Donna, Thank you. I like the approach. Best, Dennis From: "Donna@registry.godaddy" <Donna@registry.godaddy> Date: Tuesday, July 26, 2022 at 3:17 PM To: Dennis Tan Tanaka <dtantanaka@verisign.com>, Maxim Alzoba <m.alzoba@gmail.com>, "Michael.Bauland@knipp.de" <Michael.Bauland@knipp.de> Cc: "gnso-epdp-idn-team@icann.org" <gnso-epdp-idn-team@icann.org> Subject: [EXTERNAL] RE: [Gnso-epdp-idn-team] primary labels and consequences of retiring them 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. Hi Dennis, all Thank you for raising the question about what we mean when we use the term ‘primary’ and the potential consequences associated with such a designation. Michael has also provided some useful input on this point. As our discussions have evolved we have referred to the ‘primary’ to make a distinction between the IDN gTLD being applied for and the associated variant labels that are either allocatable or blocked that comprise an IDN variant ‘set’. However, our discussions can be complicated, particularly as we all have different levels of understanding of IDNs, and we use words that some might consider technically incorrect, but I believe have become commonly understood terms or phrases within our group, to help distinguish a point that someone/we’re trying to address or make. I agree that the Team does need to discuss what is the most accurate terminology that we should use to describe what we intend with our recommendations. The leadership team is aware that this is a concern and have discussed developing a glossary, which I think will be helpful. In addition, we also recognize that the draft text that we have circulated to date is not consistent in the use of terms or phrases and this is something we are also looking to address. So, I want to acknowledge your concerns, but for now I would like to park these for later discussion/consideration by the Team to give the Leadership Team some time to look into how best to address the issue. Thanks Donna From: Gnso-epdp-idn-team <gnso-epdp-idn-team-bounces@icann.org> On Behalf Of Tan Tanaka, Dennis via Gnso-epdp-idn-team Sent: Tuesday, July 26, 2022 8:34 AM To: m.alzoba@gmail.com; Michael.Bauland@knipp.de Cc: gnso-epdp-idn-team@icann.org Subject: Re: [Gnso-epdp-idn-team] primary labels and consequences of retiring them Caution: This email is from an external sender. Please do not click links or open attachments unless you recognize the sender and know the content is safe. Forward suspicious emails to isitbad@. And this is the question I post to this group. Do we need to always have a label designated as “primary”, and if so, what function will it serve? My current thinking is that the primary label serves its purpose to determine the set of ALL variant labels — allocatable and blocked. Beyond this point the set is established (disposition values are captured), and so, I would argue, a “primary” label no longer have utility (other than a data point to indicate how the set was formed, and/or potential re-calculation due to rz-lgr changes). Each applied-for/delegated label should have equal standing in the set, IMO. But perhaps I am missing something. Dennis From: Gnso-epdp-idn-team <gnso-epdp-idn-team-bounces@icann.org<mailto:gnso-epdp-idn-team-bounces@icann.org>> on behalf of Maxim Alzoba <m.alzoba@gmail.com<mailto:m.alzoba@gmail.com>> Date: Tuesday, July 26, 2022 at 3:54 AM To: Michael Bauland <Michael.Bauland@knipp.de<mailto:Michael.Bauland@knipp.de>> Cc: "gnso-epdp-idn-team@icann.org<mailto:gnso-epdp-idn-team@icann.org>" <gnso-epdp-idn-team@icann.org<mailto:gnso-epdp-idn-team@icann.org>> Subject: [EXTERNAL] Re: [Gnso-epdp-idn-team] primary labels and consequences of retiring them 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. I think the procedure of the retirement needs to have a check for the existence of the primary and if not - election of the new one. Maxim Alzoba On Jul 26, 2022, at 09:42, Michael Bauland <michael.bauland@knipp.de<mailto:michael.bauland@knipp.de>> wrote: Hi Dennis, On 25.07.2022 21:46, Tan Tanaka, Dennis via Gnso-epdp-idn-team wrote: A few additional comments on 2.3 re: use of “primary label” * Some questions I reckon we will face throughout: what is the “primary” label? what function does it serve? should we use such nomenclature? … some background notes: * When we talk about variants in a set, it is important to note which of all labels is the source for variant label calculation, or as we have adopted: the primary label. As a way of context, first, the variant relationship between two code points must satisfy the symmetrical property (there are other properties, but not relevant to this topic). This means, if ‘A’ is a variant of ‘B’, then ‘B’ is a variant of ‘A’. Second, any variant relationship is also assigned a disposition value: allocatable or blocked. The disposition value is unidirectional. Which means, a variant relationship could yield an ‘allocatable’ variant in one direction (e.g., AàB: allocatable), but a ‘blocked’ variant in the other direction (e.g., BàA: blocked). These rules are encoded in the LGR. * As we progress in our deliberations about the lifecycle of gTLDs it is important to understand (and remember) how the various variant labels and disposition values came to be (to discern which ones are allocatable/withheld or blocked). However, I believe we need to remain open (or at least give thoughtful consideration) to cases where one TLD label in the set needs to be handled separately e.g., retirement, and the implications of this one change applies to the rest of the labels in the set. For instance, what if the registry operator wants to retire the “primary” label and continue to operate the rest of variants. In this case, it would be unwise, perhaps, to re-assign the attribute of “primary” to another label in the set, because that might change the profile of the set as a whole (of course, this would depend what “primary” means and how it is used). This is a very good point. Especially with asymmetric disposition we have a real problem if a primary label is retired and the other variants are kept. There might arise a situation in which none of the remaining variants can be made primary, because it would entail one of the other existing variants to become blocked instead of allocatable. The question would then be * Do we allow a variant set to exist without a primary label? * Will retiring a primary label require the registry to select another primary label and thereby possibly force them to retire other variants together with the former primary label? Take the following quite simple example. Applied for primary label: bıß Allocatable Variants: biß, bıss, biss So, someone might have all four labels in the Root zone at some point. If they now retire the primary label bıß, the other three labels could not exist at the time (without bıß also existing). Case 1: biß is made new primary Consequence: bıss becomes a blocked variant. Case 2: bıss is made new primary Consequence: biß becomes a blocked variant. Case 3: biss is made a new primary Consequence: both bıss and biß become blocked variants. Even though the example is not very likely, it's certainly possible and in other scripts might be similar examples that are more likely. So, to summarise my points: * Initially, we need a primary label to determine the disposition value of all variants. * Retiring a primary label may cause problems and we need to decide how to deal with these problems. Best regards, Michael