Dear Jeff, We would like to respond to your email suggesting that this proposal is essentially an open restricted gTLD. We suggest that differences are substantial. We are recommending a “public interest goals” framework as sought by the GAC and the Board ("For strings representing generic terms, exclusive registry access should serve a public interest goal.” - Beijing GAC Advice). Differences between our Public Interest Closed Generic TLD and an open restricted TLD: 1. There is only a single registrant, the Registry Operator. The Registry Operator will have a governance structure that ensures that the entity meets its stated goals, which is an essential feature of any entity. The governance concepts we suggest are ones well suited to meet the specific public interest goals of this type of TLD. 2. A restricted gTLD may certainly impose strict entry requirements on its registrants, as..bank and ,pharmacy do. However such gTLDs say nothing about limiting any behavior on the registry, including behavior that would be commonly regarded as not in the public interest. For example, when one of us was on the Board and leaving, there was a significant dispute between .pharmacy and the Canadian pharmacies. The entry requirements had been set in a discriminatory fashion so that the Canadian pharmacies could not comply and were therefore excluded. This clearly was done for U.S. industry purposes to restrict knowledge of alternatives and competitive possibilities between the two groups. That may be in the interests of U.S. pharmacies but it is not in the global public interest. Similarly, using the .disaster example and permitting it to be delegated to the Red Cross risks not serving the public interest. There are a number of prominent organizations besides the Red Cross that work in the area of disaster mitigation. The exclusion of all other major organizations from participation in the TLD would both bias the content of the TLD toward only the Red Cross affiliates and branches as well as limit its usefulness to global internet users. Our model would not allow this to happen, since acceptance of the TLD as being in the public interest would rely upon the support of a sufficient number of organizations, providing in the aggregate sufficient coverage of this disaster space. Only by insisting upon such a process can the required level of trust be achieved to consider the closed generic TLD to be truly in the public interest. Note: We are *not* requiring that all eligible registrants be allowed into the public interest closed generic TLD; that choice will be made by the registry and its governance group. The registry and its governance group are fully empowered to decide on the structure of the TLD and all second level domain names (and by extension third level and beyond are owned by the registry). Our rules ensure that the public interest closed generic TLD benefits the public interest and not just the registry. In our proposal, there are a number of requirements to ensure that such a situation will occur, including a "sufficient representation" issue as one criterion for approval, a governing body for those who include themselves in the application, fair entry and exit criteria for the life of the TLD, and an inability to repurpose the TLD in a way that modifies its original and mandated purpose. In addition, our proposal allows the domain space at the second and further levels to be shaped somewhat arbitrarily -- and potentially uniquely -- by the registry in the service of the public interest rather than being only a list of second level "registrants." Our proposal has no registrants other than the applicant, but it does include a collection of like minded not-for- profit organizations. This provides a possibility of some quite innovative use of the closed generic domain structure. A public interest closed generic TLD, we submit, permits a new form of TLD structure with different objectives, applications, review, and delegation processes. 3. Our closed generic public interest TLD proposal focuses solely upon TLDs that are reviewed and accepted to be in the public interest, which we would argue as being the essence of what the GAC and the Board were trying to accomplish. Accordingly, Specification 11 does not apply. Overall, our framework features exclusive registry access to a generic string -- with express provisions designed to protect the public interest. We very much look forward to hearing the reactions and opinions of others in the working group, and especially our discussion regarding the proposal at our next working group meeting. Regards, George, Alan, Kathy and Greg