Proposed Bylaws Raise Major Accountability, Openness Questions at ICANN

October 22, 2002 - This month ICANN, the organization that oversees the Internet's key addressing and naming systems, will begin implementing long-awaited reforms. A key ICANN committee has proposed to replace ICANN's current bylaws with new ones; at the end of the month the ICANN Board of Directors will review the proposal.

The proposed bylaws would make sweeping changes to ICANN, such as the adoption of a mission statement, a broadly new process for selection of the Board, and a reconfigured policy process. But though the bylaws seek to address major problems at ICANN, they will not solve them. The proposed changes are insufficient to fulfill ICANN's pressing needs for both reform and credibility.

Background

ICANN was created in 1998 to manage the Domain Name System, the IP address hierarchy, certain protocol databases and the root server system. It is a unique organization - the private-sector manager for greatly important and publicly-valued resources that affect the Internet's basic operations. ICANN's public responsibilities create a requirement that it be accountable to and credible before the worldwide Internet community.

As ICANN's President, Stuart Lynn, has emphasized, ICANN has suffered greatly from persistent uncertainty about its structure and processes. Many questions have been raised by public interest advocates who continue to urge ICANN to adopt principles of accountability and good governance in line with the public importance of its activities. In particular, ICANN should show both its critics and its supporters that it is capable of undertaking reform to address four critical issues:

• ICANN needs a clear, bounded mission. A lack of clarity about what ICANN is and what it does hinders ICANN's ability to create effective, appropriate processes for its activities, as well as leaving unresolved the potential for "mission creep." Clear, concise limits on ICANN's mission will make ICANN more internally efficient, as well as promoting trust and broader participation by key stakeholders.
   
• ICANN needs to be accountable to the Internet community. ICANN's functions are key to the Internet's operation and potentially affect numerous online interests. Strong accountability will make ICANN a more effective and trusted manager, and will promote high-quality administration of these critically-important resources.
   
• ICANN needs to incorporate the voice of the public interest into its processes. ICANN's founding documents repeatedly emphasize the importance of ensuring that the public voice be adequately represented at ICANN. Effective representation of the public interest will help guide ICANN's activities and promote wise decisions for the future, while under-representation could lead to costly and difficult backtracking when mistakes are made. By committing itself now to developing such representation, ICANN would send an important signal to potential participants worldwide.
   
• ICANN's activities must be transparent to the public. For participants at ICANN to be effective and productive, they need to have access to a broad range of information about ICANN's activities. In addition, transparency enhances ICANN's accountability by creating pressure for ICANN's policy bodies to conduct themselves fairly and deliberately at all times.
   

In March 2002, the ICANN Board kicked off its pursuit of these goals by tasking its Evolution and Reform Committee (composed of four members of the ICANN Board) with developing a long-term reform plan.

An important stage of the ERC's activity concludes with these new draft bylaws. The Board is expected to discuss the bylaws at its November 1 meeting in Shanghai, and will shortly move the reform effort from the drafting to the implementation phase. Ensuring that the bylaws properly set the stage for real, effective change at ICANN is therefore critical.

The New Bylaws Do Not Provide a Foundation to Achieve Necessary, Stated Goals

These new bylaws describe far-reaching changes at ICANN, including changes to the ICANN mission, policy structure, and accountability processes. Major provisions of the new bylaws include:

• Adds language describing ICANN's "Mission and Core Values," by which ICANN's general coordination mission is to be performed consistent with eleven values such as enhancing competition, promoting participation, speedy action, and others;
• Adds new transparency provisions, including the establishment of a new staff position for a Manager of Public Participation;
• Modifies the policy for Board reconsideration of past decisions;
• Creates an Independent Review Process for third-party review of ICANN's decisions;
• Calls for periodic review of ICANN's major committees, processes, and policy bodies;
• Creates a new staff position for a largely independent Ombudsman to handle questions and complaints from the community about ICANN;
• Revamps the Board of Directors to include two Directors from each Supporting Organization, plus eight chosen by a Nominating Committee (see below), plus the President;
• Creates a Nominating Committee to select members both of the Board of Directors and of the GNSO Council, to include representatives from business, operational, and individual users of the Internet;
• Divides the present Domain Name Supporting Organization into two new bodies: the Generic Names Supporting Organization (GNSO) and the Country Code Names Supporting Organization (CCNSO) - the GNSO structure is described in the bylaws and in an annex, the CCNSO structure has not been defined;
• Eliminates the present Protocol Supporting Organization, and maintains the Address Supporting Organization (ASO); and
• Creates or redefines a series of Advisory Committees, representatives of which will have non-voting liaison seats on the Board:
  • the Government Advisory Committee, to represent the voices of governments, constituted of government representatives;
  • the Technical Advisory Committee, to provide technical advice and guidance, constituted of individuals selected by the Board, nominated by ETSI, the ITU-T, and the IAB;
  • the Security and Stability Advisory Committee, to provide advice on matters relating to the security of the systems ICANN administered, constituted of individuals selected by the Board;
  • the Root Server System Advisory Committee, to provide advice about the operation of the Internet's thirteen root servers, constituted of the operators of those root servers as well as any other individuals selected by the Board; and
  • the At Large Advisory Committee, to provide input to the Board regarding the interests of individual Internet users, constituted of individuals selected by the Board until such time as the Committee can develop and the Board approves a mechanism for future selection of members.

ICANN's effort at reform is necessary and its goals, generally stated, are correct. Both the proposed bylaws and the accompanying "Final Report and Recommendations of the ERC" identify areas for improvement, but they lay insufficient groundwork for reform. In the space below, we identify several shortcomings in the new bylaws:

• Article I - Mission and Core Values: In its report, the ERC stated that "a clear and bounded mission is an essential pre-condition to addressing the identified problems of ICANN." Article I pursues this goal by describing ICANN's mission: to coordinate the DNS, the IP address hierarchy, certain protocol databases, and to conduct oversight of the root server system. It then hinges that mission on the pursuit of eleven "core values" for the organization.

  The statement that ICANN has a mission of "coordination" over four basic functions of the Internet does not restrict ICANN's authority within those realms. And though the "core values" attempt to provide some guidance within ICANN's mission, they are too vague and too open to interpretation to be truly useful. Future Boards could interpret their own core values any way they wished, and at any time. The provision allowing the Board to, when it feels it appropriate, emphasize certain values over others could create opportunities for a future Board to ignore some values altogether, leaving the door open for a wide range of inappropriate activities. The provision also diminishes the effectiveness of the mechanisms meant to enforce ICANN's mission - namely the Review and Reconsideration processes. As is discussed below, the unlikelihood that either of these processes will provide effective enforcement raises serious questions about the suitability of the proposed mission statement.

  As a result, Article I as drafted is not an adequate statement of ICANN's mission. Vague language and too much flexibility reserved to the Board undercut what could have been one of ICANN's most critical reforms.
   

• Article III - Transparency: The proposed Article III includes several useful suggestions for improving ICANN's transparency, including the creation of a Manager of Public Participation to assist the public in understanding and voicing its concerns about ICANN.

  However, Article III overlooks a key item in its failure to increase the openness of the Board's decision-making. The increasing tendency of the Board to make key decisions - including the upcoming redelegation of the .org domain - on closed teleconferences rather than in public runs the risk of giving ICANN the appearance of opacity. Better transparency on this front, such as a requirement that all key decisions be made in full public view, would boost participation from key parties and improve ICANN's overall credibility.
   

• Article IV - Accountability and Review: The requirement that ICANN act accountably is a touchstone of ICANN's commitments to the Internet community and its agreement with the Department of Commerce. The proposed Article IV partly pursues this goal by prescribing ICANN's adherence to its bylaws - it does not establish more general public accountability - but several clauses serve to emphasize the Board's immunity from real review and unnecessarily undermine the Article's effectiveness.

  Article IV describes two mechanisms: Board reconsideration and third-party review. In the first process, members of the community who feel they have been wronged by action or inaction by ICANN or its staff can submit a reconsideration request to be reviewed by the ICANN Board's Reconsideration Committee. The Committee, whose membership consists of some subset of the ICANN Board, would review the request and recommend to the Board that it review, or not review, the action/inaction in question. If a review is recommended, the full Board reconsiders the issue at hand, then determines whether new action is appropriate.

  Regarding the third-party review process, the draft bylaws propose that the Board enter into an arrangement with an international arbitration provider to establish an Independent Review Panel (IRP) for ICANN. Complainants would be able to submit ICANN actions believed to be inconsistent with ICANN's bylaws to a one-person panel convened by the IRP (a three-person panel may be convened upon request of either the complainant or ICANN, or both). After reviewing the request, ICANN's response, and relevant documentation, the IRP would issue a non-binding statement to the Board recommending a given course of action. Any time the IRP process is exercised, the party that does not prevail in the IRP's decision would pay the costs of the IRP.

  The Board Reconsideration process provides the Board with a means to review its past decisions, but it would not provide very strong accountability since the Reconsideration Committee would only of ICANN Directors. Its usefulness is further undercut by provisions empowering ICANN to recover "extraordinary" costs from complainants and empowering the Committee to determine whether the requesting party is adequately affected by an ICANN action to request reconsideration, both of which are potential obstacles to the process's effectiveness.

  Freed of the restrictions placed on it in the current draft, the Independent Review process could provide valuable accountability to ICANN. Unfortunately, the fact that the recommendations of the Independent Review Panel (IRP) are proposed as non-binding on ICANN leaves the IRP powerless except for the persuasive power of its arguments.
   

• Articles VI, VII, and XI - ICANN's Policy Structure: The proposed bylaws would revamp ICANN's policy structure, replacing it with an elaborate system that includes a new eighteen-member Nominating Committee to populate ICANN's policy bodies. Weaknesses in the NomCom proposal, and in the NomCom's proposed interactions with ICANN's other bodies, raise questions about the structure's overall effectiveness as a mechanism of representation and accountability.

  In the proposed system, the ICANN Board would have fifteen voting members: Two chosen by each of ICANN's three Supporting Organizations, a President ex officio, and eight chosen by the NomCom. The NomCom is described as having eighteen members, appointed by ICANN's various policy bodies:

  • Five members from the At-Large Advisory Committee;
  • Two members from the Business Users Constituency of the GNSO, one representing small business users and one representing large business users;
  • One member from the gTLD Registry Constituency of the GNSO;
  • One member from the gTLD Registrars Constituency of the GNSO;
  • One member from the CCNSO Council;
  • One member from the ISP Constituency of the GNSO;
  • One member from the ASO Council;
  • One member from the Intellectual Property Constituency of the GNSO;
  • One member selected by the Board to represent academic and other public organizations;
  • One member from the Non-commercial Users Constituency of the GNSO;
  • One member from the IETF;
  • One member from the Technical Advisory Committee; and
  • One member from the Governmental Advisory Committee.
     

  Many of the policy bodies that will appoint the NomCom do not yet exist, and will take shape only over time. In most cases, their future shapes will be derived from existing institutions - the GNSO, CCSO, ASO, GAC, Technical Advisory Committee, Root Server System Advisory Committee all either currently exist at ICANN, or will be derived from other pre-existing bodies - providing an indication of their workability.

  The At-Large Advisory Committee (ALAC), however, lacks a pre-foundation on which to build. The proposed bylaws describe a sort of self-formation for the ALAC, presumably with some involvement by the new Manager of Public Participation (described in Article III). However, no guidance about the ALAC's structure is given, nor benchmarks describing when the ALAC will be considered operational. The ALAC will be appointed by the Board for the foreseeable future, until the Board determines that a better option exists. Uncertainty about when, if ever, the Board will make such a determination has implications for ICANN's overall public legitimacy, as well as the effectiveness of its proposed policy structure.

  The ALAC is the only element of the proposed structure intended to directly address the interests of individual Internet users. As such it has been put forward as an answer to ICANN's need for representation of the public interest. But the ALAC's lack of definition creates uncertainty about whether ICANN will ever adequately support the incorporation of this key voice into its activities.

  Finally, the ALAC would name five of the eighteen members of the NomCom, more than any other constituency of ICANN. But since the ALAC has no present form, the proposed bylaws provide that it itself will be selected by the Board. Because the ALAC appoints a significant part of the NomCom, and the NomCom selects the majority of the Board, this will create a circular mechanism in which the ICANN Board will have excessive influence over the selection of its own successors. Such circularity will not help ICANN achieve its accountability goals.

  Even if questions about the ALAC were to be answered, ICANN's policy bodies under the proposed bylaws would display extensive cross-linking and circularity elsewhere. The proposed bylaws set out that ICANN's Supporting Organizations would appoint six Directors straight to the ICANN Board. At the same time, they would appoint nine members (one-half) of the NomCom. Besides selecting the rest of the Board, the NomCom also names several members of at least one, and probably all, of the Supporting Organization Councils. Even in its best form, the arrangement is excessively complex and will likely lead to less, not more, participation from those with an interest in ICANN.
   

In the final analysis, the proposed bylaws are inadequate to support the kinds of reforms ICANN requires. By reserving too much flexibility to the Board and by not committing to real representation of the public interest, the draft bylaws raise doubts about ICANN ever achieving the kind of credibility it requires. Significant amendments are necessary before they can be approved and implemented.