SARNOFF CORPORATION

1. TLD String(s) Requested.

   .i

2. Category.

   Sarnoff Corporation (“Sarnoff”) requests a new TLD primarily intended to support human domain names. Sarnoff targets a broad registrant base, large end user group and focuses primarily on commercial uses. Accordingly, Sarnoff qualifies for the general purpose category, personal group.

3. Sponsor, Registry Operator and Subcontractor.

   a. Sponsor. Unsponsored Application.

   b. Registry Operator. Sarnoff in cooperation with AtomicTangerine, Inc. ("Atomic Tangerine")^[1], submitted the application for the .i TLD on behalf of NextDNS, Inc. Sarnoff was established in 1942 as RCA Laboratories, the central research organization for the RCA Corporation. In April 1987, it became a subsidiary of SRI International as part of an agreement with GE. Sarnoff claims to be a leading global organization in video, broadcasting and displays and claims to have a worldwide reputation in the following areas of expertise: (1) communications systems and networking, (2) vision technologies, (3) optoelectronics, (4) integrated circuit systems, (5) internet technology, (6) integrated electronic and display products, (7) emulation products and technology, (8) life sciences and systems, (9) microstructure and display systems and (10) digital TV and entertainment. AtomicTangerine was launched on December 26, 1999 as a spin-off of the consulting practice of SRI International. AtomicTangerine claims to be the first e-business venture consulting firm, combining the disciplines of venture capital, technology innovation and strategic consulting, to create category killers and incubate new industries for companies of all sizes and at all stages of evolution.

   c. Subcontractor. A significant portion of the technical infrastructure, installation hosting, monitoring, administration and maintenance will be subcontracted to Exodus Communications ("Exodus"). Exodus was founded in 1994 and currently offers a range of Internet services including: (1) complex web hosting, (2) Internet connectivity, (3) Internet technologies and collaborative managed services, (4) Exodus professional services and (5) Arca Systems security services.

4. Registry-Registrar Model.

   Registration in the .i TLD will be open to any ICANN accredited registrar.

1. Summary Description of Proposal.

   The proposal calls for a personal name TLD primarily to support human domain names and secondarily to support domain names for embedded devices. The proposal pays particularly strong attention to the reliability and security aspects of running the registry for the domain.

2. Support of the Business Plan by the Technical Plan.

   a. Total Capacity. Quite adequate, assuming their projections are anywhere near correct.

   b. Projected Growth Rate. Realistic if they confine themselves to providing domain names for human beings. Perhaps an underestimation (possibly by a great deal) if they expand to include embedded devices and such usage becomes popular.

   c. Startup Period. Not completely addressed in the proposal. The overall capacity can handle a reasonably large instant migration to their service. The fact that it supports human being’s names only may limit the number of requests for domain names when it first comes on line.

   d. Fault Tolerance. Extremely well discussed. There seems little doubt they understand the issues of fault tolerance and have taken sensible precautions to avoid negative impacts from component failures.

   e. Security. The proposal is a model for registry security. If anything, certain aspects of the proposal are over-specified. Their planned precautions against cyber attacks are as good as they possibly could be, and include mechanisms to ensure their continued improvement over time.

3. Summary of Relevant Experience.

   The application does not demonstrate much recent experience with registries and registrars. The old SRI experience is of limited value, since SRI is very much in the background, and none of the principal technical people were involved in SRI’s historical experiences. Atomic Tangerine and Sarnoff seem to have little experience in these lines. They have excellent credentials in security, however, and exhibit competence with general database, reliability, and networking issues.

4. Apparent Implementation Risks.

   The proposal initially speaks of exciting possibilities for addressing users on the move and users with differing delivery requirements by providing a single personal address that can be translated into wherever the users actually want or need the data to go at the moment. However, ultimately the proposal only provides technical detail on how they will set up a highly reliable, impeccably secure registry at the scale they envision. Accordingly, one clear risk is that they will do no more than that and do nothing that actually supports real time translation of the personal domain names to differing addresses. In that case, the result would be an excellent registry that really only translated a personal domain name into a single fixed location, and did no other interesting things to support mobility, heterogeneity, and user customization.
   
   The proposal foresees relatively little software development, and thus budgets and plans for relatively little of it. The presumption is that off-the-shelf software will meet almost all the needs of the registry, and that a few relatively straightforward predictable pieces will meet all other needs. If this approach proves inadequate, the plan leaves little room for supporting heavy-duty software development.
   
   Very little is said in this proposal about how they will handle the special problems of providing name service for embedded devices. There is a strong risk that the resulting system will not handle those naming needs at all. Again, the worst case result is that it only handles human names well.
   
   The proposal contains a concrete solution on handling multiple users who want the same name. It is unclear if this solution will map well to the real world, particularly the portions of the real world where European-style human names are not used. There is thus some risk that the proposed system would be less useful outside Western countries.

5. Available of Human, Operational and Technical Resources to Cope with Unexpected Events.

   The plan has good ongoing resources to handle reliability and security concerns. The resources that would be needed to cope with failure to design all features and requirements of the system are less impressive.

6. Advancing the State of the Art.

   The proposal offers a vision for a new style of Internet service. As written, however, it offers no plan of realizing the new, unusual parts of that service, since it puts that burden on the shoulders of the registrars and their software. It also offers a nice example of a reliable secure registry service.

7. Other Comments.

   None.

1. Applicant’s Representations.

   Sarnoff in cooperation with Atomic Tangerine is filing the “Registry Operator’s Proposal” and has formed a new entity that will actually operate the Registry, entitled NextDNS. Sarnoff has a history of developing numerous new technologies and Atomic Tangerine is an e-business consulting practice. Both organizations are subsidiaries of SRI International. Sarnoff and Atomic Tangerine have 800 and 250 employees, respectively. The Registry’s mission is to create a new TLD to provide the underpinnings for the next generation of applications and services that benefit all Internet users. It plans to enable every user to have a personal domain name that serves as a service provider-invariant point of contact, through which various services and applications can be enabled. In addition to the standard personal name space, the Sarnoff application proposes additional services including but not limited to message routing and message transcoding.
   
   This plan proposes that each individual will have a personal domain and will have a unique ID for every Internet-enabled device. In addition to the standard personal name space, the Sarnoff application proposes additional services including but not limited to message routing and message transcoding.
   
   The revenue model includes three tiers of names: (1) free personal numeric domain names, (2) standard format third-level domain names at US$2 wholesale and (3) customized domain names at US$4 wholesale. The total estimated demand in year four is 76 million users of which 46 million would be paid users.

2. ICANN’s Evaluation.

   The strengths of this application includes a thorough assessment of the market by target segment. It contains a strong assessment of the marketing mix including a plan by quarter. It has significant experience in developing new services. The weakness of this application lies in a less than specific answer regarding the amount of capital required. The application includes a letter from Arnhold and S. Bleichroeder, Inc. which states that this firm has already contributed funds to Sarnoff in its efforts to become a top level domain registry and would be interested in raising additional capital. Overall, this is a stronger application in this category from a business perspective.

1. Number of Comments.

   38

2. Support for the Application.

   The proposal to provide free domain names would be helpful in developing nations.
   
   Proposal is designed for the “average Joe” and would “reduce the clutter on the net”
   
   One comment termed the application “very well though out . . . professional.”
   
   “[The plan to offer free domain names] is excellent for Internet users in developing countries who cannot afford the fees or cannot find a means to remit the fees.”

3. Opposition to Application.

   The question was raised whether .i would add anything over the current .net.
   
   Worries were expressed that .Sarnoff would become a “monopolistic registry.”
   
   “How will the PDA, cell-phone, voicemail, fax, . . . aspects of one’s ‘identity’ be designated.”
   
   Involving WHO would overly politicize the internet and feed into certain people’s paranoid conspiracy theories.

4. Substantive Comments and/or Questions.

   “Why,” one commentator queried, “would individuals want ten digit numbers as their domain names.”
   
   One commentator was concerned that $10 a year would be too much of a financial burden for individuals in some countries.