The call for the IDNs EPDP team will take place on Thursday, 17 February 2022 at 13:30 UTC for 90 minutes.

For other places see:


  1. Roll Call & SOI Updates (2 minutes)
  2. Welcome & Chair Updates (5 minutes)
  3. Deliberations on Topic D []Charter questions D1b (80 minutes)
  4. AOB (3 minutes)


EPDP Team Meeting #24 Slides - D1b.pdf




Apologies: Jennifer Chung,  Anil Kumar Jain, Jeff Neuman

Joining late: Maxim Alzoba

Notes/ Action Items

Action Items:


Action Item #1: Staff to double-check transaction fees listed on slide 11 from today’s call.



Welcome & Chair Updates

  • Reminder to provide feedback on draft language for charter questions A5 and A6 by next week Tuesday (22 February). If the groups that members represent are comfortable with the language, please confirm on the mailing list by the deadline.
  • The leadership team met earlier this week to review the project plan. While the EPDP Team is not moving through the charter questions as quickly as anticipated, we are not going to submit project change request (PCR) yet. The group will focus on progress towards key milestones in the project plan, and in particular the Initial Report target date. The EPDP Team will revisit the project plan in three months when the group has a better sense of the timing for delivery of the Initial Report.
  • Reminder: Next week’s the meeting is rescheduled to Friday (25 February) to avoid conflicts with ICANN73 Prep Week sessions.

Deliberations on Topic D []: Charter question D1b

  • Slides 2-4 – breakdown of the three parts of charter question D1b, including summary of discussions to date and additional elements for the EPDP Team to discuss. The focus of today’s discussion will be Part 2: An existing RO seeks to activate allocatable variant label(s) of its existing gTLD.
  • Slide 5 – Existing gTLDs with allocatable variants.
  • Slide 6 – Number of self-identified variants of Arabic and Chinese gTLDs from the 2012 round and the number of allocatable variants among the self-identified ones.
  • Comment: Information about self-identified variants from the 2012 round was used in the objections process as well as in the evaluation of contention sets. If the variants overlapped, they would be used in the contention set. The EPDP Team should keep in mind that these self-declared variants were not completely unused in the 2012 round.
  • Comment: All that happened during the last round is historical data. Many TLDs have changed hands since then, so other information will also need to be taken into account in planning for future processes.
  • Response: We need to marry what has happened in the past with what needs to happen in the future. We will seek to find the best path.
  • Slide 7 – Review of IDN-specific application questions from the 2012 round.
  • Comment: Using RZ-LGR would replace the need to request the applicant the IDN table for the top-level label (Q15).
  • Slide 8-9 – Overview of the application process flow in the 2012 round.
  • For consideration: what elements of the 2012 application process would still apply in cases where an existing RO seeks to activate allocatable variant label(s) of its existing gTLD?
  • Comment: While self-identified variants were taken into account to some extent during the 2012 round, with the RZ-LGR there are more variant possibilities now.
  • Question: which parts of the process did self-identified variants go through in the 2012 round?
  • Response: RZ-LGR did not exist, but the IDN Tables needed to be included in the application. Self-identified variants were likely identified by the applicant drawing the IDN Tables. Self-identified variants were taken into account in string similarity evaluation and objections processes. They were parts of these elements in relation to the IDN application, but the self-identified variants were not evaluated themselves.
  • Comment: While we can consider which potential applicants might apply for variants, we don’t know for sure if they will be interested in doing so. One question to consider is whether they can just go ahead and obtain variants between application rounds. If existing gTLD operators do apply for a variant, it’s important to consider staffing needs to be put into place to conduct the evaluation. One possibility is to make sure there is an opportunity for existing gTLD operators to apply for their variants before the next round opens.
  • Comment: It is important to consider evaluation elements that do not need to be completed repeatedly. There should be a simplified round existing gTLD operators do apply for a variant.
  • For discussion: Based on the information provided, what process should be used for an existing RO to activate allocatable variant label(s) of its existing gTLD? Could this happen on an ad hoc basis or is a more structured opportunity needed? Would a simplified round make sense? What application questions are appropriate?
  • Comment: A simplified round might be seen as a workaround for certain applicants. It will be viewed by some as preferential treatment. From one perspective, SubPro did not support special application rounds for specific types of applications.
  • Comment: The EPDP Team should not use the term “application” in discussing the delegation of variants. It should use the term “activation” of variants because there is some support in the EPDP Team for having single application for a variant set.
  • Comment: The process for requesting and obtaining variants should be ongoing, whether for the previous round or future rounds. Successful applicants should be able to activate allocatable variants at any time. Since the process operates on a cost recovery basis, when we talk about activation of variants, we should consider principles to take into account in the calculation of fees.
  • Comment: The terminology of “apply” vs. “activated” is important and may inform how these labels are delegated in the root zone.
  • Comment: In principle, simple processes are better. ICANN already has processes for evaluating labels in general and IDN labels specifically, including evaluating technical competency for IDN services. We can consider a simple process for existing ROs to obtain variants, with the addition of a few questions regarding management of variants. No one is frontrunning here. The existing ROs are entitled to those variants.
  • Comment: In addition to the technical evaluations, there were other processes from the 2012 round we need to take into account, for example the EPDP Team should consider whether string similarity is still relevant and if so, how to we incorporate it.
  • Comment: In general, activation of variant TLDs should not be only possible in a future round, because this may take a lot of time. However, some applicants will be at a disadvantage if this is the case -- specifically, an applicant for a label that the applicant considers to be a variant that is not a variant according to the RZ-LGR. This applicant would have to wait until the next round to apply. Example: .quebec.
  • Staff Comment: Regarding the framing of deciding between a simple vs. complex process – the core issue is that we need to ensure that the variant is being evaluated to make sure it is being introduced and managed in a safe and secure manner. The EPDP Team needs to make sure that the necessary elements are taken into account in completing the evaluation.  There are two elements of the evaluation processes: those involving the applicant vs. those involving the string itself. The EPDP Team may want to take that into account when considering which elements are applicable to evaluating variants. In the 2012 round, the processes involving the string itself allowed the community/public to provide their input on the string.
  • Comment: If a special process is needed, it might be able to draw on the model of the fast track process for ccTLDs.
  • Data point presented by staff: There is a considerable difference between the number of self-identified variants from the 2012 round and those generated through the RZ-LGR. There were approximately 170 self-identified variants from the 2012 application round excluding Arabic script, whereas the number of variant labels generated through the RZ-LGR is around 1,000 if you exclude those in the Arabic script. For the numbers including the Arabic script, the number of self-identified variants is still around 200. The variant labels generated through the RZ-LGR is around and 20,000 if you include the Arabic script.
  • Next Steps: Develop a strawman proposal of a discrete process for further consideration.
  • Slide 10 – Recap of discussion on fees (part 3 of charter question D1b)
  • Slide 11 – Overview of fees from the 2012 round
  • Comment: Applicants from the 2012 round have already paid the application fee of $185.000. This should potentially be considered in the activation of variants. Would applying for a variants set include additional costs, or should it be considered that the applicant has already paid their fees in 2012? The revenue neutral model should be taken into account.
  • Comment: In the TMCH, each string is treated separately with an associated fee. We should look at this model as we consider the question.


Action Item: Staff to double-check transaction fees listed on slide 11 from today’s call.



  • For further discussion: Consider whether the meeting should be 2 hours every other week.


  • No labels