Return to tldreport Forum - Message Thread - FAQ

Username: khansen
Date/Time: Tue, November 14, 2000 at 6:47 PM GMT
Browser: Microsoft Internet Explorer V5.5 using Windows NT 4.0
Score: 5
Subject: JVTeam Response to .per Evaluation

Message:
 

 
To: ICANN
From: NeuLevel (formerly JVTeam)

Comments on Report on New TLD Applications  – Appendix B, Summary of JVTeam .per Application

NeuLevel (previously known as JVTeam) commends ICANN for the thorough evaluation performed by the ICANN staff and review team of the new gTLD Applications provided in its November 9th Report on New TLD Applications.  In the interest of providing ICANN and the Internet community with information related to areas in which the reports suggests clarification or elaboration is required, we would like to offer the following comments to the summary of NeuLevel’s .per Application authored by the review team and found in Appendix B of the Report.  Consistent with the open and transparent manner in which the selection process is being conducted, we are posting these comments to the ICANN Public New gTLD Comment Forum.

The following comments are prefixed by the section heading of the Summary to which they refer.

B.1 Use of SLD in lieu of multiple sub-domains for family and surname: We are open to alternative structures for a personal name space, such as .per.  Our reference to structuring it as an SLD is simply the initial proposed default structure.  A number of interesting concepts exist to provide structure and increased access to desirable names while also providing uniqueness to a personal name space.  These include the use of multiple levels as name separators, geographic or cultural SLD specifiers, and use of certificate identifiers or postal codes as uniqueness numbers.  We propose to work with ICANN and the internet community to finalize a suitable structure for a personal name space.  We think that developing a personal name space meeting community needs would be of significant value in the long run and also requires some flexibility in the ultimate structure in order to achieve that goal.

B.2.d Co-active SRS Data Centers:   The databases in each of the two SRS data centers (existing NeuStar owned and operated facilities in Chicago, IL., and Sterling, VA.) are both active for updates.  Commits to the logical database from an SRS application process are simultaneously posted to both physical database servers using a distributed two-phase commit protocol that guarantees that the update is posted to both copies in a synchronous real-time fashion.  Please see Section III.2.13.1 of our Application for more information.
B.2.e Security: The billing and collection system, consistent with the rest of the SRS, utilizes advanced security measures including digital certificates for strong session authentication, and object-level signatures for strong authentication and non-repudiation of critical transactions requiring multi-entity authentication (e.g. transfers).   Please see Section III.2.6.2 of our Application.  In addition, individual userids and passwords are employed to identify individual registrar personnel in transaction logs.  NeuLevel will generally employ extensive military-grade-style security technologies, including biometric systems for physical data center access, and physical security token authentication (e.g. smartcard) for remote administration access, consistent with existing practices of its parent companies (NeuStar and Melbourne IT).  But more important than the technology, we subject ourselves to regular Code of Conduct audits which include an audit of system/data security measures to ensure not only consistent security policy implementation but openness and transparency of our adherence to these measures.

B.6 XRP Protocol: The new proposed registry-registrar protocol in our Application is called extensible registry protocol (XRP).

B.6 gTLD structure and impact on caching: As mentioned in our comment to B.1 (above), some further development and trailing of structures for a personal name space such as .per are appropriate.  We agree that one of the many considerations in addition those mentioned in B.1 is scalability.  Caching behavior has a significant impact in DNS scalability.  However, we believe that many of the potential .per structures, as well as the default SLD structure proposed, can be established to not adversely impact caching behavior and therefore ultimate scalability.  The statistical behavior of caching of to a personal name space is comparable to that associated with queries to other individual identifiers, like telephone numbers, with which we have extensive familiarity.

B.7 Availability of names and uniqueness suggestions: As commented in reference to B.1 (above), providing access to an adequate number of convenient names is a key requirement for the ultimate structure of a personal name space.  The proposed default structure of .per as an SLD provides a uniqueness mechanism to allow users to employ their own names, and a standardized way for SRS client systems (registrars) to utilize information provided in XRP responses to present to the user alternative options or variants where required.  We believe that the complete development of these algorithms should be done collectively with the community of interest (and/or ICANN at its option) and propose to do so as part of the launch plan for the gTLD in the Application.

NeuLevel hopes the above comments and clarifications will be of assistance to ICANN in facilitating a complete assessment. We appreciate the opportunity to respond, and as always, are available to provide further clarification where needed.

Ken Hansen
NeuLevel    
     

 


Message Thread:


Privacy Policy | Terms of Service | Cookies Policy