Reconsideration Request 00-13
Received: 15 December, 2000

Dear Reconsideration Committee:

     Our firm represents Image Online Design with respect to its ICANN TLD application. Below please find Image Online Design's (hereinafter "IOD") request for reconsideration submitted in accord with the ICANN reconsideration policy. Additionally I am attaching this request as a Word Perfect file. Please feel free to contact me via e-mail at djs@cph.com should you need any additional information.

Name, Address, and Contact Information for the Requesting Party, Including Postal and Email Addresses.

Image Online Design, Inc.
1241 Johnson Avenue, Suite 201
San Luis Obispo, California 93401 USA
+1 805 543-4716

The Specific Action of ICANN for Which Review or Reconsideration Is Sought

     Review and reconsideration is sought for two related actions of ICANN. The first action under review is the ICANN Staff and TLD Evaluation Team's report, and related comments, made to the ICANN Board during the Annual Meeting, 15 and 16 November, 2000. See generally, ICANN-LA 2000 Public Meeting Archive, available at http://cyber..law.harvard.edu/icann/la2000/archive/.

The second action under review is ICANN Board Resolution 00.89, which reads:
"RESOLVED [00.89], the Board selects the following proposals for negotiations toward appropriate agreements between ICANN and the registry operator or sponsoring organization, or both: JVTeam (.biz), Afilias (.info), Global Name Registry (.name), RegistryPro (.pro), Museum Domain Management Association (.museum), Société Internationale de Télécommunications Aéronautiques (.aero), Cooperative League of the USA dba National Cooperative Business Association (.coop);"

ICANN | Preliminary Report | 16 November 2000, available at http://www.icann.org/minutes/prelim-report-16nov00.htm#SecondAnnualMeeting.

The Date of the Action(s)

     Both actions occurred during the ICANN Annual Meeting, on 15 and 16 November, 2000. ICANN Board Resolution 00.89 was passed on 16 November, 2000.

The Manner by Which the Requesting Party Will Be Affected by the Action

     Image Online Design will be substantially harmed by ICANN's action. Specifically, Image Online Design will be excluded from competition for gTLD registry services in the U.S. Government root server system. This harm is irreparable because a first-to-market effect will benefit those competitors selected and dilute the opportunity to compete for subsequent market participants.

     Additionally, Image Online Design remitted a $50,000 application fee to ICANN for the purposes of having its application fairly and equitably reviewed. Because ICANN's review of Image Online Design's application was performed contrary to ICANN's bylaws, it has suffered harm to the extent of its application fee.

Whether a Temporary Stay of the Action Is Requested

     A temporary stay of the action is not requested.

What Specific Steps the Requesting Party Asks ICANN to Take - i.e., Whether and How the Action Should Be Reversed, Cancelled, or Modified

     Image Online Design requests ICANN take the following steps: First, that Staff and Evaluation Team reassess and revise their evaluation of Image Online Design's TLD application for .Web consistent with this request, and expediently communicate its revised evaluation to the ICANN Board for their reconsideration.

     Second, Image Online Design requests that the ICANN Board reassess its decision to exclude Image Online Design from Board Resolution 00.89 and direct the Staff to expeditiously proceed with negotiations with IOD.

The Grounds on Which the Action Should Be Reversed, Cancelled, or Modified

     ICANN failed to treat IOD's TLD application equitably in violation of its by-laws and breached its obligation to fairly and competently evaluate each TLD application. ICANN is required to treat all applicants equally. In pertinent part, ICANN's amended by-laws read as follows:

"Art. IV§(c) The Corporation shall not apply its standards, policies, procedures or practices inequitably or single out any particular party for disparate treatment unless justified by substantial and reasonable cause, such as the promotion of effective competition."

ICANN, Amended Bylaws, July 16, 2000, http://www.icann.org/general/bylaws.htm#IV.

     ICANN also breached its duty to fairly and competently evaluate IOD's TLD Application. ICANN and its staff is obligated to fairly and competently evaluate each Applicant's application. In pertinent part, ICANN's stated criteria reads as follows:

"The ICANN staff is responsible for gathering information about submitted applications, evaluating the applications and associated information, and making recommendations to the Board based on the applications, associated information, and evaluations."

ICANN, Criteria for Assessing TLD Proposals (15 August 2000), http://www.icann.org/tlds/tld-criteria-15aug00.htm.


     A careful analysis of the evaluation team's report and comments on both IOD's application and Afilias' application reveals that IOD was held to a much higher standard than other applicants. One example of this higher standard, which is discussed in greater detail below, is found in the staff's "Fault Tolerance" section, where the substantially similar technical configurations of IOD and Afilias are treated in completely different manners. Afilias' fault tolerance plan receives a passing acceptance in the staff report, while the same fault tolerance plan, which has been actually implemented by Image Online Design, receives a long and critical evaluation, concluding that the plan is both unrealistic and insufficient.

     Throughout the examples below, comparisons are made between IOD's application and the application of Afilias, a similarly situated applicant. Of note is that IOD's application is compared to the application submitted by Afilias because they were the approved applicant for an open, commercial top-level domain. Accordingly, we look to their application as an example of the threshold by which ICANN determined an acceptable application.

      a. The Fault Tolerance of IOD's Registry Was Incorrectly Cited As Inadequate

     The fault tolerance of IOD's registry system is essentially identical to the fault tolerance of Afilias' registry system. Despite this fact, IOD's system was criticized while Afilias' system was praised. This demonstrates the inequitable treatment by the staff of IOD's application.

     With respect to the design of the registry system, and specifically its fault tolerance, the staff report has the following to say about Afilias:

"Fault Tolerance. Equipment is co-located in Exodus managed sites. Redundancy is based on clusters of servers fed by load balancers using standard designs. Registry data bases are replicated on a dynamic basis allowing immediate switch over to an operating data base. Registry data base will be mirrored at a remote location. The remote location will operate as a cold back up to the primary site. DNS servers are located world wide." Staff Report, at B.2.d.

On the other hand, the staff report takes issue with Image Online Design's:

"Fault Tolerance. DNS services will be hosted by UltraDNS, which provides strong fault tolerance. The registry service will be hosted by GST Telecommunications, which (unlike UltraDNS) does not itself offer fault tolerant capabilities beyond multiple network links to the facility. It is not clear that IOD's design for the registry has sufficient fault tolerance. IOD plans to bring a second mirror facility on line on the East Coast, but it is unclear when this would happen. Procedures for ensuring proper distributed operations between these facilities are little discussed in the proposal. To provide the high level of availability expected from a large registry applicant should consider including warm or hot fail-over capability in the initial deployment. The proposal forecasts an impressive 99.99% uptime for the registry service." Id.

     Again, Afilias was praised for its similar system while IOD was criticized. It is clear that the design proposed by Afilias is functionally identical to the design proposed by Image Online Design, which also based its redundancy on clusters of servers fed by load balancers using standard designs.

     Another example of the unfair and inequitable analysis by the ICANN staff is that Image Online Design does not give a timetable for the creation of mirror data centers. Moreover, neither Afilias nor IOD provided a timetable for deploying their mirror data centers. However, only IOD was criticized on this point. See generally, staff report summary, at B.2.d.

      b. IOD's staffing plans were criticized when the substantially identical plans of other Applicants were praised.

     The staff summary criticized IOD for its need to hire employees if it were awarded the .WEB registry and its plan to rely on executive search firms for its technical and executive requirements. However, Afilias received no criticism when it identified the same staffing needs. Specifically, Section D13.1.7 of Afilias' application stated:

"As of the date of the application, Afilias will consist of its Board of Managers, with no additional executive or administrative staff. Afilias will benefit from the collective experience of its participating Member Registrars to transfer the expertise required to build its staff within six months of approval of the application by ICANN, enabling Afilias to set a target date to begin offering its services in Q3 2000. High-growth Member Registrars, including Tucows, Register.com, Inc. and Network Solutions, Inc. have proven capabilities for scaling their staff to meet the needs at peak level demands, and Afilias will benefit from the transfer of knowledge from its Member Registrars to successfully build its staff as demand levels require."

Individuals of IOD's management team was criticized for holding jobs unrelated to the .WEB registry pending approval of its application by ICANN. Such criticism was not levied at the management teams of other Applicants.

For example, Section C.2 of the Staff Summary of Image Online Design's application states, in part,

"... two of the four employees, the Chief Executive Officer and the Chief Operating Officer, are performing two roles. The planned management is working at Toyota San Luis Obispo as the Chief Executive Officer and Business Manager. This dual responsibility could potentially become problematic for the registry operation" (emphasis added).

     In sharp contrast, no criticism was made of the staffing portion of Afilias' application despite the fact that each member of the Afilias' executive board also held a full time job unrelated to the registry to be awarded. Indeed, Afilias' executive board was praised for their "dual roles."

     Further, the staff report incorrectly assumes that the "dual role" of IOD's management team would continue if IOD had been awarded the ".WEB" registry. Indeed, had the staff inquired, it would have realized that both identified individuals intend to assume full time management positions with IOD upon the award of the ".WEB" registry, pending the hiring of additional members of the management team. In this respect, Afilias and IOD have essentially identical plans to hire additional professional management team members.

     Again, in this context, IOD was criticized and Afilias was praised.


     The staff's failure to fully comprehend and analyze IOD's application resulted in the presentation of inaccurate information to the Board. The three primary areas of the application adversely affected were technical, financial and personnel. Each is discussed in further detail below.

     The notation of inaccuracies contained in the Staff Report and Board presentation discussed herein is not intended to be a complete list of errors. While there may be additional errors not cited in this request, the errors brought to the attention of this panel are more than sufficient to support the grant of our request for reconsideration.

      1. Technical Inaccuracies
            a. IOD's Operational Registry Should Have Been Treated as an Advantage By The Staff and Not as a Detriment.

     The staff report suggests that IOD's registry, with 20,000 registered names, represents a very small system, and called into question IOD's ability to run a large global registry. The truth of the matter is that our registry represents over 4 years of research and development. From start to finish, the progress to date represents at least 75% of the development and 95% of the research required to create a large operational registry.

     Furthermore, IOD has a functioning registry with scalability already in place. The fact that IOD has an operational registry with years of experience was inappropriately ignored, or treated as a detriment.

            b. Despite Superior Peak Transaction Capabilities Eliminating the Need for Initial "Demand Throttling" Schemes, IOD's Application was Assessed As Inadequate

     A materially inaccurate assessment of IOD's peak transaction capabilities was presented to the Board. Image Online Design's approach to the problem of a "land rush" or initial surge in registrations (and the necessary need for capacity) was to design and build our system to handle the traffic. A system properly designed to accommodate the initial rush of traffic is also suited for day-to-day operations. Other applicants, like Afilias, chose to build a less capable registry system and to instead implement a demand throttling system where by registrars queue domain name registration requests and submit them to the registry in a round robin fashion. This method is not necessary if the registry system is properly robust. The "demand throttle" system introduces an artificial scarcity to the initial operation of the registry, and demands, by definition, that registrars take "pre-registrations" in preparation for submission in the round-robin process, leading to abuses such as exorbitant pricing for preferential queue placement.

            c. IOD's TPS Was Incorrectly Reported as Deficient, When In Fact it Far Exceeds The Maximum Anticipated Load.

     A materially inaccurate assessment of IOD's transactions per second (TPS) rating was presented to the Board. The technical evaluation team incorrectly reported to the Board that the transactions per second (TPS) rating of IOD's system as 28.6, an unrealistically low rating. This error seems to be a mis-read of section D15.2.10 (Peak Capacities), paragraph 4, which actually describes the number of sequential registrations each machine in IOD's cluster is capable of handling. The paragraph describes that each machine can handle 250 simultaneous and consecutive requests, and that the load simulator used 28.6 requests per second as a real-world load multiplier. It is of concern that the technical evaluation team confused the rate of the load simulator with the TPS rating, and then reported its inaccurate understanding to the Board in a manner that adversely affected IOD.

            d. IOD's Registry System was Incorrectly Characterized as Having Poor Fault Tolerant Properties

     The staff improperly analyzed IOD's application with respect to the highly fault tolerant operation of its registry. Specifically the staff failed to properly attribute the numerous fault tolerant properties detailed in the registry operators section of IOD application. Several of these properties are detailed below:

     As noted in section D15.3 of IOD's application, "The GST network is fully redundant and all facilities are built to the highest telecommunications standards. Every hub site has its own indefinite power generation capability and diverse and redundant fiber routes (emphasis added)."

     As noted in section D15.2.13, "Hot spares (extra hardware that is formatted and initialized for use) are kept in the data center for the occurrence of hardware failure. In the case of catastrophic failure of the data center itself, the duplicate data center will take over operations while repairs or relocation occurs."

     Section D15.2.12 states, "In the event of a power failure, the transfer switch in the facility will automatically detect the failure and switch the power source from commercial to generator power." This section clearly shows that power redundancy, as well as all other physical redundancy requirements are already complete.

     Section D15.2.7 of IOD's application clearly states that "a minimum of 2 redundant data centers ensures security of all data. The registry database data is replicated, in real-time, between the two data centers by way of a dedicated, secure connection. The backup procedures described in D15.2.13 will be undertaken in both data centers. Backup systems will be maintained on-site as well as off-site storage of critical data in secure locations. In the case of a catastrophic failure of any data center, alternative sites will have current data and will be able to recover functionality immediately."

     As clearly outlined above, no clear or accurate reading by the ICANN staff could have produced a conclusion that IOD had insufficient fault tolerance properties.

      2. Financial Inaccuracies
            a. The Financial Team Inaccurately Reported IOD's Credit and Capital Funding Were Deficient When in Fact Both Are More Than Adequate At All Times Of Operation

     A materially inaccurate assessment of IOD's funding was presented to the Board. The ICANN reviewer states that IOD only holds $450,000 in cash as of the beginning of the projected period and states that this is a significantly weaker position than the other applicants. This is inconsistent with Section D13.2.13 of IOD's application that clearly states IOD has solid credit commitments of up to $2,000,000 prior to being selected by ICANN, and over $6,000,000 thereafter.. Specifically, IOD made the following statement in Section D13.2.13:

"Prior to commencement of our operation as a primary TLD, we believe that cash flow from current operations would be significant enough to fund operational and primary TLD startup costs. Further, we have two separate $1,000,000 credit commitments from financial institutions which can be found at Appendices D13.4.4_A and D13.4.4_B. After .Web enters the root, we believe that our operations will provide significant cash flow to cover operational needs and that we would retire any outside credit commitments in the first quarter of fiscal year one."

     Because IOD does not project that any additional funds will be needed in the operations of the company, the amount of available credit was not included on the projected financial statements. Accordingly, IOD's capital position is not weaker than the other applicants as was misrepresented to the Board.

"Holding only $450,000 is a significantly weaker capital position than the capital positions of the other applicants." Finally, based upon its historical experience, Image Online Design has not demonstrated the ability to grow, even when performing other services such as web hosting and design. Overall, the other applications in this group are significantly more realistic and would result in much more viable competition for the .com registry. staff report, at C.2.

     This erroneous statement omits the letters of credit, and incorrectly compares us to other applicants, thereby removing any objectivity. This is a major error because all subsequent financial projections were based on $450,000 total capital.

            b. The Staff's Incorrectly Reported That Allowing Registrars Access to a New Registry Would Create Serious Stability Problems." "Moreover, an Attempt to So Significantly Revise a Registry So Soon after Launch Would Be a Serious Stability Problem. Staff report, at fn. 3"

     The Staff's Incorrectly Reported That Allowing Registrars Access To A New Registry Would Create Serious Stability Problems. However, Network Solutions was able to develop its Registry-Registrar test bed without threatening the stability of their registry. IOD now benefits from NSI publishing of the RRP standard, and is confident that we can implement this published standard without inducing any instability into our registry.

     To the extent the adoption of the use of independent Registrars prior to the end of the first year of operation might be perceived as materially reducing IOD's revenue, IOD believes this to be incorrect. The use of independent Registrars would most likely result in a greater number of registrations because of the access to additional markets and consumers, not readily available to IOD, but clearly available to the independent Registrars. Accordingly, it is anticipated that a test bed during the first year would result in a greater number of registrations with limited overall effect and an increased total income stream, even if the revenue per registration declined slightly. This increase in income is anticipated regardless of whether IOD makes a move towards using independent Registrars before the end of its first year of operation or after.

     Further, if during the transition from IOD's role as Registry/Registrar to using independent Registrars, a temporary access to a credit line becomes necessary, IOD has already established that it has access to up to $2,000,000 for just this type of contingency.

            c. The Staff Incorrectly Evaluated IOD's Projected Market Share

     The business/financial evaluation team concluded that there were significant deficiencies in the business plan submitted by Image Online Design, particularly compared to other applications in this group. The staff notes:

"First, Image Online Design expects to obtain a 15 to 23 percent market share of all new registrations in the very first quarter of operation, even with additional competition from other new top-level domains." Staff Report, at III B.1.a.

     However, we strongly believe that this is a very realistic projection. When a new TLD is introduced, there will be quite a rush to register names in it. Accordingly, 15 to 23 percent is conservative considering the first-quarter demand. Further considering that ICANN approved only 2 generic and open domains (.info and .biz), and that one of them (.biz) is limited in that it requires approval and is more expensive at in the startup phase, 15 to 23 percent is not unrealistic.. Indeed, we proposed 1.6 million domains registered in the first year, whereas Afilias proposed even more; "Afilias is expecting total registrations of 2.6 and 16.2 million registrations, respectively." Staff Report, at C.1.

Turning next to IOD's application, the staff concluded that:

"This combination creates a very large influx of money to finance operations, with Image Online Design's cash balance increasing from $450,000 to $37.4 million in three months at the 50 percent confidence level, which is 83 times larger. The need for this influx presumably is the motivation for Image Online Design's insistence on being the sole registrar during startup." Staff Report, at C.2

Even by the end of the fourth year, after other registrants have been permitted to compete for three years, Image Online Design estimates that it will still obtain a 30 percent registrar market share within the TLD, and that it will do so with a $20.00 registrar markup. This is inconsistent with experience in .com, .net and .org." Staff Report, at C.2.

     While the Staff reports the correct percentage from our financial projections, they contest the accuracy of our assertion. However, the data used to calculate the projections was based on marketing statistics, projections by our staff, and expert opinion. Further, recent reports tend to corroborate our assertion. Digital MASS reports that NSI currently enjoys a 40% market share, even more than IOD projects, despite health competition from registrars. Digital MASS , available at http://digitalmass.boston.com/columns/internet/1018.html.

            d. The Staff Failed to Consider That The Market Determines the Price of Domain Names At Both the Registry And Registrar Level.

     The Staff Failed to Consider That The Market Determines the Price of Domain Names At Both the Registry And Registrar Level.

"Despite this new competition, moreover, Image Online Design anticipates maintaining its $15 registry price throughout the forecast period. This is at least two and a half times the registry prices anticipated by others in this category. This higher price is likely to deter registrars and potential registrants." Staff Report, at C.2.

     Consider the following examples of this free market: The cost of a .MU domain is $50 a year, as is .CC and .TO. The cost of a .NU domain is $30 a year. Specifically, .WS charges $35 per year for domains of 4 characters or more. 3-character domains are $1000 per year, and 2-character domains are $2500 per year. While .TV domains are $35 for non-generic names, generic names are considerably more expensive. For example, sex.tv, was registered for a reported price of $1,000,000 per year. Truly, the market decides the price of domain names.

      3. Business Location Inaccuracies
            a. San Luis Obispo was Incorrectly Characterized as an Area with Limited Resources from Which to Draw

     Further inaccurate conclusions regarding IOD's application includes the erroneous assumption that IOD will experience difficulty in hiring a sufficient number of competent support staff from the San Luis Obispo area. For example, the ICANN Staff Summary of Image Online Design's application for .Web (http://www.icann.org/tlds/report/web1.html, Section B.4) states, in part,

"To operate the type of registry and registrar operation envisioned, the applicant would need to do extensive open hiring of management, technical and support team members. This hiring would need to be accomplished in an area with limited available technical personnel" (emphasis added).

     The statement that the "hiring would need to be accomplished in an area with limited available technical personnel" is both incorrect and misleading. Rather, the California Central Coast has a large pool of technical talent, as well as being geographically close to both the Los Angeles and San Francisco/Silicon Valley areas. Hiring from these markets is expected. Indeed, IOD currently retains a number of contractors and consultants from these markets. The statement is further misleading in that it fails to take into account that the San Luis Obispo area is a very attractive region, and with appropriate salary and relocation compensation, hiring from out of the area is expected desirable. Also, IOD is not required to maintain 100% staffing levels in the San Luis Obispo area. The location of the East Coast data center has not been finalized, and it is expected that the available hiring capabilities of the area will be a primary concern in selecting that location.


     During the deliberations, the Board members repeatedly poised important questions regarding IOD and its pending application. Many of these pertinent questions did not receive responsive answers. Other answers, while responsive in nature, were factually incorrect.

     A fundamental lack of understanding as to IOD's application by the staff and evaluation team, as detailed in Section B above, led to an inappropriate and premature tainting of IOD's application. During the Board's deliberations, the evaluation team imbrued its incomplete and inaccurate understanding of IOD's application onto the Board. This is evidenced by numerous statements by Board members favorable to IOD, only to have those statements contradicted by the staff, based on an inaccurate understanding of the content of IOD's application. In addition, despite this information being readily available to the staff, specific requests for information by Board members went unfulfilled, despite this information being available to the staff.

     Because the board was unable to receive complete and accurate information about IOD's application, the board's action to exclude IOD application was unfair. Several examples are detailed below.

At 36:35 (all times cited refer to Public meeting archive, available at http://cyber.law.harvard.edu/icann/la2000/archive/) Dr. Cerf began the consideration of Image Online Design by noting that IOD had been operating for five years. Dr. Cerf stated that he was unsure whether the IOD's registry was compatible with the rest of the Internet. Mr. Louis Touton never addressed this question, but rather replied that IOD had operated outside of the standard root servers, had some 18,000 registrations, which represented somewhat of a surge, and quoted expected market share levels. Mr. Touton then stated that he did not know if IOD supported the RRP, and then, without further comment, turned the discussion to "Affinity Internet's" application. However, as shown in Section D15.2.1 of IOD's application, IOD's registry presently operates with existing Internet DNS Servers. Further, as discussed below, IOD had specific stated plans to implement registrar protocols. Specifically, the comments made were:

Cerf: "I'm also fairly impressed by the IOD comment that it's been operating a registry for a period of time for 5 years ... what I was not sure of ... was whether that operation has been known to be compatible with the rest of the Internet system and the registrar protocols ... if it would actually function with the rest of the network? So I guess I'd be inclined to throw that one on the list of potential candidates."

Touton: "IOD is operating outside of the standard root pointed to by a number of alternate roots over the last few years. It is currently ... had 18,000 registrations, which we understand from some data that we've looked at represents a bit of a surge. Perhaps prompted by a possibility of this very process. The system as proposed and presently in operation is a proprietary single-registrar system. As proposed they projected that they'd continue to be the sole registrar for one year and then it would drop to 80% and 70% and 60% share as they implemented a shared registration system and Charlie will have to remind me if it has the same protocol as the NSI system or not, but it was a thin-registry model, and during the first year they propose to charge $35 for every registration. After the introduction of additional registrars, they would charge a registry fee of $15. They projected that by the end of the 4th year they might be able to achieve 3.4 million (I may be wrong on this point 4) registrations."

"So that's a summary of the aspects of that. But before we go to that, let me turn to JD who has the Affinity information."

At 1:11:00, Dr. Cerf again brought up IOD, inviting a discussion by commenting on the statement by staff that IOD would not be accepting multiple registrars. Mr. Touton answered non-responsively, and then turned the discussion back to ".biz." Specifically, the exchange was as follows:

Cerf: "For example, your discussion of the IOD application for .Web in which there was no possibility for multiple registrars."

Touton: "We put 2 on the table, we got them halfway discussed, and both are in a state of incompleteness. Let's try to finish up the discussion of Affinity, and then there's more to say about IOD as well."

At 1:19:40, Mr. Touton states that, after the applications were submitted, IOD said that they would be accepting registrars (note that this was well after Dr. Cerf's comments that IOD would not be accepting registrars, which went uncorrected by Mr. Touton at the time). Mr. Touton incorrectly stated that IOD did not have such RRP software now, and that the evaluation team thought it was a larger task than the 30 to 60 days that IOD proposed. This was an incorrect statement. In its application, Section D15.2.2, IOD clearly stated that the software was already in development. Mr. Touton then incorrectly compared the IOD software development to the development cycle at NSI, stating that it would take longer than anticipated. Again, this is certainly incorrect, as NSI had to develop its RRP "from scratch," whereas IOD must merely implement the published RRP standard. Id., see also, IOD Application, D15.2.2_A.

Mr. Touton then incorrectly asserted that IOD's $450,000 cash-in-bank figure was IOD's sole capital contribution, noting that IOD had $2 million in credit, but that it wasn't clear that IOD would be using it, or what it was for. In truth, IOD has $2 million in credit at the present time, which increases to $6 million upon acceptance (with one letter of credit becoming "open" and subject to a request for any amount) by ICANN. Specifically, the comments made were:

Touton: "After submission of the application and in response to a combination of public comments and perhaps a staff question as well IOD indicated that they would be willing to accelerate the time in which they'd make their RRP available for a test bed. They don't have such software now, but Chris Ambler, the CTO of IOD indicated not yesterday but at a forum earlier in the week that he'd talked to some programmers that thought it could be done in 30 to 60 days. The view of the evaluation team is that first of all it's a process question of that kind of amendment but leaving that aside, based on the experience of Network Solutions in implementing their SRS, it's a larger-scale task than that, and also that if the introduction of the shared registry system were to occur some time after the expanded operation of the registry that would be a very destabalizing effect if that were done shortly afterward. It takes a lot of phasing, and we had a lot of trouble in the case of NSI where it took .... well we started working on it late 98 and introduced it in June 99 and it still had a lot of problems after that. One other aspect presented in the report were the resource levels available. Their company has an equity available now of $450k and I believe they have indicated that there are credit lines of 2 million dollars though they haven't committed to use them for this purpose. The principles of the company are involved in an auto dealership and it wasn't clear what the credit lines were for. Okay, I think I've now fairly [inaudible] the IOD Design application."

At 1:36:15, Ken Fockler took up the issue of IOD's willingness to accelerate the adoption of registrars, and suggested that ICANN could negotiate such a business model with IOD, as Mr. Touton had suggested with respect to Afilias. Mr. Fockler concluded by saying that he supported pioneers, and that "IOD is a candidate for my basket." Dr. Cerf agreed. Mr. Touton, however, immediately changed the subject to whether or not Afilias was in the basket, and led the board to approve it as such. At no time did the board consider IOD as being "in the basket," and at no time did Mr. Touton present the issue for a decision. Specifically, the comments made were:

Fockler: "I think you said earlier ... and referring to statements they made earlier in the week, I heard it a little bit stronger than just referring to a few programmers, that they were already discussing that arrangement with people about their shared registry, and that they were already consulting with people to speed it up.

     We could negotiate some things if we want to, business models will change... I have a strong empathy and support for the pioneers and those that go ahead and get things done... I have a strong tendency to see, with the IOD experience, that they have shown they can run a registry. They're a candidate for my basket... [applause]"

Cerf: "Well, we're allowed to put more than one thing in the basket, so could we try to get one more in?"

Touton: "Let's go back to the Afilias, and then we'll treat that as a request to put IOD in the basket. How are the feelings to put Afilias in the basket?"

Touton: "IOD. Ken put IOD on the table. Just to ... and the report, and we went into a great deal of detail on this, although the business and technical team initially both thought that it wasn't one of the stronger applications in this group, because there were some public comments favoring it, we looked at it in more detail to see how it compared with the others. In terms of the .web ... three .web proposals ... there was [long pause] ... in some of the investments that we talked about, the expected market we talked about, the price we talked about, in addition to those, well... let me talk a little bit more about the financials. As described in the evaluation report, because IOD is a relatively small company with limited resources, at least in their application their general approach was to have a period of exclusivity especially during the land rush period, where at $35 each they could fund the growth of their business. In terms of technical parameters as quoted, the performances are about the same but Afilias and NeuStar both offer a thicker higher-service registry vs. the thin registry that IOD plans, ultimately, either 30/60 days or later to implement. ... that's kind of an overview of the three .webs, I know that's contrary to our procedure, but..."

Cerf: "I'm having a terrible oscillation internally because the IOD guys have been at this for a long time, and I have some sympathy for pioneers, having a few hours on my own, at the same time I'm very concerned about the viability of the proposition. It sounds like this only works for them if they're a combined registrar/registry... and most of the intent, I think, in asking for proposal, was to get registries on the table. The fact that they might be able to get a protocol developed and tested is - I'm sure they can eventually, I'm always concerned about software because it always takes a bit longer than anyone wishes or thinks. So I'm still hesitating on technical grounds that getting multiple registrars sounds like a technical hurdle, and that causes me to hesitate to put them in the basket."

Roberts: "The economic model he proposed is exactly what the NSF did with NSI 6 years ago, and that got us into a lot of trouble, and the absence of competition in the proposal is troublesome."

Touton: "This may be such a fundamental change that it changes a lot in the business model, and I think that's right. It hasn't been worked through be a coherent proposal, based on this change."

Dyson: "IOD doesn't seem to cut it for these 2 reasons. One is the business model that I don't see how it works. You're now looking at a much larger world where there will be competition, and I don't think that initial surge is going to translate at $35 a head when you have other options out there..."

Kraaijenbrink: "Why I'm not putting IOD in the basket is the fact that they contravene everything we've worked for in the past few years. Separation of registry and registrar and bringing down price. So I was a little bit amazed to see a proposal like it."

Touton: "Any other comments on IOD? [silence] Alright. Would someone like to go to the next category then?"

At 1:42:00, without solicitation from the Board, the staff identified a list of problems, the majority of which IOD in this Request For Reconsideration has shown to be false and/or incorrect assertions. Staff member Mr. Touton acknowledged the inappropriateness of this action by stating, "I know that's contrary to our procedure." This prompted Dr. Cerf to conclude, "this sounds as if this only works for them if they combined registry/registrar," which is untrue, but was based on the stated opinion of Mr. Touton. Mike Roberts then commented that IOD's "economic model ... is exactly what the NSF did with NSI," and expressed his distaste. Mr. Touton then concludes by saying that IOD "hasn't worked through to be a consistent proposal based on this change," referring to the commitment to accelerate the acceptance of registrars.

     Taken as a whole it is apparent that the ICANN Board was not presented with required information to render an accurate decision with respect to IOD's application. Numerous questions posed by Board members to the staff received either non-responsive or wholly inaccurate answers. For example, Dr. Cerf likely would have placed IOD in his basket but was "[hesitant] on technical grounds." Dr. Cerf's hesitations were founded in factual inaccuracies presented by the Board. Similarly, Esther Dyson notes "IOD doesn't seem to cut it for these two reasons," both were inaccuracies presented by the Board. Accordingly, the Board's decision to not select IOD's application was based on inadequate information as required to reach an informed decision.


     ICANN should grant IOD's request for reconsideration and approve IOD for further negotiations with the ICANN Board.

     IOD has demonstrated that the ICANN Board's denial of IOD's application for the ".WEB" registry was based on an erroneous understanding of the content of IOD's application. This fundamental misunderstanding by the ICANN Board was a direct result of the staff's failure to fully analyze and comprehend IOD's application, the staff's failure to answer material questions presented by the ICANN Board, and the staff's inaccurate and non responsive answers to specific questions poised by the ICANN Board during the November 2000 meeting.

     If the ICANN Board had been presented with an accurate and fair assessment of IOD's application, the ICANN Board would have concluded that IOD's application was equal to, or superior to, the applications actually approved. But for this misunderstanding, IOD would not have been excluded from Board Resolution 00.89 and would have been invited to proceed with negotiations with the ICANN Board.

Any Documents the Requesting Party Wishes to Submit in Support of its Request. The following documents are hereby fully incorporated by reference:

Image Online Design's TLD application for .web, available at: http://www.icann.org/tlds/web1/.

Afilias' application TLD application for .info, .site, and ..web, available at: http://www.icann.org/tlds/info1/.

ICANN-LA 2000 Public Meeting Archive, available at http://cyber.law.harvard.edu/icann/la2000/archive/.