CDT: ICANN Study

As the Internet becomes an essential part of life for millions of people worldwide, once obscure questions about who manages its underlying technical systems are taking on new public importance.

The Internet community is on the eve of an unprecedented election designed to allow millions of computer users around the world to vote for board members of the Internet Corporation for Assigned Names and Numbers (ICANN), an organization that will manage certain crucial Internet technical systems.

But nearly every member of the Internet community with whom we spoke, as well as respected outside observers, identified fundamental problems with the current plan for the election being put in place by ICANN. Some of these problems can be addressed through improvements in the proposed election process, but others are the products of conflicting goals for the election itself. These problems are compounded by the fact that most of the electorate envisioned by ICANN does not know what ICANN is or what it does. In short, what we found is a proposed election process for ICANN viewed with almost uniform skepticism by informed observers.

ICANN faces the daunting goal of seeking a fair ballot, free from capture or fraud, from a potential electorate of millions of Internet users worldwide who have little knowledge of ICANN and little understanding of its mission, in order to select a high-quality board of technically-capable members -- all by September of this year. Realistically, without substantial changes to the proposed process, it is difficult to see how this is possible.

Until these fundamental concerns are resolved, this election can only be viewed as a risky experiment in democracy that must be dramatically improved for it to confer legitimacy on ICANN.

This report explores concerns with ICANN's current At-Large election model and a range of options to address them. Authored by two U.S. public interest groups with expertise in election systems and Internet policy, with support from the Markle Foundation, this report is the product of three months of extensive analysis and consultation with a broad cross-section of stakeholders and election experts. This report does not represent all of the disparate concerns about these elections, nor does it provide a comprehensive "silver-bullet" solution for ICANN to follow in response to those concerns. Rather, it is designed to be a resource to assist the Internet community in having a meaningful debate about the proposed plan for the election before proceeding with it.

ICANN and its Elections: ICANN's 19-member Board of Directors (including its appointed president) was designed to give those affected by ICANN's decisions some voice in its governance. Nine interim Board members were seated by appointment in October 1998. Last fall, an additional nine directors were elected by the Supporting Organizations representing defined Internet addressing, protocol, and domain name stakeholders. The original nine interim directors are to be replaced by nine new members who are to be elected "At-Large" from among the broader Internet community of public stakeholders, in an election to be conducted by September 2000. While a matter of great debate within ICANN, exactly how that election is supposed to take place is only now being established.

Major features of the current election plan include:

Electorate - Any person with an email address and physical address who registers can vote.

Quorum - A minimum of 5,000 registered members is needed.

Internet voting - Members will vote electronically using PIN numbers received in the mail.

Indirect elections - Members will vote for an 18-person "At-Large Council", which in turn elects the nine At-Large Board members.

Two-stage election - One-third of the seats will be filled first, with a second election for the remaining seats to follow an evaluation of the first election.

Timing - The election of all new At-Large board members is to be completed by September 30, 2000, with the first stage of elections starting this spring and ending by July 2000.

ICANN's Mission: In theory and by design, ICANN's mission is narrow and technical. But even technical decisions about who gets which names and numbers can have broader policy implications. And because of ICANN's central control over Internet functions, it will face pressures to make broad policy decisions.

ICANN is likely unknown to its very broad potential electorate of Internet users worldwide. Its mission and the limitations on its authority have not been stated plainly, and ICANN's activities in areas like intellectual property protection may create further confusion. Without a mission that is clearly delimited, there is a risk that many voters will not understand what they are voting for or why.

Election Goals: There is little agreement among those consulted on the ultimate role of the At-Large election in advancing ICANN's mission. Its goals appear to be several:

To give a voice in ICANN's governance to those bound by and/or affected by ICANN's decisions and who are not currently represented in the ICANN structure - a diverse population potentially including tens of millions of people all around the world.

To select high-quality board members capable of fulfilling ICANN's responsibility for management and stability of essential technical systems.

To fairly represent the diverse interests of those affected by ICANN decisions, as expressed by an engaged and educated electorate.

To avoid "capture" of the board through disproportionate representation of any one organization or interest group.

To complete the election and seat the new board by September 30, 2000.

These goals are in tension with each other. For example, including millions of potential voters who today have little knowledge of ICANN and are unlikely to vote, creates the potential for capture by a motivated interest group. On the other hand, completing the election by September 2000 will make it difficult to have an educated electorate that will participate broadly in the election and thereby diminish the risk of capture.

Concerns with the Current Election: Almost every person we consulted expressed skepticism about the current election plan. The concerns raised most frequently include:

Public Engagement -- Few in the potential electorate understand ICANN or its mission.

Capture -- There is a 5000-person quorum, yet millions are eligible to vote. This system is highly vulnerable to "capture" by motivated groups at the expense of those without the resources to organize voters.

Scope of the Electorate -- Not everyone with an email address is a sufficiently interested and knowledgeable party to vote.

Indirect Election -- An indirect election disengages voters because they do not even get to vote for board members.

Accountability -- At-large council members have no accountability for what Board members they support, and there are no current prospects for "slate-making" that would help voters better understand the candidates for whom they are voting.

Nominations -- There is no clear nomination process and no plan for campaigns that educate the electorate.

Diversity -- Minority viewpoints could be submerged by a two-stage, winner-take-all election process where only a few people are elected worldwide at a time.

Fair elections -- There are few safeguards to ensure fairness, guard against fraud and corruption, or provide for monitoring of the election by a third party.

If ICANN proceeds according to its current election plan, it is unlikely that these concerns will be addressed.

Options and recommendations: The full report explores a wide range of options to address the concerns set forth above. These include options relating to the election structure, the membership, the process for putting candidates on the ballot, the campaign rules and regulations, the voting system to be used and other miscellaneous matters.

No single set of options will perfectly answer all of the concerns and at the same time achieve all of the goals for the election. But based on our outreach and analysis, Common Cause and CDT believe the following ideas hold out the most promise for improving the fairness and legitimacy of ICANN's elections:

1. ICANN should develop and promulgate, in some binding fashion, language to delimit the scope of ICANN's decision-making authority to technical management issues. ICANN must do a better job of explaining to the public what it can and cannot do, and make those limitations a clear and binding part of its structure.

2. ICANN should establish an independent election authority to set the rules for the election, and then audit and monitor the conduct of the election.

3. The "electorate" for the At-large board members should broadly include individual Internet users. Legitimacy can only come if ICANN operates with the consent of the governed - those bound by and affected by its policies. Finding the appropriately inclusive membership that can be practically engaged in ICANN remains a major challenge. At this time, we believe the planned membership open to all persons with an email address is a best, though still a highly problematic, first approximation that should at least be attempted and then carefully evaluated.

4. The electorate should directly elect the At-Large board directors. Having an "At-large Council" choose the board members diminishes accountability and both the reality and appearance of an open, inclusive, representative and democratic election. On the other hand, if a system of indirect elections is maintained, some accountability mechanisms - such as a commitment to a slate - should be developed for the At-Large Council members.

5. ICANN should develop a clear candidate nominating process and campaign rules. A clear process should be established for putting candidates on a ballot - including a hybrid of a nominating committee and open petitions, coupled with baseline campaign rules.

6. ICANN should use some form of a proportional-representation voting for its At-Large elections.

7. ICANN should institutionalize periodic review of the At-Large membership and elections process. A clear sunset for the current rules should be established, along with a review process, to evaluate this year's election and suggest improvements for moving ahead. This evaluation should be done by a group independent of the existing board.

8. ICANN should place the goal of holding effective elections ahead of the goal of completing them by September 2000. We recognize that ICANN faces intense pressure to hold its elections this year. Regardless of how the election proceeds, however, it must ultimately be viewed as a first experiment in democracy for this new technical management body.

Given the outstanding concerns about matters ranging from membership to capture, the unanswered questions about how the electorate will develop over time, and the rapidly changing nature of the Internet itself, ICANN cannot claim that its proposed election system, as it now stands, will confer legitimacy on it. Much work is needed to address and resolve the myriad of serious issues that have been raised. ICANN's approach to these elections should be an iterative process of democracy: one that looks at this year's election as a tentative first step, with rules for subsequent elections to be established in the future upon further community deliberation.

This study is a joint effort of the Common Cause and the Center for Democracy and Technology.

Common Cause (http://www.commoncause.org) is a leading advocate for openness and accountability in government. A non-profit public interest group based in Washington, D.C., Common Cause brings its knowledge of fair and democratic electoral processes to this study, as well as a fresh perspective on issues of Internet governance.
The Center for Democracy and Technology (CDT) (http://www.cdt.org), also based in Washington, is a non-profit public interest group dedicated to promoting individual liberty and democratic values online. CDT has extensive experience with Internet policy and technology issues relating to free expression, individual privacy, and open access to the Internet.

The full report is available online at http://www.commoncause.org/icann/ and http://www.cdt.org/dns.
For more information, please contact:
Andy Draheim of Common Cause <adraheim@commoncause.org> or
Lusan Chua of CDT <lchua@cdt.org>.