ICANN is opening a public comment forum for the annual review of the IDN ccTLD Fast Track Process. The input received will be used to determine appropriate updates to the current version of the Final Implementation Plan of the IDN ccTLD Fast Track Process [PDF, 851 KB] (FIP) and for the proposed IDN ccTLD policy [PDF, 376 KB], which is currently under consideration. Community members, technical community, and specifically those who were directly involved in the IDN ccTLD Fast Track Process, including requesters, public authorities and end users of the delegated IDN ccTLDs, are kindly asked to provide feedback on their experience in the process.
Section I: Description and Explanation
To help initiate the feedback process on the review of the IDN ccTLD Fast Track Process, ICANN staff has listed the following topics as ones that can benefit from further discussion. However, it should be noted that the review is not limited to these aspects, and comments are welcome on any part of the Fast Track Process.
Upper case U-Labels in the context of string similarity
The string similarity evaluation was implemented in the original FIP [PDF, 497 KB] based on the recommendations of the Internationalized Domain Names Working Group (IDNC WG). It was introduced to minimize the risk of end-user confusion due to similarity between two strings. Minimizing the risk of string confusion has at least two distinct functions: (i) provide a predictable user experience, where user can unambiguously use a domain name "in common fonts, in small sizes at typical screen resolutions", and (ii) contribute to a secure user experience, where user is protected from the possible spoofing and phishing threats. String confusion issues can involve two or more strings that are identical or are so confusingly similar that they cannot coexist in the DNS, as stated in Section 5.5 String Confusion and Contention of the current FIP [PDF, 851 KB].
For the scripts using both upper and lower case letters, IDNA2008 protocol generally disallows the use of upper case letters in a valid U-Label. However, the end-user can still input letters in upper case form for a U-label in applications, which can then map upper case forms of the letters into their lower case forms for further processing, as also discussed in the informational RFC 5895 and RFC 5992 Appendix A. Many applications, like browsers, implement such mapping from upper case to lower case for different scripts, e.g. for Greek and Cyrillic. Thus, both the upper case and lower case letters remain relevant to the discussion around string similarity.
Based on the context, community feedback is sought on whether upper case forms of U-Labels, where such labels are disallowed by IDNA2008 protocol, should be considered relevant for the string similarity review of IDN ccTLD labels. Further, as ASCII TLDs may also be used in upper case or lower case, for string similarity purposes, should the applicable form(s) of U-labels be compared with both upper case and lower case forms of labels in ASCII?
Viability of EPSRP as a second review
On 27 June 2013, the ICANN Board approved the proposed amendments of the FIP to implement a two-panel process for string similarity review in the IDN ccTLD Fast Track Process. The first string similarity review was maintained as a function of the external and independent DNS Stability Panel, also acting as the "Similarity Review Panel". To provide a second and final similarity review of the string(s) found to be confusingly similar by the "Similarity Review Panel", and using a different framework, an external and independent "Extended Process Similarity Review Panel" (EPSRP) conducts a review of the requested IDN ccTLD string, only if so requested by the applicant. The EPSRP reviews the requested string(s) on the basis of the framework described in the Guidelines for the Extended Process Similarity Review Panel [PDF, 86 KB] and with experimental details and results as available in the reports published for the first set of evaluations.
Due to the nature of the methodology, it takes significant time and effort to design the experiments, collect and analyze the data and publish the results. In the current phase, it has taken a team of four experts up to eight months to conduct the first set of evaluations for three strings, requiring more than 400 person-hours for designing, conducting and reporting on the experiments. The methodology has been clearly documented, for independent verification, along with detailed experimental results, providing added transparency in the process.
The community is invited to comment on the effectiveness of the EPSRP methodology in terms of the experimentation designed, results achieved and the transparency provided. How effective is the EPSRP method in terms of the time taken and effort required? Should this methodology be taken forward or revised? In the latter case, in what ways?
Inclusion of Name Collision in the Fast Track Process
ICANN's mission and core values call for ICANN to preserve and enhance the operational stability, reliability, security, and global interoperability of the Internet's system of unique identifiers (names, IP numbers and protocol parameters). In pursuing this mission and values and following the direction of its Board of Directors we have prepared an informational resource for new ccTLD managers to consider regarding the name collision risks.
Feedback is requested on whether the FIP should be updated to include a section on Name Collision with recommendation to follow the guidelines available at https://www.icann.org/resources/pages/cctld-mitigation-2014-10-02-en.
Using machine readable XML format for IDN Tables
As part of the IDN ccTLD Fast Track application process, applicants are required to submit data in the form of IDN Tables (or language tables) which present the list of code points, variants and any other rules which will be supported at second level. The FIP encourages the requesters to review the existing IDN Tables posted in the IANA repository to promote re-use for consistency.
ICANN, in collaboration with the community, has been working over past few years to develop a machine-readable format for representing this data. The work has been motivated by the representation of such data for the Root Zone, now being referred to as Label Generation Ruleset (LGR). However, the framework is equally usable for representing data and rules at second and other levels. ICANN is also developing an LGR Tool to create, use and manage LGRs in this machine-readable format. Having IDN Tables in this format will allow a consistent representation across all levels and across registries, thus promoting their re-use.
What should be considered when updating the FIP to support submission of IDN Tables in the machine readable format?
Section II: Background
When the ICANN Board of Directors approved the Final Implementation Plan for IDN ccTLD Fast Track Process (FIP) at their meeting in Seoul, South Korea, 30 October 2009, they also directed the staff "to monitor the operation of the IDN ccTLD Fast Track Process at regular intervals to ensure its smooth operation, and, subject to Board review, update the process when new technology or policies become available, with the goal to efficiently meet the needs of Fast Track Process requesters, and to best meet the needs of the global Internet community."
FIP has been revised multiple times since its initial approval. The various versions (in reverse chronological order) are given below:
- Current version of FIP, revised on 5 November 2013, available at: https://www.icann.org/en/system/files/files/idn-cctld-implementation-plan-05nov13-en.pdf [PDF, 851 KB]
- FIP version revised on 4 June 2012 available at: http://www.icann.org/en/resources/idn/fast-track/idn-cctld-implementation-plan-04jun12-en.pdf [PDF, 886 KB]
- FIP version revised on 15 December 2011 available at: http://www.icann.org/en/resources/idn/fast-track/idn-cctld-implementation-plan-15dec11-en.pdf [PDF, 804 KB]
- Original FIP version, published on 16 November 2009, available at: http://www.icann.org/en/topics/idn/fast-track/idn-cctld-implementation-plan-16nov09-en.pdf [PDF, 497 KB]
Section III: Relevant Resources
- Latest version of the Final Implementation Plan for the IDN ccTLD Fast Track Process: https://www.icann.org/en/system/files/files/idn-cctld-implementation-plan-05nov13-en.pdf [PDF, 851 KB]
- Guidelines for EPSRP: https://www.icann.org/en/system/files/files/epsrp-guidelines-04dec13-en.pdf [PDF, 86 KB]
- Guidelines for Name Collision: https://www.icann.org/resources/pages/cctld-mitigation-2014-10-02-en
- Machine readable XML format for Linguistic Data: https://tools.ietf.org/html/draft-davies-idntables
Section IV: Additional Information
Details for successfully evaluated IDN ccTLD Fast Track applications are available at: https://www.icann.org/resources/pages/string-evaluation-completion-2014-02-19-en.