This declaration emanates from a recognition that while ICANN is a well-analyzed institution, few credible proposals for reform exist. Accordingly, this proposal first outlines certain premises that we consider to be necessary and important. The principles incorporated in this document were arrived at from various discussions with interested parties in Accra at the recent ICANN meeting, statements and discussions held online following the publication of the Lynn Proposal in addition to the private thinking of the authors of this document. It is upon these premises and principles that we proceed to elaborate on a proposed Board structure for ICANN that we think are consistent with those principles. This elaboration also takes into account the proposals made and the concerns expressed by Stuart Lynn, President of ICANN, about perceived deficiencies in ICANN’s processes and structure.

ICANN has been selected by the United States’ government to run the DNS from among a number of proposals submitted “precisely because it was open, consensus-based, and rooted in the Internet community”[1]. The goal of reforming ICANN is to make it work better. Consensus-based decision-making may not always be appropriate where commercial interests conflict. This does not mean that in future ICANN should be closed, hierarchical, and divorced from the industries, communities, and consumers that it serves.

Our goal is to foster a productive discussion among the parties that make up ICANN, or who have an interest in the good management of the DNS. We all have a serious interest in keeping ICANN, or something very much like it, able to perform those necessary functions of domain name management for which it was created. Let us cooperate in the task of making it work well.

The purpose of this document is to provide a departure point for positive and directive discussion of ICANN reform. The authors do not see this document as an endpoint but as a starting point and are hopeful that interested and informed parties will expand on the ideas expressed herein, especially within areas of their specific subject matter. We expect interested parties to dissect this document and pursue specific discussions on the various points and principles made. An analysis of this nature will allow for the creation of a structure that supports and informed statement of the specific ideals that will allow ICANN to work well.

This discussion will be fostered through the website at http://www.byte.org/heathrow and the associated mailing list found at the website. This discussion is not intended to be exhaustive, nor should it be (although it may be) the only thread of its kind. The more voices that weigh in, the more refined the output will be and the more valid the conclusions will become. However, this discussion must necessarily be rational, informed and must be managed towards the goals stated in this document.

Premises

Our premises are that

1. The governance of the DNS is a matter of the common convenience in being able to locate Internet-based resources, which governance should be suitable for the mundane purposes and nature of the problems that the DNS solves;

2. The Internet, and the Domain Name System in particular, work because of a high degree of voluntary cooperation among carriers, root servers, registrars, registries, and standards-setting organizations;

3. The Internet, and the DNS in particular, cannot be made to work without that high degree of cooperation among private interests and actors;

4. Private actors and interests are capable of self-government: of being represented and having their conflicts resolved within a structure of largely voluntary, self-interested compliance, assisted by an authority capable of enforcing necessary common rules with contractual or other means of enforcement;

5. Governments, domain name registrants and users, and users of the DNS itself have legitimate and distinct interests in the well-being of these necessary, if temporary, arrangements;

6. Governments have particular interests in the management of country codes, which interest is separate and distinct from that of ICANN with respect to country codes[i] [further commentary follows at the end of this document].

From these premises several positions naturally follow.

The legitimacy of ICANN

ICANN’s basic idea is legitimate: a private-sector organization, founded in corporate law rather than international treaty, is an appropriate political and legal arrangement to engage in the management of the DNS. Private interests in the DNS are best expressed and dealt with directly, rather than through the indirect apparatus of governments.

The appropriate test of success or failure

The criterion of success of the DNS is the common convenience of mankind in being able to locate resources on the Internet efficiently. Failure of the DNS to serve that function might be repaired or compensated for by the use of other resource locators. Fixing the governance of the DNS will not in and of itself fix any failures of the DNS, however strong and appropriate governance[ii] will provide a suitable vehicle for addressing these failures when required. Ambiguities in the governing documents of ICANN, or failures to achieve contracts with all players involved in the DNS within a certain period of time, are not the same as an actual failure of the Domain Name System to perform its functions, and should not be treated as such. One should not confuse failure to achieve agreements with an actual failure of that necessary cooperation on which the DNS relies. Indeed, proposals should not seek to compel a cooperation that already exists. Obedience can be compelled, but not cooperation.

Participation in self-government

We are ICANN.

ICANN should not be a structure that stands outside and above the governed; it should embody the principles of self-government. This does not imply that every part of the structure is elected or democratic; there is a role for the ascription of expertise and of prestigious individuals to the tasks at hand. Yet the function of ICANN is to maintain the public interest in a convenient, effective and affordable domain name system, for as long as we need one, and not longer. It is not the basis of democratic world government, or even the governance of the Internet. Its role is humble; its goal is to be useful. When the time comes for it to die, it should be able to disappear.

Principles that will guide ICANN reform

1. The governance of the DNS should be appropriate and proportionate to the nature and needs of the DNS. Accordingly, the governance of the DNS should not outlast the useful life of the DNS. This result is more likely to be achieved if governance of the DNS is more responsive to popular demand for domain names and a coherent working DNS than to formal arrangements among states.

2. Owing to the role of states in the management of country codes, the role of a central manager of the DNS, such as ICANN, is naturally larger in relation to generic TLDs than it is in relation to country codes.

3. Those who wish to participate in the management of the DNS should contribute to the funding of it, possibly with some exception for non-profit entities.

4. Businesses need a structure that resolves issues quickly. The future structure of ICANN or its successor should be tested against this criterion. This means that rules must specify voters, election procedures, and what constitutes a majority suitable to decide an issue. Consensus is not a sufficiently precise basis for action. The decisions of stakeholders should be capable of being enforced or supported by ICANN’s Board.

5. Registrants of domain names have a greater claim on the attention of ICANN than end users who do not own domain names. The interests of domain name registrants and users are broad and varied. ICANN must provide a mechanism by which the interests of Registrants are actively taken into account.

6. Internet Users who are not registrants[iii], being without a contractual link to ICANN, have interests that are distinct from those of governments. Internet users are affected by whether the symbols they type resolve to websites they seek. Some avenue of participation in DNS management, apart from registrars themselves, is desirable, and it is appropriate that they pay for this input in some measure.

Organizational Proposal

We acknowledge that these principles do not automatically lead to each and every one of the organizational proposals set forth below. People will have honest differences of opinion regarding the appropriate role of the electoral principle, to any limitations on the democratic principle, the weight to be given to certain constituencies, and the time it should take to transit from a nominated to a partly or fully elected structure.

The chief questions concern:

- How is policy to be decided?

- What are the appropriate checks and balances?

Board Construction

Our assumptions are that:

- There will be an ICANN, or a successor organization for the same purposes, and that it will be governed by a Board of Directors comprised of fifteen individuals;

- Five of the Board seats will be or represent, respectively, the CEO of ICANN, the Government Forum, the Technical Forum, the gTLD Forum, and the ccTLD Forum;

- There will be some form of membership which will be represent the interest of domain name registrants, who will also be entitled to seats on the Board of ICANN;

- These representatives will in time be directly elected in ways that we have not yet devised;

- There will be a transitional period during which they will be selected by a nomination committee with different criteria;

- A nomination committee will select another five members of the Board of ICANN, and this arrangement will be perpetual.

Making policy

We foresee that policy will be made both in the policy forums and consented to by the board, and made on occasion by the board of ICANN itself. We are quite sure that there will be conflicts of opinion within forums, among the forums, and between the forums and the Board. This is normal. That consensus would be achievable is the one area where the original assumptions of ICANN, based as they were in the culture of the academics who created the Internet, has proven to be in error.

Contrary to the current idea that the Board should only endorse a previously made consensus, the Board of ICANN should be able to initiate decisions as well as to receive policy proposals from below for advice and consent.

The inevitability of conflicts and the need for balances

Preparing for conflict should be the task of those who want a new ICANN constitution. The resolution of conflict should not be assumed away because new structures are envisaged, or new players invited to participate. This is why we emphasize the importance of stated jurisdictions for forums, explicit criteria for the selection of people, and explicit voting rules for the resolution of issues.

None of this is beyond the competence of good lawyers and constitution-makers to deal with.

The constitution of ICANN should specify the types of decision which call for majorities, and which call for super-majorities, such as a two-thirds vote.

We propose that when policy is devised by a forum that is vested with the subject matter, a simple majority of the board should suffice to set policy. This implies that the jurisdiction of forums shall be specified in ICANN’s revised constitution.

We propose that when policy is devised by the Board, and imposed on a forum or forums against its or their wills, that a two-thirds majority of the Board be required.

The Board, without serious consideration and a super majority, should not overturn a forum vested with the subject matter in its decisions.

The need for criteria

We foresee that there will need to be different criteria for the different decisions and situations that the new structure will generate.

Choosing outsiders for the nomination committee

The Lynn proposals foresee that there will be a nomination committee to select ten Board members, composed of one representative each of:

- ISPs

- The GAC

- The gTLD forum

- Business

- The technical forum, and

- The CEO of ICANN

- Plus an unspecified number of outsiders.

The constitution should specify the criteria for the selection of the nominating committee.

The Domain Name Users/Registrant Forum

Which interests will be entitled to sit on this forum?

Business

Which business interests require appropriate representation in the users’ forum?

Domain Name Users/Registrants Board Representatives

What criteria should be specified to govern the selection of the five users’ representatives in the interim process of selection of these people for the Board?

NCI Board Representatives

What criteria should be specified to govern the selection of the five “best interest of ICANN” representatives in the process of selection of these people for the Board?

Subject matters for policy forums

The exclusive and the shared jurisdictions of the policy forums will need to be devised.

Commentary

[1] Stuart Lynn, “A unique, authoritative root for the DNS”, 9 July 2001, http://www.icann.org/icp/icp-3.htm, at page 13

[i] Scope of National Authority vis-à-vis Country Code Top-Level Domains - There need be a clearly delineated division of powers in the governance of ccTLDs between ICANN and nation states. This sum of these powers should be loosely equivalent to those exercised by ICANN with respect to gTLDs.

The specific powers of ICANN with respect to ccTLDs should be narrow in scope and express. Areas not specifically attributable to ICANN should be assumed to be the responsibility of nation states.

Different countries will approach Internet governance differently. Some will exercise control through the organs of government. Others will deal with these issues through broad-based groups on interested parties. This document expresses no opinion on these varied approaches, but instead will try and propose a framework that allows for the broad range of approaches.

Areas of ICANN Responsibility

- Delegation of nameservers;

- Maintenance of appropriate delegation in root servers;

- Minimum level of operational standards;

- OTHERS?

Areas of National Authority

- Nexus requirements;

- Transfer policy;

- Registrar accreditation;

- Competition issues;

- OTHERS?

We reiterate that areas of authority not specifically enumerated should be deemed to be areas of national authority. In addition, as with all elements of this document, these delineations of authority should be open to change over time as appropriate.

With respect to funding we propose a tiered structure based upon number of registrations with the lowest tier incurring no fee. Specific proposals must be tabled based on an analysis of specific registration numbers in the various ccTLDs.

[ii] Clarification Regarding “Strong and Appropriate Governance” – The phrase “strong & appropriate governance” is intended to describe a structure that possesses the resources, conviction and capability to act in the best interests of the DNS and larger Internet without requiring the structure to inappropriately inure to potentially stronger, self-interested actors in order to gain political or economic leverage. This leverage must be naturally inherent in the structure and can only be obtained through serious and concerted effort by all involved.

[iii] Clarification Regarding “Internet Users”/”Non-Contracted Interests” - For the purposes of this document, Internet users that are not registrants have also been referred to as “Non-contracted Internets”. These entities, business’ or individuals, are valued stakeholders that require a voice in the ICANN process. However, these interests are distinct and separate from most others represented within the process due to the lack of legal relationship with the other actors. It is not the intent of this proposal to diminish or redefine this group, but rather to hold them as separate so as to ensure that the interests of those that do have a direct or indirect legal relationship with ICANN are also appropriately dealt with.