IDNA Protocol Review and Proposals for Changes
In follow up to the RFC4690 that was made available in September and announced on http://www.icann.org on 19 October 2006, a review of the IDNA protocol is underway.
The review is managed through the IAB and the IETF and was initiated to revise the protocol from being based on the Unicode version 3.2 (as was the latest version at the time of the development of the protocol in 2003) to the current version 5.0 of Unicode. This latter version contains many more characters that hence otherwise would not be available for usage in domain names.
As previously announced, in September, the basic framework discussing such review was published as RFC4690. This paper is discussing issues related to language specific character issues where the same script is used across different language, issues related to cases where languages can be expressed by using more than one script, bi-directional cases as well as the topic of visually confusing characters.
In response to some of these issues a set of internet-drafts providing suggested solutions and revisions to the IDNA protocol was released publicly in the past couple of weeks. ICANN urges all interested Internet users to take part in this development work. Of particular importance the proposals includes revising the IDNA protocol to be based on an inclusion list of characters. The three internet drafts can be found via the links below and the last link provides an initial version of the inclusion list of characters. Part of the work that is still underway includes a process by which the community will be able to add characters to this inclusion based list, however, initially if these changes are accepted and implemented it means that not all characters will initially be available for registration in domain names and only those characters passing through the mentioned process will be made available.
- An overview with proposed issues and changes for IDNA:
- A suggestion for solving an IDNA problem in right-to-left scripts by revising the stringprep profile:
- An initial suggested inclusion based IDNA Unicode Codepoints based on Unicode 5.0
For more information about IDNs see http://www.icann.org/topics/idn