A Plan for Action Regarding New gTLDs

18 October 2002
Posted: 8 November 2002

A Plan for Action Regarding New gTLDs
(by Stuart Lynn, ICANN President)

At its August 23rd meeting, the ICANN Board of Directors directed me to produce a "plan for action" with respect to the implementation of the Report of the New TLD Evaluation Process Planning Task Force. This document and its attachments are in response to that request.

There are three parts to this response. Each Part contains its own independent set of recommendations, and each Part can be addressed and pursued separately:

I. A Proposal for the Board to Extend the Proof of Concept in Parallel with Evaluation of the new gTLDs: This proposal provides the Board with the option, should it so wish to do so, of extending the original "Proof of Concept" to solicit up to three more sponsored TLDs, subject to certain conditions, following a suitable, but short, bidding process. Applicants for sponsored TLDs from the previous round would be invited to update their proposals and resubmit them, but new proposals would also be invited.

II. A Plan for Implementing the Key Recommendations of the NTEPPTF Report regarding the Evaluation of New gTLDs: This plan for the Board's consideration proposed how to implement the principal recommendations of the NTEPPTF Report for a Monitoring Program and an Evaluation, while recognizing the limitations of time and funding.

III. A Recommendation that the Board seek DNSO (or its successor) advice on how to evolve the top level generic namespace: This recommends that the Board ask the DNSO (or its successor depending on what evolves in Shanghai) to develop a recommendation on how to evolve the generic namespace, with specific focus on the question as to whether and how to rationalize the space (that is, to develop a taxonomy for top level domains); whether to expand the space strictly according to market demand; or whether to pursue some third alternative; and to develop other principles that should govern the evolution of the namespace. This is an important step towards fulfilling the requirements of Paragraph IIC 8(b) of the recently amended Memorandum of Understanding with the United States Department of Commerce regarding the creation and implementation of selection criteria for new and existing TLD registries, including public explanation of the process, selection criteria, and the rationale for selection decisions.

The first topic provides the option of parallel processing to allow some movement forward while longer term directions are under consideration. The second topic concentrates on evaluation of what has already been accomplished, to provide guidance for the future – at least to the extent that resources allow. The third topic addresses some key policy concerns that will surely be factors in any future expansion of the top level generic namespace.

I plan to discuss these approaches in the Public Forum at the Shanghai meeting. However, except possibly for Part III (which is in effect seeking a recommendation from the DNSO), I am recommending that no specific actions be taken until the Annual General Meeting in December to allow time for posting and community feedback. In the meantime, some of the steps recommended in the "parallel processing" document (Part I) can be pursued as a prerequisite to a go ahead (or not) decision in December.

Stuart Lynn
President, ICANN
October 18, 2002

Click here to submit a comment on this action plan.
Comments will be most useful if they are made by 10 December 2002.

Table of Contents

A Plan for Action Regarding New gTLDs

Table of Contents

Part I: A Limited Proposal for Moving Forward with Another Round of New Sponsored TLDs





New Policy?

Appendix A: Background

Appendix B: Documents from the Tralliance Corporation/ IATA

Part II. Implementing the NTEPPTF Report Recommendations for Evaluation of New gTLDs

Initial Evaluation Project

Monitoring Program

Evaluation Methodology and Schedule



Part III: Evolving the Top Level gTLD Namespace

Part I: A Limited Proposal for Moving Forward with Another Round of New Sponsored TLDs


This document is a personal assessment by the ICANN President in partial response to a request from the ICANN Board of Directors to develop a recommendation as to whether and how ICANN might move forward with a limited new round of TLDs in parallel with conducting an evaluation of those new TLDs introduced over the past twelve months.

It is the conclusion of this assessment that the ICANN Board of Directors could consider authorizing the President to initiate immediately a new round of proposals for up to three sponsored TLDs, subject to validating two conditions (see Recommendations). This would be launched as an extension of the original Proof of Concept following similar, if somewhat streamlined, criteria and groundrules. It is also recommended that special financial consideration be given in this new round of proposals to those proposals that were submitted but not accepted in the previous round, recognizing that such proposals may need to be updated.

In Part III of this Action Plan for new gTLDs, it is separately recommended that the ICANN Board of Directors request the DNSO or its successor to develop a recommendation on how to proceed with the gTLD namespace, in particular considering whether to rationalize the namespace according to some taxonomy as a basis for expansion, whether to continue to respond piecemeal to justified market demand, or whether according to some other alternative. Proceeding with the limited extension to the Proof of Concept, however, should not wait upon the outcome of this namespace study.


With ICANN's reform process heading toward a conclusion, the organization will soon have the energy and resources to return its attention back to the key issue of adding new top-level domains to the DNS. Since the initial group of seven new TLDs was selected in November 2000, the ensuing two years have witnessed the expected sequence of events, albeit more slowly than anticipated: the negotiation and signing of registry agreements, the preparation of technical and administrative operations, the launch phase, intensive marketing, the introduction of dispute resolution mechanisms, the propagation of zone files, and so forth. These seven new gTLDs were authorized as a "proof of concept" to gain understanding of the practical and policy issues involved in such an undertaking.

At the same time, ICANN convened a New TLD Evaluation Planning Process Task Force (NTEPPTF), which issued its Final Report on 15 July 2002. The NTEPPTF Final Report recommended that ICANN promptly undertake a comprehensive, multi-stage evaluation process in which the most critical questions would be addressed first by an evaluation team under contract to ICANN, with evaluation reports to be completed by the middle of 2003, and that a monitoring program be launched immediately to focus on the effect of new TLDs on root server system performance, DNS operational performance problems, accuracy and completeness of Whois data, charter and contract compliance by registries, and startup and landrush problems. The NTEPPTF also recommended that the ICANN Board "should consider to what extent it can initiate planning of and proposal solicitations for new rounds of gTLDs in parallel with the Evaluation and Monitoring projects recommended above."

In light of these recommendations and the long time horizons they necessarily entail, ICANN Board members asked me to prepare a report on the possibility of moving forward with some sort of limited new TLD program, even while the full-scale evaluation and monitoring are being conducted according to the recommendations of the NTEPPTF. I emphasize that this is a personal assessment and does not necessarily reflect the views of NTEPPTF or of any of its members.

More background is given in Appendix A: Background. In Appendix B are attached documents that have been forwarded by Ron Andruff, President of Tralliance Corporation; and from Patrick Murphy, Senior Director for Business Development, IATA, regarding their new arrangement for the purposes of seeking a .travel gTLD.


The underlying premise of this report is that most of the vexing problems over the course of the new gTLD process have arisen in connection with the unsponsored TLDs, not with the sponsored TLDs. This is not surprising, and does not reflect negatively on the selected operators of those new unsponsored TLDs – it rather reflects the simple reality that they undertook vastly more difficult and complex tasks than are required to launch the smaller, specialized sTLDs operated in the interest of a particular, defined community.

Experiences with the new sponsored TLDs since November 2000, although somewhat limited in that these launches are not yet fully mature, seem to indicate that sponsored TLDs can be added smoothly and with little fanfare The hardest parts lie in determining by objective criteria whether a proposed TLD sponsor in fact stands as an accountable, legitimate policy entity with demonstrable support within the community it proposes to serve; and in establishing and implementing a charter that balances the need to ensure that the sponsorship is limited to a well-defined community against being so overly prescriptive as to encompass too small a segment of the relevant community. This is not to minimize the difficulties in launching even new sponsored TLDs: addressing the organizational, policy, administrative, legal, and technical issues took considerable time and effort on the part of those responsible. But many if not most of these problems were internally focused, and did not affect external interactions by others with the sponsored TLDs.

In sum, the sponsored TLDs appear so far to have generated relatively few problems, raise the least concern, and can fill easily demonstrable community needs. If these premises are correct, ICANN could, should it so wish, move forward with the selection and other preparations for introduction of additional sponsored TLDs at a measured but expedited pace without awaiting completion of its full evaluation of the first group of new gTLDs.

Some in the IETF have expressed concern about the unknown effect on the performance of root name servers of wantonly adding many new gTLDs with rapidly increasing zones. On the other hand, the tools for assessing how many would constitute "wanton" (as opposed to "measured and sustainable") do not appear to exist. However, I have not encountered any real technical objection to the addition of a relatively modest number of new gTLDs, particularly if the "shape" of those gTLDs is not too "flat". It is hard to find anyone who would argue that tens or perhaps even hundreds of new small- to medium-sized gTLDs could not be safely added (as opposed to thousands or tens of thousands – numbers which have raised concerns among some in the technical community), particularly if there were careful monitoring of overall DNS performance as the new gTLDs were introduced. After all, new ccTLDs have been successfully added to the root steadily over the years. Furthermore, neither the PSO nor the IETF commented negatively on the original recommendation of the Names Council to add a limited number of new TLDs that led to the "Proof of Concept".

Sponsored TLDs on the whole also raise fewer business concerns than unsponsored TLDs. Since registrants must come from a carefully defined and limited community, there are fewer worries about trademark infringement and cybersquatting. This is an assertion that needs to be tested against the reality of experience with the existing sponsored TLDs, at least one of which has yet to start accepting registrations via ICANN-accredited registrars. Likewise, sponsored TLDs entail fewer and less complex start-up and launch phase worries. Within the defined communities to be served by unsponsored TLDs, there is likely to be less contention among multiple applicants for the same domain name. The sponsoring organizations are well-positioned to sort out any of these issues that might arise, acting in the best interests of their particular communities. Sponsored TLDS are, by definition, limited with regard to the number of registrations, so they are unlikely to grow to the size of the largest unsponsored TLDs, although clearly even sponsored TLDs with very generic names could grow quite large quite rapidly.

Some would argue that continuing to carve up the TLD space, even with sponsored TLDs, in a somewhat ad hoc manner is bad policy, that there should be some established model for how new TLDs should be introduced, a taxonomy as it were. That is certainly a point to be considered by the Board in setting the selection criteria for any future round of sTLDs. Others believe this question should be left to the marketplace to decide and there should be no constraints on proposals other than their technical adequacy. Both points of view are valid. In any new round of new TLDs, the Board would have to decide whether to simply reiterate the selection criteria from the new TLD process used in 2000, or whether changes (such as a rationalization of the name space and associated uTLD strings) are desirable. In a companion document, I have addressed this issue in greater detail.

There are many other significant, unanswered questions regarding new unsponsored TLDs – questions whose answers should, in my view, await completion of the full-scale evaluation recommended by the NTEPPTF (see Part II of this overall Action Plan for new gTLDs). In any event, there appears to be little demand right now for new unsponsored TLDs  – at least, no one is banging at my door. Some of the enthusiasm of a couple of years ago has perhaps been tempered by the realities of the marketplace.

By contrast, the most significant unanswered questions regarding sponsored TLDs are relatively few:

1. Are the organizations coordinating the new sponsored registries able to manage the required technical and business responsibilities of a sponsored TLD? So far, we have seen no evidence of any incipient failures. The new sponsored TLDs are all managed by experienced organizations, and the sizes of the registries are relatively small – no larger, for example, than a small- to medium-sized ccTLD. In any event, the effect of failure or interruption of service – while highly undesirable and not to be taken lightly – is less dramatic than could occur with a large unsponsored TLD.

2. Have registrations been accepted by these new sponsored TLD registries that do not meet the defined criteria? (Put another way, is there any evidence that sponsored TLDs are havens for cybersquatting or other registration abuses likely to cause concerns among significant portions of the Internet community?) Here again, we have seen no evidence of significant problems, but this can be verified with a limited but rapid test based on sampling. Some might argue that it is too early to tell, but the time between now and the launch of a new round of request for proposals should provide an adequate sample. The sponsored TLDs have all implemented careful post-registration screening prior to entry of a given registration in the zone file, thereby guarding against abusive or noncompliant registrations.

The concern that these sponsored TLDs might go the way of .org (which was originally formed with the intention of limiting registrations, a goal that has long been abandoned) does not appear to be justified. The new sponsored registries are contractually bound to follow their charters. Besides, it is in the interests of the communities they serve to maintain the unique definition of their sTLD.

It might also be asked whether the addition of a limited number of small-to-medium-sized new sponsored TLDs would have a significant potential effect on the performance of the root? Conventional wisdom seems to indicate not, at least I have heard no one express such a concern. As indicated earlier in this document, the Names Council had recommended that a limited number of new TLDs be added to the root, and there was no negative reaction to this recommendation from the PSO or the IETF. It may also be practical to add new sponsored TLDs at a sufficiently gradual pace so the effect can be reliably monitored to ensure that we do not harm DNS performance as a result. I assume there would be no practical concern in adding three new TLDs to the seven already selected as part of the Proof of Concept.

These questions (coupled perhaps with the general question of how to expand the namespace mentioned elsewhere in this document and treated at greater length in Part III of this Action Plan) seem to represent the paramount outstanding issues that should be considered before moving ahead with even a limited round of sponsored TLDs. Given their nature, I believe that the two questions posed above could be addressed and answered to a reasonable degree of satisfaction within a very short time period. This assumes that other major questions would be addressed as part of the new gTLD evaluation as recommended by NTEPPTF, but that these major issues mostly affect unsponsored TLDs, and not sponsored TLDs. Or to put it another way, if the above questions are satisfactorily addressed, there would be little risk in moving ahead with a small number of sponsored TLDs now following the same groundrules as the last round.

In Section 4 of its Final Report, the NTEPPTF recognized and detailed both the opportunities and the risks of moving forward with "parallel processing," noting that:

The Board, therefore, may be tempted to consider the extent to which it can move forward on certain fronts – such as planning for the next round of gTLD applications – in parallel with the evaluation process. The Board may (or may not) find that the problems and risks associated with moving forward with new sponsored gTLDs are less than those associated with new unsponsored gTLDs, and therefore may wish to move ahead faster with introducing requests for proposals for new sponsored gTLDs.

Certainly much of the planning for new gTLDs can be done in parallel with the evaluation, as can much of the proposal solicitation and selection  – provided that proposers understand the risks of submitting proposals with no guarantee that any will be selected. The key checkpoint is the actual entry of any newly selected gTLD into the root zone file, assuming proper procedures and understandings are implemented. Much, however, can be accomplished in parallel on the way to that point of no return.

This assessment goes further, suggesting that in fact there is little risk – subject to addressing the two questions above – in moving forward with a limited number of sponsored TLDs even to the point of full operation, extending as it were the "proof of concept" to include a few more sTLDs.

Stakeholder commentary on the NTEPPTF approach has been consistent with these general assessments of opportunity and risk, though there is disagreement about the relative weight of the opportunities and the risks. For example, the Copyright Coalition on Domain Names stated that:

We agree that many of the risks of most concern to copyright owners may be less with sponsored TLDs, in which the sponsoring organization plays a gatekeeper function, than in unsponsored TLDs. Even here, though, it makes sense to have at least some idea of how the existing sponsored TLDs are functioning before kicking off a new round.

For a full record of comments and submissions, see the NTEPPTF Final Report.

In the context of parallel processing, I should reemphasize the broader question mentioned earlier that ICANN must begin to grapple with, though its resolution is not essential to this proposal, which calls for a limited extension of ICANN's ongoing "proof of concept" round of new TLDs. That is the question of namespace taxonomy: whether new TLDs should be added in piecemeal response to justified demand, that is, according to demonstrated market demand for new TLD strings as defined by would-be sponsoring organizations; or should the name space be rationalized or "taxonomized", as it were, with ICANN defining the TLD strings in some comprehensive fashion, and requesting bids for the various individual components of this "taxonomized" space? There are valid arguments on both sides of this question. As outlined in Part III of this Action Plan, I believe that this question should be posed directly to the DNSO for its advice and recommendations. It should be noted that developing an acceptable taxonomy could take some considerable time, and that any bid process is likely to be quite lengthy.

In short, it seems reasonable that (1) some degree of parallel processing is feasible and desirable, and (2) that sponsored TLDs raise far fewer concerns and carry far fewer risks than unsponsored TLDs. There are two questions to be answered relating to unsponsored TLDs that, if resolved quickly, could decisively tilt the balance of reasonable opinion in favor of a new round of unsponsored TLDs, even as the full-scale evaluation is proceeding.


I therefore recommend that the Board instruct the staff to solicit proposals for up to three more sponsored TLDs as an extension of the "Proof of Concept" and following similar, if somewhat streamlined, criteria and ground rules, subject to funding a rapid study, based on sampling techniques as appropriate, to assess to what extent, if any, the new sponsored TLDs have admitted registrants that lie outside of their charters.

This study could be completed very rapidly. In normal times, these questions could be answered within a month or so. However, at this time, staff time is very limited by reform planning and potential implementation, so there could be some slippage. ICANN will also possibly be in the midst of reform transition. The Board will therefore need to assess the priority of moving forward in this way against other competing claims for action. The reality is that it could take through the first quarter of 2003 to complete the study. ICANN would undoubtedly obtain good cooperation from the existing sTLD registry operators. With similar cooperation from the DNSO constituencies, an objective answer could be obtained and documented within a one-month period.

It should be taken as a given that the addition of three more sponsored gTLDs will not cause technical problems of performance. There was nothing magical about the original number of seven as part of the "Proof of Concept". The number ten would just have fit within the original advice given by the Names Council without any technical objection from the PSO or the IETF.

Eventually, the question of namespace taxonomy will need to be addressed as a prerequisite to substantive expansion of the top level namespace, regardless of any interim action on additional sponsored TLDs. That issue, in fact, must be addressed in fulfillment of the tasks posed by the amended Memorandum of Understanding with the U.S. Department of Commerce. The resolution of this question, however, should not hold up the limited extension of the Proof of Concept recommended in this document.

Finally, let me say a few words about how a new round of sponsored TLD proposals would be evaluated. In my conception, all of the applicants that submitted unsuccessful proposals for new sponsored TLDs in 2000 would be invited to update and resubmit their proposals; in addition, new sponsored TLD proposals would be accepted from new proposers. Cost allocation would have to be assessed, but there should be a differential fee recognizing, in all due fairness, that the original proposals from the 2000 round of applicants from 2000 (who have already paid $50,000 to ICANN) have already been subjected to evaluation.

To accelerate the process and to reduce costs, it would also be my recommendation that the independent technical and financial evaluation process be streamlined, particularly as to review of financial capabilities. As to policies and contracts, for the same reasons and to avoid re-negotiation of agreements that were already heavily-negotiated, my strong recommendation to the Board would be to require use of the existing contractual framework for the new sponsored TLDs.

New Policy?

It could reasonably be asked whether proceeding in this manner is de facto creating new policy, whether it is in fact reifying the sponsored/unsponsored and restricted/unrestricted framework that governed the "proof of concept" of the first seven new gTLDs, and the legal documents that enshrined their creation. After all, one of the purposes of the evaluation is precisely to determine whether this is a useful framework or whether perhaps it needs modification or more.

As already noted, I would suggest that if the Board wishes to pursue this proposal, it regard it as an extension of the "proof of concept" and defer consideration of the larger policy question until after the evaluation is complete. There was nothing magic in the number seven for new gTLDs. It could just as well have been three, fourteen, or twenty.

Expanding the number of sponsored TLDs within the proof of concept would also yield more data for the evaluation. For example, proposers would know in advance much more about what they could expect – such as the agreements they would be expected to embrace – than was known by the original proposers; it would be of interest to see how many would submit proposals forearmed with this knowledge and with a more precise regimen of selection criteria. We might indeed learn a bit more about the nature of the "market" for these kinds of new TLDs.

Appendix A: Background

Let me briefly recap the background that has led us to where we are.

In November, 2000, the ICANN Board authorized seven new generic Top Level Domains (gTLDs) to be added to the DNS. These seven new gTLDs were authorized as a "proof of concept" to gain understanding of the practical and policy issues involved in such an undertaking. Though the last 15 years have seen the steady addition of (at least initially) small country-code top-level domain (ccTLD) registries to the DNS root, ICANN was venturing into uncharted territory by approving a set of globally open registries expected to attract massive numbers of registrations at launch, raising a range of novel technical and policy issues.

Three of these new gTLDs (.museum, .aero, .coop) were "sponsored" TLDs or "sTLDs", that is, they were delegated to not-for-profit policy organizations that were demonstrably supported by and accountable to a defined community. As such, sponsored TLDs operate within an agreed charter that clearly defines who can be a registrant, and how policy decisions will be made within that community. The remaining four gTLDs (.biz, .info, .name, .pro) were "unsponsored" TLDs or "uTLDs", operating commercially more like the traditional .com, .net, and .org registries. Three of the unsponsored TLDs were also "restricted," insofar as they applied defined criteria to determine eligibility for registration (.biz, for example, is for business organizations).

There were many issues that constrained the ICANN Board to adopt a cautious approach towards the pace it was willing to launch new gTLDs (whether sTLDs or uTLDs). The technical community was concerned that the effect of adding many new gTLDs to the root was unknown, particularly on performance, and particularly if the "shape" of those new gTLDs turned out to be flat. A central advantage of a small root zone file is the relative stability of the information in that file, meaning that changes in the network locations of top-level domain nameservers are rare, so that the potential of errors is minimized and so that long-persistence TLD resource records can be employed. The Business and Intellectual Property constituencies were concerned that the need for new gTLDs was not compelling; perhaps even more importantly, new gTLDs created a potentially significant expansion of problems with cybersquatting and trademark infringement. It was also uncertain how best to handle a range of start-up problems in allocating domain names to registrants in the face of multiple competing demands for the same name, and in the face of trademark claims. The ISP constituency was concerned that simultaneous addition of a large number of popular gTLDs would create a large customer-education and support burden for its members. Registrars, while desiring the additional business opportunities that new gTLDs could bring, were concerned that the new policies in new gTLDs ensure fair treatment to all registrars, and not create burdensome technical interface requirements. Ultimately, the policy recommendations developed by ICANN's Domain Name Supporting Organization (DNSO) called for ICANN to add new gTLDs to the root in a careful and disciplined manner, starting with a proof of concept round that would allow for the testing of different models and the careful evaluation of their successes and shortcomings.

As the process unfolded, there were other concerns, as well. There were differing views as to how much involvement ICANN should have in constraining new registries through its legal framework. Some felt that ICANN should have minimal involvement, adopting a laissez-faire attitude towards the marketplace. Others felt that more oversight was needed to create a level playing field for competition to work fairly and effectively, or to protect consumers. For example, some felt that business plans of prospective new gTLDs were not of ICANN's concern; others felt a review of business plans was necessary to provide some comfort that the proposers had the wherewithal to launch a registry business that could survive in the marketplace, at least in the first year of operation, and that any large-scale failures would significantly and needlessly damage Internet users' confidence in the reliability of the DNS. Many of these differences of perspective still exist.

These seven new gTLDs were selected from among 44 complete proposals submitted. The proposers were each required to pay US$50,000 as a non-refundable proposal fee, with the full knowledge that only a few would be selected and that those who were not selected could not expect any special privilege in any future undertaking. The amount of the fee was calculated based on the estimated future total costs (across all proposals, both those that were selected and those that were not) to ICANN of evaluation and selection, negotiation of agreements, implementation, evaluation of the results and other follow-up activities, including any possible litigation.

The selection criteria were quite broad. The following criteria were used in making recommendations to the Board;

  • The need to maintain the Internet's stability, and especially the protection of domain-name holders from the effects of registry or registration-system failure.
  • The extent to which selection of the proposal would lead to an effective "proof of concept" concerning the introduction of top-level domains in the future, including the diversity the proposal would bring to the program, such as fully open top level domains, restricted and chartered domains with limited scope, noncommercial domains, and personal domains; and a variety of business models and geographic locations.
  • The contribution that a particular new gTLD might make to enhance competition for registration services at the registry and registrar level.
  • The extent to which the proposal would enhance the diversity of the DNS and of registration services generally;
  • The contribution that a particular new gTLD might make to enhance the utility of the DNS.
  • The evaluation of delegation of policy-formulation functions for special-purpose gTLDs to appropriate organizations.
  • The extent to which the proposal would meet previously unmet types of needs.
  • The extent to which a proposal incorporated appropriate protections of rights of others, including intellectual property rights, in connection with the operation of the gTLD, especially during the start-up phases.

The Board made its choice of the seven new gTLDs following extensive community input. The choice was made without prejudice as to the future status of the 40 or so proposals that were not accepted at that time.

Consistent with the vast weight of the public comments, the Board had difficulty addressing proposals where registration in the proposed gTLD was based on the content of websites. In the end, the Board accepted several "semantic" gTLDs. Any restrictions on registrants were based on who registrants are (i.e. a museum, a cooperative, a professional, a business, etc.), not on the contents of the registrant's website.

Appendix B: Documents from the Tralliance Corporation/ IATA

The International Air Transport Association (IATA) was one of the original bidders in the 2000 round of solicitations for proposals for new gTLDs but was not successful in the initial round. They were seeking a .travel gTLD.

Since that time, another corporation, the Tralliance Corporation (president: Ron Andruff) has been urging ICANN to launch a new round of gTLD solicitations, because they, too, are interested in seeking a .travel TLD.

The appended documents indicate the new arrangement under which IATA and the Tralliance Corporation would provide a single combined bid for .travel, should a new round of solicitations arise.

It should be emphasized that others of the original bidders for sponsored gTLDs have also shown continuing interest in being considered in a new round of solicitations

Part II. Implementing the NTEPPTF Report Recommendations for Evaluation of New gTLDs

The Report of the New TLD Evaluation Process Planning Task Force (NTEPPTF) contained a series of recommendations to the ICANN Board. This part of the overall Action Plan regarding new gTLDs addresses implementation of the most critical recommendations contained in that Report.

Besides recommendations regarding budget and schedule, these recommendations can essentially be grouped as follows:

1. As a matter of highest priority, immediately launch a program to address the Questions specified in Section 7 of the NTEPPTF Report as being of Criticality 1.

2. Immediately launch a Monitoring Program with regard to the existing new gTLDs as specified in Section 8 of the NTEPPTF Report.

3. With respect to the above recommendations, follow the Methodologies recommended in Section 9 of the NTEPPTF Report.

4. Consider to what extent the Board wishes to initiate planning of and proposal solicitations for new rounds of gTLDs in parallel with proceeding with the first two recommendations.

This last recommendation is addressed in Part I of this Action Plan. The overall recommendation of this document is that the Board authorize the President to proceed with dispatch with the first three recommendations subject to the constraints of the balance of funding available. The three recommendations and the funding/schedule questions are addressed in more detail below.

Initial Evaluation Project

The NTEPPTF Report poses a series of questions to be addressed as the key parts of any evaluation. The Report also provides commentaries as suggestions on how to answer these questions. In the Report twelve of these questions are designated as Criticality 1. I designate a project to address these Criticality 1 questions as the Initial Evaluation Project.

Many of these questions are complex, and providing satisfactory responses, even at a high level of analysis, may exceed the limits of funding available. Some parts of some of the questions can be addressed by volunteer task forces or by ICANN staff. However, I recommend (see below) that a limited RFP be issued to solicit an Evaluation Team or Teams. Bidders would be asked to state how they would address some or all of the twelve Questions at different levels of funding. A given bidder, for example, may only choose to bid on the Legal questions.

Some of the Questions overlap the Monitoring Program described in the next section, particularly at this time when the gTLDs are much further along into their production cycles than was envisaged when the Monitoring Project was initially proposed. Obviously, duplication of effort must be avoided.

For ease of reference, the twelve Criticality 1 Questions are (Appendix 1 of the NTEPPTF Report also provides comments on how answers to these Questions might be derived):


1. Has there been any measurable or otherwise determinable effect on the DNS performance, security, and stability with the introduction of the new gTLDs? (Note: this Question can be answered in the same way as the first item of the Monitoring Program).

2. Have the new TLD registries incorporated technologies, including new technologies, that can adversely affect the performance of the DNS, violate DNS technical standards, or cause existing applications to fail?


3. How effective have start-up mechanisms been in protecting trademark owners against cybersquatting and other abusive registrations?

4. How often and how successfully have advance filtering and other mechanisms for enforcement of registration restrictions been used, both in sponsored gTLDs and in restricted unsponsored gTLDs?

5. To what extent and in what timeframe have the registry operators provided free, realtime access to a fully searchable Whois database?

6. What effect have the new gTLDs had on the scope and competitiveness of the domain name market, in terms of opening new markets, and in their effectiveness on existing TLDs and registrants?

7. Are adequate management policies and safeguards in place to ensure protection against accidental or malicious acts that could substantially interfere with continuity of service?

8. How effective were the different start-up mechanisms employed from both a functional and operational perspective? To what extent did they achieve their objectives or, conversely, cause consumer confusion, delays, legal issues, operational problems, or impediments to smooth implementation?


9. How well do the [legal] agreements provide a framework for the addition of new gTLDs?

10. Have the new gTLDS encountered any legal or regulatory problems that were not considered at the outset and, if so, how could they be avoided?

11. Have there been an unusual number of legal disputes during the startup period and how well have they been addressed?


12. To what extent were the Board's original objectives met through processes that were used for selection, approval, negotiation, and implementation? How could these processes have been streamlined?

Monitoring Program

Section 8 of the NTEPPTF Report recommends a six-part Monitoring Program to be launched immediately. I comment on how to proceed with each of these part separately.

As noted in the previous section, events have overtaken some of these monitoring considerations to the point now that many of them address recording events that have already occurred rather than monitoring on-going events. In this regard, they overlap the Initial Evaluation Project defined in the previous section, and the requirements can be combined.

The components of the Monitoring Program are:

1. Monitor the effect on the performance of the root of the addition of the new gTLDs as the number of registrants grows, focusing (given the limitations of measuring tools that can distinguish cause and effect) on detecting significant changes and effects that are demonstrable over time.

This can be combined with Question 1 in the Initial Evaluation Project.

I recommend that the RSSAC be asked (a) what monitoring tools are available that can provide the information requested; (b) how such a monitoring program can be implemented; and (c) whether there is any reason so far to believe that the addition of the seven new gTLDs has had any significant or detectable effect of the performance of the root.

The new gTLDs are growing steadily, but not by leaps and bounds. It may well be that the RSSAC would advise us that there has been no detectable effect at this point that can be directly attributable to the new gTLDs.

2. Monitor any operational performance problems that affect the stability of the DNS, particularly focusing on the ability to enter new registrants into the various registries, and monitor any failures associated with access to those registrants.

I recommend that each new gTLD registry and accredited registrars be asked to self-report on these matters documenting known problems that have occurred in this area. Subject to confidentiality agreements, these reports would be posted for feedback and comment.

3. Monitor the accuracy and completeness of the Whois data.

I recommend that a sampling study be initiated for a 6-month period, addressed at both assessing the current status of the Whois databases, and monitoring changes as new registrations are added.

4. Monitor the degree to which registries are conforming with their charters and other key contractual provisions, particularly with regard to including only registrants that conform to charter specifications.

This primarily applies to sponsored gTLDs. Such a study has been separately recommended as part of the proposal for another round of new sponsored gTLDs (see Part I of this overall Action Plan). This study can and should be launched immediately.

5. Monitor any particular start-up issues, particularly with regard to sunrise or landrush problems.

The start-up periods are now over. Each new gTLD should be requested to document and self-report their experiences in this area. Since, however, there may be confidentiality or legal problems that may limit the feasibility of this, an independent study should be contracted to the extent that funds allow.

6. Monitor compliance with contractual provisions regarding acquiring data and making it available to ICANN.

ICANN can report on this.

Much of the above can be overseen and even accomplished by the proposed Coordinator of New gTLD Planning and Evaluation, once this position has been filled (see next section). In the interim, some temporary staff support will be needed to move this project forward.

Evaluation Methodology and Schedule

NTEPPTF recommended, in Section 9 of its report, that addressing the above Questions and Monitoring Programs be conducted at a "high-level", without becoming "bogged down in extensive detail". Furthermore, existing data (such as that provided under the agreements between the new gTLDs and ICANN) be used wherever possible.

NTEPPTF also recommended that the studies be guided by a TLD Evaluation Team (TEAC). The members of TEAC agreed to serve on the TEAC should that be of value to the Board. For reasons of continuity, I recommend that the TEAC consist of members of the NTEPPTF.

The NTEPPTF also recommended that ICANN solicit bids for an Evaluation Team or Teams to conduct an evaluation of the Questions listed in the previous section. As discussed in the previous section, I believe this should be done within the limits of the funding available; and that volunteer assistance be solicited wherever possible to complement that effort.

In Section 9 of its report, the NTEPPTF proposed an aggressive schedule for completing this evaluation. That schedule can no longer be met, and given the realities of Reform and Evolution transition (should the ERC recommendations be adopted by the Board), it will be very difficult to adhere to an aggressive schedule. ICANN is in process of recruiting a new gTLD Planning and Evaluation Coordinator, and that staff position needs to be in place before an aggressive schedule can be pursued. At this stage, it is recommended that ICANN proceed with the process of soliciting for an Evaluation Team(s), but define a precise schedule until the Coordinator is on board.


The balance of funds available for the new gTLD project is $385,000. On advice of Counsel, I propose that $35,000 be retained as a contingency for future litigation (there is still residual litigation relating to an outstanding lawsuit). The balance of $350,000 is available for the evaluation, including the Monitoring Project.

I propose that the Board authorize me, proceeding with the guidance of the TEAC, to solicit bids as outlined above to determine what can be accomplished at different levels of expenditure: $250,000, $300,000, $350,000. Wherever possible and meaningful, I, acting with the advice of the TEAC, will solicit volunteer support to address the questions raised, but expectations should not be raised too high in this regard.


Summarizing, therefore the recommendations contained in the above, I recommend:

1. The Board should appoint a TLD Evaluation Advisory Committee (TEAC) consisting of former members of the NTEPPTF.

2. The Board should authorize the President, proceeding with the guidance of the TEAC, to solicit bids for an Evaluation Team(s) as outlined above to propose what can be accomplished at towards accomplishing the Evaluation Project (consisting of addressing the Questions defined as Criticality 1 in the NTEPPTF Report) as summarized above, at the different levels of expenditure: $200,000, $250,000, $300,000, perhaps complemented by volunteer support; and to present that information to the Board as the basis for a decision on proceeding forward with the Evaluation.

3. The Board authorize the President to spend up to $50,000 towards implementing right away those aspects of the Monitoring Project summarized above that are not being addressed as part of the Evaluation Project.

4. The Board should recognize that, because of other staff commitments, a precise schedule cannot be defined until the New gTLD Planning and Evaluation Coordinator is recruited and on board.

Part III: Evolving the Top Level gTLD Namespace

Since well before ICANN's new TLD process in 2000, there has been discussion about the best way to address the semantic element of new TLD creation.

Some have argued that the best approach is to allow each would-be TLD registry operator to select which new TLD string(s) it wishes to operate (on the basis of its own analysis, market research and/or reference to the intended community to be served), with little or no editorial role for ICANN in selecting new TLD proposals. This is the approach ICANN took in 2000: each proposer was asked to specify the TLD string(s) it wanted, including an explanation of its rationale and, for sponsored TLD proposals, the relationship of the string to the intended community to be served and to any restrictions or limitations on registration.

Others have argued that the ICANN community should undertake to "rationalize" the new TLD space according to a defined taxonomy of names. This is the approach that was taken when the Domain Name System was first designed and deployed in the mid-1980s. [See RFC 1034, RFC 1035, RFC 1591]. RFC 1591 documented post-hoc the initial taxonomy decisions, which established seven three-letter "generic" top-level domains (EDU, COM, NET, ORG, GOV, MIL, and INT) and an expandable number of "country-code" top-level domains, drawn from the ISO 3166-1 table. In other words, the initial architects of the DNS made a collective judgment about the best taxonomic structure for the DNS, which the IANA then implemented by identifying registry operators for each generic and country-code TLD and delegating registry authority, according to the policies documented in RFC 1591.

There are any number of pros and cons to each of these approaches:

  • The laissez-faire approach puts responsibility for TLD string selection in the hands of those who are arguably in the best position to make market-driven or community-driven judgments – the potential registry operators. For example, the proposer of .museum was arguably in the best position to determine which particular string of letters would be most acceptable to the global community of museums, which spans a vast range of cultures and languages.
    • On the other hand, the laissez-faire approach may result in a DNS that is confused and confusing, undermining the basic objective that domain names should be easy-to-remember pointers to the IP addresses that identify nodes, services, and links on the Internet.
  • The taxonomic approach would allow the broad community of ICANN stakeholders the ability to participate in designing a rational and comprehensive set of available TLD strings, with the objective of articulating a system with the greatest utility for all Internet users. This approach would be arguably most likely to result in a coherent system of easy-to-remember TLD strings that make sense to users.
    • On the other hand, the taxonomic approach would require ICANN to undertake the doubtless difficult, contentious, and time-consuming process of creating a global consensus around a particular list of TLD strings. It could interpose that process between now and the start of ICANN's next round of new TLDs selection.

Paragraph II.C 8 of the Amendment to the Memorandum of Understanding between ICANN and the United States Department of Commerce signed in September, 2002 states that ICANN should "Continue the process of implementing new top level domains (TLDs), which process shall include consideration and evaluation of . . . (b) The creation and implementation of selection criteria for new and existing TLD registries, including public explanation of the process, selection criteria, and the rationale for selection decisions." The need to understand how the top level TLD namespace should be expanded is indeed a prerequisite to defining such "selection criteria".

Recommendation: As ICANN proceeds with its new TLD evaluation process  – and, if the Board concurs, with an additional round of new sponsored TLDs – this basic question of taxonomic rationalization should be addressed within the ICANN process. Accordingly, it is my recommendation to the ICANN Board that the DNSO and its Names Council be requested to develop and submit its advice and guidance on the issue.

Click here to submit a comment on this action plan.
Comments will be most useful if they are made by 10 December 2002.

Comments concerning the layout, construction and functionality of this site
should be sent to webmaster@icann.org.

Page Updated 11-Nov-2002
©2002  The Internet Corporation for Assigned Names and Numbers. All rights reserved.