DESIGN ARCHITETTURA DIGITALE ANALOGICO SPA

1. TLD String(s) Requested.

   .pid

2. Category.

   New Services, Message Routing.
   
   Design Architettura Digitale Analogico SPA, (“DADA”) requests the .pid TLD to provide a name space which represents individuals, usually by their actual name, for message routing services. Message routing is a new service generally not available on the Internet today and qualifies DADA in the new services category and message routing group.

3. Sponsor, Registry Operator and Subcontractor.

   a. Sponsor. Unsponsored Application.

   b. Registry Operator. DADA is a limited liability company organized in January of 1995. DADA claims to be one of the leading Internet companies in Italy. It derives revenue wholly from the sale of web-based products and services with respect to the following activities: (1) acting as a community portal (offering services to DADA's community members), (2) acting as an Internet service provider, (3) providing technological services to businesses such as domain name registration, housing, hosting, development of software, web sites, Intranet and VPN, (4) online advertising, (5) e-commerce and (6) incubator services.

   c. Subcontractor. None.

4. Registry-Registrar Model.

   Domain names registered in the .pid TLD will be registered through registrars only. DADA does not anticipate any direct contact between the registrant and the registry. In its initial start-up phase, DADA plans to limit the number of registrars to five. Future registrars will be selected by the registry based on compliance with ICANN rules and whether they will offer the widest and most differentiated range of services.

1. Summary Description of Proposal.

   The proposal suggests creation of a .pid TLD. This TLD would be primarily used for registering humans. Unlike standard domain names, a .pid domain name would be expected to change name/address mappings regularly, routing messages to different addresses or devices as the matching person moved or changed tasks. They propose to automatically create and control common family names as second level domains. They offer no novel solutions for handling duplicate given names within a given family name.

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

   a. Total Capacity. The technical approach requires consultation of a central server at DADA for the majority of name lookups. It’s unclear if the proposed network and processing capacity would support that load.

   b. Projected Growth Rate. The applicant projects a rather aggressive growth rate over 5 years, projected 120 million .pid addresses in that time. Some aspects of their technical plan (e.g., customer support) do not seem equal to that level of load.

   c. Startup Period. The applicant does not seem to expect a large surge of activity when their proposed TLD first comes on line, and offer few details on how they might cope with such a surge. Much of the important functionality of the service is to be provided by the registrars, and it is unclear if registrars are ready to provide such services initially.

   d. Fault Tolerance. The proposed support for the system will reside in a single building in Florence, Italy. The applicant does not suggest any plans to develop other facilities anywhere in the world to support failures of the facility’s machines, network connections, or software. The design is essentially for a DB cluster consisting of two Sun 4500 machines. No plans are discussed to handle power outages, the unlikely (but not impossible) failure of both machines, catastrophic failures (fire), or network problems in the vicinity of Florence. The backup strategy does not discuss off-site copies of the backup data.
   
   The DNS service for the domain is similar in character to the service for the registry itself, consisting of a pair of replicated servers in a single building.

   e. Security. Registrars will be authenticated to the registry service by passwords. No discussion of dealing with the known shortcomings of passwords is provided. No discussion of authenticating the registry to the registrars is provided. The proposers provide some discussion of basic security mechanisms for the registry facility, but do not discuss issues of logging and auditing, monitoring of the facility for security problems, use of firewalls, etc.

3. Summary of Relevant Experience.

   The applicant has provided a DNS registrar service in Italy. The applicants provide little information on experience with handling other registration systems, large database systems, or systems, other than DNS registries, with strong reliability requirements. They do have 1.6 million subscriptions to the Supereva portal they run.

4. Apparent Implementation Risks.

   The applicant suggests a new style of name usage in the Internet. However, they seem to rely on research and development by registrar servers to provide much of the core technologies that differentiate the service from a standard service. There is significant risk that no one will do this work, reducing the proposed service to a standard personal Internet name service.
   
   They do not discuss how these multiple third parties would coordinate to provide coherent services that interoperate properly, so there is significant risk that the service will be fragmented and hard to use.
   
   The proposers do not provide fault tolerance for anything beyond failures of machines and software in their own single facility. There are risks that the service will be interrupted, suffer poor performance, offer highly variable performance to different users in different locations, and perhaps entirely lose its database of name/registrar mappings in case of a catastrophe.
   
   The proposers take a narrow view of the security concerns of the system.

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

   The applicants are essentially an ISP. They do not discuss the human resources that will allow them to perform the research and development necessary to make the proposed system work. They seem to rely entirely on an limited recruiting base, so their ability to grow and cope will depend on competition for a very limited pool of workers.

6. Advancing the State of the Art.

   The proposed TLD and its uses could significantly change the way people use the Internet to communicate to each other.

7. Other Comments.

   The proposal suggests an interesting idea, but is lacking in details required to ensure that a usable, reliable service results.

1. Applicant’s Representations.

   DADA is an Italian Internet company whose services include domain name registration, ISP, community portal, business services, online advertising, e-commerce, and a start-up incubator. There are currently five employees. At June 30, 2000, DADA had six-month revenues and net assets of approximately 13.9 million Lira and 169 million Lira, respectively.
   
   The revenue model includes a $6 annual registration fee and the following fees from registrars: $10,000 set-up fee, $10,000 training fee, $15,000 annual software license fee, and a $5,000 annual rights fee. The market is defined as those with Internet access and mobile phone subscribers; essentially mobile people who have multiple points of contact such as phone, fax, pager, mobile phone, e-mail, and web pages. The initial target market is the 655 million current mobile phone customers.

2. ICANN’s Evaluation.

   The strengths of this application include a creative approach to a new TLD and significant resources available. The weaknesses of this application include weaker implementation strategy and market demand assessment. Overall, there are stronger applications in this category from a business plan perspective.

1. Number of Comments.

   22

2. Support for the Application.

   One comment called the concept “very interesting and innovative.”
   
   Another supporter wrote “everyone, as a person, deserves his/her own space on the web.”

3. Opposition to Application.

   One writer expressed concerns about Dada S.P.A.’s track record as a registrar for .it in Italy.
   
   “[There] will be only one john.smith.pid: the ONE AND ONLY JOHN SMITH. This leads to . . . a run for the name.”