I was asked to testify before the Subcommittee on Science, Technology, and Space of the United States Senate Committee on Commerce, Science and Transportation.
The oral and written submissions may be found below.
Good afternoon.
Mr. Chairman and distinguished Senators. Thank you for giving me the opportunity to appear.
Today I would like to speak on the topic of ICANN reform.
Everyone agrees that ICANN is in dire need of reform. The only argument is over the shape of that reform.
I have been involved in the Internet since 1974.
I was elected to the Board of Directors of ICANN in year 2000.
I am the only person on ICANN's Board of Directors who was elected by the Internet community of North America.
My seat on ICANN's Board of Directors, and the seat of every other publicly elected Director, will cease to exist on October 31 of this year.
ICANN's plan is repudiation of the concept of public participation and public accountability. ICANN replaces the voting and representation of the at-large membership, which I remind you is the process that elected me, with what can be best described as a garden club. This garden club will have no voice and no authority in ICANN's affairs.
For years ICANN has promised to Congress and to the public that there would be elections, accountability, and transparent decision-making. ICANN has retracted those promises. There is no trace of elections, accountability, or transparency in ICANN's proposals.
ICANN will tell you about its public forums and public processes and its broad consultation with "stakeholders". As a Director I can tell you from experience that these are facades. Most of ICANN's decisions are made by its staff, often without consultation with the Board of Directors.
In light of this experience I encourage to view with deep skepticism what ICANN has said to you about the nature of its processes.
The key to salvaging ICANN lies in process and accountability. ICANN will not be saved, as ICANN proposes, by simply adding more layers of insulation and increasing budgets and staff.
The following factors have contributed to ICANN's present troubles:
The so-called "evolution" proposals created by ICANN are reforms only in the sense that they reshape ICANN at the surface.
The reconstituted ICANN that they envision is one that exacerbates rather than cures ICANN's flaws.
Instead of increasing responsibility and accountability to the public, ICANN's proposals impose more impenetrable walls between ICANN's decision-makers and the public.
Instead of creating a nourishing flow of new ideas into ICANN, ICANN's proposals create councils and nominating committees that will create an even more insular body.
Instead of being a body of limited powers, ICANN's proposals would create a body with ever-increasing powers over the Internet, and an ever-growing bureaucracy to exercise those powers.
It is my suggestion that:
All of this is occurring against a backdrop of concern about what might happen if ICANN were to disappear or become non functional.
ICANN does not have its hands on any of the technical knobs or levers that control the Internet. Those are firmly in the hands of ISPs, Network Solutions/Verisign, and those who operate the root DNS servers.
Were ICANN to vanish the Internet would continue to run. Few would notice the absence.
There's an old joke about corporate life that says when all else fails it's time to reorganize. May I suggest in closing that all that ICANN is proposing is nothing more than a hollow reorganization.
Thank you very much for this opportunity to speak, I will be happy to answer any questions you may have at this time.
Prepared
Statement of
Karl Auerbach
before the
Subcommittee on Science, Technology, and Space
of the
Senate Committee on Commerce, Science and Transportation
June 12, 2002
Good afternoon.
My name is Karl Auerbach.
Today I would like to speak on the topic of ICANN reform.
I have been involved in the Internet since 1974.
I am a computer engineer. I am a principal of a new (and yet unnamed) start-up to create products that will make the Internet more reliable, secure, and efficient.
I am also an attorney. I graduated cum laude in 1978 from Loyola of Los Angeles specializing in commercial, international, and administrative law. Although I maintain my status as the member of the California Bar and the Intellectual Property Section of the California Bar, I am not engaged in active practice. I have been named a Yeun Fellow of Law and Technology at the California Institute of Technology (CalTech) and Loyola Law School of Los Angeles.
I have been a founder, principal, or first employee in several Internet related start-up companies. These have provided me with a broad base of experience in commerce and technology. I have direct experience with the needs and obligations of Internet related businesses.
I have been active in the core design and standardization body of the Internet, the Internet Engineering Task Force (IETF), since the mid 1980's. And I have been a member of the Internet Society (ISOC) since its formation.
I have been deeply involved during the last several years with the evolution and activities of what has become ICANN. I am a founding member of the Boston Working Group, one of the groups that submitted organizational proposals to NTIA in 1998 in response to the so-called "White Paper." I have recently been named to the Board of Directors of the Domain Name Rights Coalition.
I speak for myself; I am here at my own expense and on my own time.
ICANN
I am the only person on ICANN's Board of Directors who was elected by the Internet users of North America.
My seat on ICANN's Board of Directors, and the seat of every other publicly elected Director, will cease to exist on October 31 of this year. On that date real public representation within ICANN will end. After that date, ICANN will be effectively controlled by a small group of privileged "stakeholders".
Many commentators have noted ICANN's weaknesses. Among these are excessive secrecy, lack of public process, lack of accountability, insufficient oversight by the Board of Directors, and poor business practices.
As a Director it is my job to work to correct these weaknesses. ICANN has chosen to either ignore my input or resist my suggestions. For example, ICANN has ignored a paper I wrote shortly after September 11 - Protecting the Internet's Domain Name System - online at http://www.cavebear.com/archive/rw/steps-to-protect-dns.htm.
ICANN has even demanded that I relinquish my rights and obligations as a Director as a pre-condition of inspecting ICANN's financial and other records. I have been forced to assert my rights as a Director through a legal action that is presently before the California courts.
I spoke before this subcommittee sixteen months ago - on February 14, 2001 - my comments of that date are as apt today as they were then. One may review those comments online at http://www.cavebear.com/archive/cavebear/growl/issue_6.htm
Reforming ICANN - The Need To Know History and Establish Principles
There is no doubt that ICANN is in need of reform.
I submitted my own suggestions to ICANN about how ICANN might be reformed. That paper is available online at A Plan To Reform ICANN: A Functional Approach - http://www.cavebear.com/archive/rw/apfi.htm
It would be an error to blindly reform ICANN without knowing where we want to go and why. Without this the result could easily be as flawed as the ICANN of today.
So let me first state in short form what it is that has gone wrong with ICANN.
Then I will then discuss various principles that should guide the reform process and discuss why those principles are necessary.
What Has Gone Wrong With ICANN?
Sixteen months ago I said to this subcommittee:
ICANN is ill designed,
has been ill operated,
has brought upon itself significant ill will within the Internet community,
and has greatly exceeded its proper scope.
I believe that significant restructuring of ICANN is needed so that the corporation can fulfill its purposes and fulfill its obligations towards its stated beneficiaries.
Unfortunately the last sixteen months have brought no improvement.
The following factors have contributed to ICANN's present troubles:
The so-called "evolution" proposals created by ICANN are reforms only in the sense that they reshape ICANN. The reconstituted ICANN that they envision is one that exacerbates rather than cures ICANN's flaws. Instead of increasing responsibility and accountability to the public, ICANN's proposals impose more impenetrable walls between ICANN's decision-makers and the public. Instead of creating a nourishing flow of new ideas into ICANN, ICANN's proposals create councils and nominating committees that will create an even more insular body. Instead of being a body of limited powers, ICANN's proposals would create a body with ever-increasing powers over the Internet, and an ever-growing bureaucracy to exercise those powers.
Defining ICANN's Charter
The most important aspect of any reform is to define the duties and limitations of the reformed body. Vague phrases such as "internet stability" are inadequate; they must be replaced with precise formulations.
ICANN's technical duties should be clearly enumerated and precisely defined. The phrase "technical duties" contains more than a hint of irony - ICANN has done little, if anything, during the nearly three and a half years of its existence that can be construed as "technical".
Why is the charter so important?
Phrases such as "stability of the internet" have been shown to be so elastic as to be virtually meaningless. In addition, ICANN has ignored clear mandates, such as the obligation written into its own By-Laws to operate in an open and transparent manner.
ICANN has grown into areas where its presence is not desirable - such as the regulation of business practices unrelated to the technical operation of the Internet - and away from areas where ICANN's presence is desirable.
Two examples may illustrate:
A clear charter for ICANN is imperative. This charter must not only make positive statements about what ICANN is to do but must also establish inviolable walls that constrain ICANN's authority, role, and scope.
Certain matters of heightened importance should require a supermajority vote of its Board of Directors. ICANN's Board of Directors should not have the unilateral power to resolve the meaning of any ambiguities that might arise.
The charter ought to require that extraordinary questions be placed before an "at-large" body of interested members of the public Internet community.
Prototype charters for ICANN have been suggested by a number of people. Nearly all of these call for a significantly smaller, more limited structure that is more publicly accountable and representative than the charter published by ICANN's staff.
These proposals include:
Abandon The Word "Stakeholder"
The Internet is so pervasive today that everyone has an interest in the Internet.
ICANN has been crippled from its inception by the concept that there are "stakeholders" in the Internet and that only "stakeholders" should be allowed into the forums of Internet governance.
The "stakeholder" concept has limited ICANN's flexibility and forced ICANN into channels that are dictated by the "stakeholder" definitions. By denying people and organizations the ability to form fluid coalitions and relationships according to their self-perceived interests the "stakeholder" concept has made compromise within ICANN exceedingly difficult and rare.
The reform of ICANN must avoid placing people and entities into pre-defined classifications. Instead, the reformed ICANN must allow people and entities to work in concert (or in opposition) as the ebb and flow of their self-interests dictate.
Structural Separation - One Job Per Entity
The third principle of reform is to create structural boundaries that prevent mission creep and constrain ICANN to do only what it is intended to do.
ICANN's multiplicity of roles has created a situation in which it is difficult to ascertain responsibility for decision making. For example, ICANN has accepted a no-cost purchase order from the Department of Commerce to provide something known as the "IANA function". This "function" is not well defined, however, the "IANA" function is most frequently invoked by ICANN with respect to the actual creation of new top-level domains and the maintenance of country-code (ccTLD) top-level domains. Because of the multiplicity of roles, it is often impossible to tell whether a decision is made by ICANN acting as ICANN or ICANN acting as IANA. Accountability to the public disappears when it is impossible to determine who or what is making a decision or taking an action.
My suggestion is that ICANN be reorganized into several separate and independent entities that share neither personnel, nor office space, nor any other resources. This separation of functions is the core idea of my proposal A Plan To Reform ICANN: A Functional Approach - http://www.cavebear.com/archive/rw/apfi.htm
It is not uncommon, or improper, for entities to split themselves into separate parts with minimal residual linkages - Hewlett Packard/Agilent, AT&T/Lucent, and Cabletron/Riverstone/Enterasys are but three examples of companies that have done this.
In my proposal I suggest that ICANN be divided into six entities - three operational entities, each having a clear task and minimal discretion about how to do that task, and three policy bodies, each focused on a single policy area.
The three operational entities would be:
And the three policy entities would be:
This may seem like an excessive number of entities. However, these functions do, in fact, exist within ICANN today but as a hodgepodge. By separating them into distinct bodies they may be more precisely defined and be more visible and thus more amenable to oversight and to be held to account for their actions.
The three operational entities encompass the core aspects of technical reliability of the Internet. These three entities could be quickly and inexpensively established. I believe that they could rapidly obtain wide acceptance and support.
The three policy entities encompass the difficult, value-laden debates over Internet policy that are so necessary but which have so burdened ICANN. The actual day-to-day operations of the Internet would be insulated from these debates by the clearly drawn lines that divide the operational entities from the policy entities.
The United States Department of Commerce Must Actively Oversee ICANN
This statement will cause a great outcry among those who live in other nations. That outcry is legitimate and must be heard.
However, as long as ICANN obtains its authority from agreements with the US Department of Commerce, the DoC ought to ensure that ICANN remains true to its promises.
The National Telecommunications and Information Administration (NTIA) has been extremely lax and has permitted ICANN to evade promises made to NTIA and to the public. A few examples will suffice:
There is a widely held perception that NTIA has been captured by ICANN.
Assistant Secretary Nancy Victory has recently taken the reins of NTIA. There is already evidence of a new and more open attitude. I have every hope that the problems of the past will not recur.
I recognize that the Department of Commerce has to walk a very fine line when dealing with the private corporation that is ICANN. However, the DoC should not use that fine line as an excuse for silence. In fact, if not in law, ICANN is a creation of the Department of Commerce. And ICANN receives its authority from the several legal agreements that exist between the Department of Commerce and ICANN.
Recently there have been several petitions to the Department of Commerce asking the DoC to consider something more than an automatic, unthinking renewal or extension of the various legal agreements between the DoC and ICANN. I wonder about the effectiveness of this approach unless the DoC clearly invites and considers competing proposals from others who might wish to assume one or more of ICANN's tasks.
Full and Meaningful Public Participation
No matter what form ICANN may take, it must include full and meaningful public participation.
The so-called "forums" that populate ICANN and its plans do not constitute "full and meaningful" public participation. In addition, ICANN's board and staff have frequently disregarded proposals and recommendations that have been promulgated by these forums, even when directly compliant with ICANN's bylaws.
As I have suggested previously, ICANN could be split into several pieces. This division would have the ancillary benefit of allowing the form of public participation to correspond to the degree of discretion of the particular ICANN-chunk in question. For those ICANN-chunks that have well defined administrative roles with little discretion, the obligation of public participation could be satisfied by a simple notice-and-comment mechanism. For those ICANN-chunks that are formed around more contentious policy-making tasks, full and meaningful public participation would require more - such as the ability of the public to fill a majority of the seats on a board of directors or board of trustees.
True Accountability To The Public
True accountability has several aspects:
Meaningful public participation and public accountability will bring a much-needed flow of new faces and ideas into ICANN. This will reduce ICANN's sense of "us versus them". And it will almost certainly indirectly result in the eventual hiring of replacement staff and supporting professionals and firms who have more capable financial, business, and legal skills and who comprehend that the role of staff is not to supplant the Board of Directors.
What Would Happen To The Internet If ICANN Were To Vanish?
Much of the debate over ICANN is colored by the fear of what might occur were there to be no ICANN.
ICANN does not have its hands on any of the technical knobs or levers that control the Internet. Those are firmly in the hands of ISPs, Network Solutions/Verisign, and those who operate the root DNS servers.
Were ICANN to vanish the Internet would continue to run. Few would notice the absence.
Were there no ICANN the DNS registration businesses would continue to accept money and register names. With the passage of time the already low standards of this business might erode further.
The UDRP (Uniform Dispute Resolution Policy) system runs largely by itself. The Federal ACPA (Anti Cybersquatting Consumer Protection Act) would remain in place.
ICANN has already established a glacial pace for the introduction of new top-level domains. ICANN's absence will not cause perceptible additional delay in the creation of new top-level domains.
ICANN has already abrogated the making of IP address allocation policy to the regional IP address registries; those registries will continue to do what they have always done with or without ICANN.
ICANN has no agreements with the root server operators; the root servers will continue to be operated as an ad hoc confederation, as has been the case for many years.
The only function that would be immediately affected would be the IANA function. IANA is an important clerical job, particularly with regard to the country-code top-level domains (ccTLDs.) IANA is not a big job, nor does it have real-time impact on the Internet. (In fact there is a credible body of evidence to suggest that ICANN delays certain clerical tasks on behalf of ccTLDs for months on end in an effort to coerce ccTLDs to sign contracts with ICANN.)
There are those who will try to divert outside reforms of ICANN by asserting that touching ICANN will cause the Internet to collapse or otherwise be damaged. The truth is quite the reverse - ICANN's ties to the technical and operational stability of the Internet are tenuous at best. A full inquiry into ICANN, a full reform of ICANN, or a complete rebid of the agreements under which ICANN operates would not damage the Internet.
Supplementary
Statement of
Karl Auerbach
before the
Subcommittee on Science, Technology, and Space
of the
Senate Committee on Commerce, Science and Transportation
June 21, 2002
This document contains supplementary statements to my oral and written statements to the subcommittee on June 12, 2002.
.INT and .ARPA
A question was asked during the hearing about the difficulty of transferring the .int and .arpa top-level-domains (TLDs) away from ICANN.
It would be a trivial exercise to do this. Neither the .int nor the .arpa top-level-domains impose more than the most trivial of burdens upon ICANN.
At the present time, .int is a largely moribund top-level domain; keeping it up to date is essentially a trivial exercise and involves no matters of policy or discretionary decision-making.
The .arpa top-level domain is joined at the hip with the in-addr second level domain; they are essentially a single unit. Administration of .in-addr.arpa has long been delegated to the regional IP address registries (RIRs) such as ARIN, RIPE, and APNIC. The administrative burden of the .arpa piece alone is a trivial exercise and involves no matters of policy or discretionary decision making.
Thus the effort to transfer either .int or .arpa away from ICANN would involve an effort of almost vanishing small size and complexity.
Security of the Domain Name System
ICANN's response to the question of security of the Domain Name System has been inadequate.
DNS security is not merely a question of prevention; the larger portion is the question of recovery after a natural or human event.
ICANN has chartered a committee of experts in the field of computer and network security to look into the issue. This committee's skills are largely academic and focus on the technical aspects of network protocols and operating system security. The committee lacks expertise in matters of physical protection, personnel security, and other non-technical aspects of security. Similarly the committee lacks expertise that pertains to recovery and re-establishment of service after a security event. ICANN's committee has no experts in restorative measures or data repository techniques. The committee does not have expertise in matters such as finance (i.e. where will the cash come from when one needs to buy new computers and obtain new office space after a security event or natural disaster?) Nor does the committee have expertise concerning the gathering and protection of evidence (i.e. how will one investigate and prosecute the wrongdoers?)
At the same time, ICANN has quietly failed to recognize concrete and easy measures to protect DNS or enhance its recoverability that could have been taken immediately and at nearly no cost.
For example, ICANN has ignored a document presented to ICANN shortly after September 11 by this Director: Protecting the Internet's Domain Name System - http://www.cavebear.com/archive/rw/steps-to-protect-dns.htm.