document

On a more substantive level, NETmundial’s output document represents several achievements. The outcome document does not have a binding nature, but it has become an important reference point for global discussions due to the multistakeholder approach used for its development. It placed human rights as the cornerstone of the Internet governance ecosystem, it reinforced the commitment to strengthen the Internet Governance Forum, it managed to cover topics that were considered sensitive and were initially not part of the agenda (such as mass surveillance), it supported distributed mechanisms of governance, it emphasized the importance of enhancing both democratic and multistakeholder participation, and it ultimately “represents substantial progress towards public interest-driven Internet governance.”[62]

In spite of the importance of the Multistakeholder Statement of São Paulo, there were several issues that were perceived as shortcomings in the outcome document. It is important to highlight them because there seems to be some misunderstandings about the rationale underlying the treatment of certain topics. This lack of clarity may indicate procedural imperfections in the drafting process, which made it difficult to achieve enough transparency and strike the appropriate balance between competing interests.

  1. Net neutrality: in the drafts of the outcome document there is reference to “equal treatment of all protocols and data” and also to “recognizing technical management principles for efficient and improved network operation” and that Internet governance should encourage “the free flow of data packets/information.” These references were removed. In turn, a bullet point on the final NETmundial Multistakeholder Statement merely recognizes the need to “continue the discussion of the Open Internet including how to enable freedom of expression, competition, consumer choice, meaningful transparency and appropriate network management and recommend that this be addressed at forums such as the next IGF.”

  2. Intermediary liability: while there was no specific provision on intermediary liability in the draft outcome document, the new text included the principle that “Intermediary liability limitations should be implemented in a way that respects and promotes economic growth, innovation, creativity and the free flow of information. In this regard, cooperation among all stakeholders should be encouraged to address and deter illegal activity, consistent with fair process.” This formulation is problematic in the eyes of civil society because the focus on economic aspects prevails over the protection of human rights. Furthermore, the second sentence of the text contains a provision which recalls the OECD’s controversial language of voluntary measures to deter infringement in accordance with “fair process”[63]—as opposed to the more stringent right to “due process”—which forms an integral part of human rights jurisprudence.

  3. Right to access, share, create and distribute information on the Internet: this right was qualified with the expression “consistent with the rights of authors and creators as established in law,” which allows the legislator to create exceptions to the aforementioned rights without explicitly requiring the respect of the conditions identified under human rights law, including necessity and proportionality for the attainment of a legitimate objective.

  4. Reference to “necessary and proportionate” in relation to the conditions under which mass surveillance may be permitted: this point was not included in the original draft document, but was included during drafting after having been repeatedly supported at the public comment period and at the meeting.[64] However, the final text has no mention of this particular limitation.

  5. Structural separation of IANA functions and ICANN policy process: the issue of separation was another point that was not included in the original draft document, but was raised by a large number of actors during the meeting and was included in the final draft that was submitted to the HLMC.[65] The paragraph that was approved followed a last-minute suggestion by ICANN, presented during the controversial HLMC meeting that occurred just before the plenary session. The final text features diluted language, according to which states “it is desirable to discuss the adequate relation between the policy and operational aspects.”

Finally, some of the most important and controversial issues on the agenda remain open after NETmundial, such as the roles and responsibilities of different stakeholders. Remanding the exact definition of these processes to another forum is a consequence of the fact that NETmundial was ill equipped (and never intended) to address highly detailed and complex questions. A similar argument, stemming not only from the technical but also from the political and economic complexity (due to the difficulty of reconciling opposing positions) can be made for other important issues contained in the non-exhaustive list for further discussion, such as jurisdiction on the Internet, net neutrality, and mechanisms of oversight or benchmarking of compliance with the principles: these problems call for complex assessments which make rough consensus more problematic, and therefore warrant separate treatment.

[62] “Association for Progressive Communications (APC) Statement on NETmundial,” APC, April 2014, http://www.apc.org/en/pubs/association-progressive-communications-apc-stateme.
[63] See OECD, “Communiqué on Principles for Internet Policy-Making,” The Internet Economy: Generating Innovation and Growth, http://www.oecd.org/internet/innovation/48289796.pdf. Civil society did not endorse the Communiqué in light of the vague terminology and the risk that this would lead to censorship. For an account of the civil society reaction, see Milton Mueller, “Civil Society Defects from OECD Internet policy principles”, Internet Governance Project, June 28, 2011, http://www.internetgovernance.org/2011/06/28/civil-society-defects-from-oecd-internet-policy-principles/.
[64] The Secretariat of NETmundial produced a second version of the outcome document which tried to reflect the comments received. It was treated as a supporting document to the EMC. This version explicitly mentioned “necessary and proportional.” During the drafting exercise, the same expression was included but it got removed after some members of the drafting group withdrew their support. This swing of opinion happened after conversations with some observers, which is evidence of the efficacy of lobbying during the drafting exercise.
[65] For more information about the separation of IANA functions, see Milton Mueller and Brenden Kuerbis, “Roadmap for globalizing IANA: four principles and a proposal for reform,” Internet Governance Project, 2014, http://www.internetgovernance.org/wordpress/wp-content/uploads/ICANNreformglobalizingIANAfinal.pdf.

VI. Best Practices and Templates

Most of the opinions expressed after NETmundial assert that its open process is a key legacy of the event. For many years, the production of outcome documents prepared in a multistakeholder manner with the involvement of a large number of global actors was deemed infeasible. It is important, therefore, to identify positive aspects and areas for improvements of this unique experiment, particularly with respect to transparency, inclusiveness, and accountability.

The list below is an attempt to summarize some lessons that can be taken from NETmundial, which were discussed in more detail in the previous sections of this case study.

Structured multistakeholder dialogue was key to building trust. After the Snowden revelations, trust among actors declined. “Fragmentation of the Internet [was] a very real risk,” according to some analysis at that time.[66] Instead of contributing to the distrust, Brazil came up with an approach for discussing topics in a collaborative fashion. In a moment of crisis, energies were channeled to a clear, common, and positive goal.

It was possible to produce bottom-up, multistakeholder outcomes in a global environment. For many years the production of outcome documents prepared in a multistakeholder manner with the involvement of a large number of global actors was deemed infeasible. NETmundial used an open process of consultations followed by drafting groups to effectively produce an outcome text.

Multistakeholder committees were an important tool for strengthening the bottom-up nature of NETmundial. The establishment of multistakeholder committees in charge of specific aspects of the meeting helped to channel concerns and opinions. Nevertheless, stakeholder groups were inconsistent in how they chose representatives for these committees. This point is raised by the NETmundial Multistakeholder Statement, which mentions that, “Stakeholder representatives appointed to multistakeholder Internet governance processes should be selected through open, democratic, and transparent processes. Different stakeholder groups should self-manage their processes based on inclusive, publicly known, well defined and accountable mechanisms.”

Lack of clarity on decision-making guidelines and drafting reduced accountability. The set of guidelines that were used by the EMC and the drafting groups to choose the topics and positions that were included in the outcome document should have been publicized prior to the public consultation. Moreover, there was insufficient clarity about whether the members of the drafting group represented the interests of their stakeholders or were limited to incorporating the content of public contributions. During the intermediary stages of the drafting, at least succinct reasons for choosing not to follow a particular set of suggestions could have been provided. At NETmundial, this was neither done in the process leading to the first draft of the outcome document, nor in the phase that took into account the comments made on that first draft. Repudiation of the final document by the representative from Russia, and the dissatisfaction of India and various nongovernmental stakeholders with the way in which inputs were fed into the text illustrated that the lack of clear criteria generates a sense of alienation from the process.

Flexibility in the procedures was crucial to the achievement of results within a limited time frame, but sufficient time for the assessment of contributions is essential to further democratic participation. Procedures followed by the NETmundial committees were largely inspired by the principle of flexibility, which made particular sense in light of the very limited time-frame. Nevertheless, flexibility should never be seen as a justification for failure to respect essential democratic principles, such as allocating sufficient time to choose committee representatives and justifying why some proposals were not included in the draft outcome document. In addition, the process could better respect the fact that stakeholders often have different needs when it comes to the time needed to develop a response. For example, governments have to follow an internal hierarchy before they are able to present an official contribution in a public consultation period. Moreover, depending on the content of the outcome document, they may also need to consult with their principals before endorsing it. Developing procedures that take into account these differences is key to promoting equal participation.

The specialization of tasks and the assistance of experts facilitated the drafting exercise. While drafting the first version of the outcome document, the EMC divided itself in two groups, one devoted to principles and other devoted to the roadmap. The final work was reviewed by the whole EMC. This division into relatively small and issue-oriented groups was a good way to structure the drafting process. In addition, the use of trusted experts to assist the drafting group on technical or other specific issues was helpful. While the source of the legitimacy of the drafting group was political–their constituencies should choose them–the legitimacy of the assistants was based on expertise.

Concrete management of the process and the setup of the meeting had an impact on equality among participants. As noted, an innovation of the NETmundial meeting was the creation of different microphone queues for each stakeholder group and for remote participants. This put every stakeholder in a situation of formal equality during the plenary sessions. In other stages of the process, however, this equality was maintained to a lesser degree. For instance, the involvement of certain governments through the HLMC (where governments had more seats than any other constituency) afforded them the opportunity to intervene with more weight in the last stage of negotiation of the outcome document. This meeting of the HLMC added an element of top-down decision making into the bottom-up process.

On-site participation was enriched and complemented by online consultations. NETmundial used online channels for enhancing openness and participation in the process of shaping the outcome document. Open-ended consultations, in which participants were able to contribute their ideas freely was combined with platforms that allowed adding comments to specific provisions of the text.

Remote participation mechanisms was key to fostering wider participation. NETmundial provided channels for the online participation of individuals and also encouraged the creation of hubs, where several people gathered together to follow the webcast of NETmundial and interact with the meeting. Remote participants were given similar opportunities for participation as onsite participants. In practical terms remote participation represented a fifth microphone with the same time limit as the other four onsite microphones.

More could have been done to improve multilingual participation. Multistakeholder participation ought to promote substantive, instead of merely formal equality. While the organization of NETmundial accommodated the needs of the speakers of seven different languages, it should be noted that the online preparatory phase, including the original submissions and the comments on the proposed text, was only possible in English. This may have led to underrepresentation of non-English-speakers and insufficient discussion of the topics that were of relevance to them. If the number of times an issue was raised was considered criteria for its inclusion in the outcome document, lack of linguistic diversity may have affected the outcome document.

[66] “Internet Society Urges President Obama to Effect Immediate and Meaningful Changes to U.S. Government Surveillance Practices,” Internet Society, January 15, 2014, http://www.internetsociety.org/news/internet-society-urges-president-obama-effect-%E2%80%98immediate-and-meaningful%E2%80%99-changes-us-government.