[Gnso-epdp-idn-team] Reminder: Share Current Views on String Similarity Review Hybrid Model

[Tan Tanaka, Dennis]

Hi Ariel, et al

On behalf the RySG representatives I’m sharing the stakeholder views on the Hybrid Model.

We look forward to our conversation during next meeting.

Best,
Dennis

----
RySG’s position regarding IDN EPDP Small Group Hybrid Model Recommendation on String Similarity Basis for Comparison for gTLD labels and corresponding variant labels

The RySG agrees that the Small Group’s recommendation is conservative, and that this is a desirable goal, along with the inclusion and stability principles, for DNS labels considered for the root.

The RySG also acknowledges that the Small Team did not consider the operational and procedural implications of the recommended Hybrid Model for the string similarity review process. In this regard, the RySG would like to submit the following observations which may be considered in the IDN EPDP deliberations to finalize the policy recommendation and/or implementation guidelines:

  1.  The procedure to develop the RZ-LGR was not designed to identify visually similar labels as a primary objective; It may capture some cases but not all of them. For this reason, we recommend that this EPDP should be careful to not construe that the variant set produced by the RZ-LGR is a “complete” set of all visually similar labels; furthermore, if the intent is to capture and incorporate possible confusable labels of a gTLD, the string similarity process should not rely solely on the output of the RZ-LGR (for visually confusable labels), but it may use it as an input to determine the basis for comparison.
  2.  In general, a conservative approach to processing candidate gTLDs for the root is a good approach. That said, incorporating ineligible labels (blocked variants) into the string similarity process is unprecedented, so careful consideration should be given to the implementation to achieve consistency and predictability of outcomes.
  3.  As the Small Group noted, there are several reasons to mark a variant as “blocked”. Some blocked variant labels are not well-formed labels e.g., mix-script labels, which do not conform to protocol or generally acceptable guidelines for DNS labels, including IDNs. Therefore, as long as mix-script labels are prohibited from gTLD labels (with the notable exception of Japanese), the RySG suggests that the anticipated procedure should identify and remove mix-script labels from further consideration to avoid unnecessary work.
----

From: Gnso-epdp-idn-team <gnso-epdp-idn-team-bounces at icann.org> on behalf of Ariel Liang <ariel.liang at icann.org>
Date: Friday, October 7, 2022 at 9:48 AM
To: "gnso-epdp-idn-team at icann.org" <gnso-epdp-idn-team at icann.org>
Subject: [EXTERNAL] [Gnso-epdp-idn-team] Reminder: Share Current Views on String Similarity Review Hybrid Model


Caution: This email originated from outside the organization. Do not click links or open attachments unless you recognize the sender and know the content is safe.

Dear All,

In the upcoming meeting on Thursday, 13 October, the team is expected to discuss the String Similarity Review topic and the hybrid model proposed by the small group. During the meeting yesterday, several members provided a preview of their groups’ positions on the hybrid model. To facilitate an effective discussion next week, the leadership team is requesting members, especially those who could not make it to yesterday’s call, to circulate your groups’ current views on the hybrid model on the mailing list by EOB Tuesday, 11 October. Please also share the rationale, caveats, and other considerations (if any).

Thank you,
Steve, Emily, Ariel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mm.icann.org/pipermail/gnso-epdp-idn-team/attachments/20221010/2ee1171a/attachment.html>