Thanks both of you for the explanations. But I am still concerned that we are making this more difficult than it would need to be. We seem to assume that Registries have the option, worse, the right to f… sorry, to make a whole mess when it is not needed at all, for no good reason. More on that before next call. Amadeu
Missatge enviat per Michael Bauland via Gnso-latin-diacritics <gnso-latin-diacritics@icann.org> el dia 22 maig 2026 a les 9:00:
Hi Amadeu,
as Saewon wrote, the question about handling of diacritics at the second level is indeed pictured in Figure 5.
I'll describe a bit more, what is displayed in Figure 3.
It tries to explain in more detail how the same entity principle and variant relationship works in our cases. In IDN EPDP, all domains are variants of each other and the variant set covers all TLDs. So a domain under one TLD will be a variant under all other (variant) TLDs. Therefore the same entity requirement is easy there: just require that all domains in the variant set need to belong to the same entity.
This does not work in our case. We have no variant relationships crossing the TLD boundary (because the TLDs are not variants of each other). This means instead of having a single variant sent cover all domains in all TLDs, we will need to work with multiple variant sets. Even though we are not working variants at the top level, we still have variant relationships at the second level (defined by the registry's second level LGRs).
The graphic then shows different scenarios, depending on which code points are available in the different TLDs. For simplicity, we only work with one ASCII and one LD TLD (it would work analogously with more than one LD TLD): Scenario 1: Both TLDs have the same LGR, i.e., allowing the same repertoire and having the same variant relationships. This is easy, we just have one variant set per TLD.
Scenario 2: ASCII domain has the following variant set: example.atld exampleV1.atld exampleV2.atld exampleV3.atld
For the LD TLD, the label exampleV2 is not in the repertoire, and therefore not allowed. But instead exampleV4 exists and is a variant of exmapleV3. Then the same entity set contains example.ldtld (because it's the same label) exampleV1.ldtld (because it's the same label) exampleV3.ldtld (because it's the same label) and also exampleV4.ldtld (it's not the same label as in the ASCII set, but it's a variant of one of the labels in the LD TLD).
Scenario 3: This is also similar, but the primary domain name from the ASCII TLD's variant set (example.atld) is not part of the repertoire of the LD TLD. But still all labels that exist in the ASCII TLD's variant set and are available in the LD TLD, must belong to the same entity set, plus all variants of those labels.
So, the graphic is describing Rec 39 with some examples, how it works and why it's necessary.
I hope this helps a bit with understanding this. I know it's a complicated topic and without being able to see the actual examples in the graphic, it makes it even more difficult.
Let me know, if there is something unclear and I'll be happy to explain a bit more.
Cheers,
Michael
-- ____________________________________________________________________ | | | knipp | Knipp Medien und Kommunikation GmbH ------- Technologiepark Martin-Schmeisser-Weg 9 44227 Dortmund Germany
Dipl.-Informatiker Fon: +49 231 9703-0 Fax: +49 231 9703-200 Dr. Michael Bauland SIP: Michael.Bauland@knipp.de Software Development E-mail: Michael.Bauland@knipp.de
Register Court: Amtsgericht Dortmund, HRB 13728
Chief Executive Officers: Dietmar Knipp, Elmar Knipp
Certified according DIN ISO/IEC 27001:2017 _______________________________________________ Gnso-latin-diacritics mailing list -- gnso-latin-diacritics@icann.org To unsubscribe send an email to gnso-latin-diacritics-leave@icann.org