[UA-discuss] [EXTERNAL] Re: Ideas for engagement with stakeholders in Germany

Mon Nov 25 19:02:54 UTC 2019

On 11/25/2019 4:22 AM, Maxim Alzoba wrote:
> Hello Mark,
>
> the topic is quite complex one (from the operational and legal 
> perspective),
> we should not invent rights not existing in the real world (and we 
> should not forget abut rights of the Registrants).
>
> Potentially some variants of the same word might be in the hands of 
> different rightful registrants, and who is going to decide
> which one wins?

With variants you have the following issues:

(1) if they are imposed after the fact, you'll need to have some policy 
for grandfathering

That's why, for the root zone, we categorically ruled out, for example, 
any variants (even implied by transitivity) between ASCII code points. 
And we test all proposed LGRs against delegated TLDs.

(2) if they are allocatable, you have issues with multiplicity.

A longer label may have several code points with variants, which naively 
would lead to huge multiplicities. That's why, in the root zone, we go 
to some lengths defining allocatable (code point) variants so that 
allocatable variant labels are limited to a very small number; usually 
by requiring all variants to be of the same type (e.g. all 
traditional/all simplified for Chinese).

(3) if they are allocatable, you have issues with predictability.

Not all applicants want all possible allocatable variants actually 
delegated. That's why, in the root zone, we've tried to limit 
allocatable variants to contexts where both users and applicants clearly 
understand their scope (e.g. Chinese simplified and traditional).

(4) if they are blocked variants, they need to be full equivalents, not 
merely similar

Otherwise, transitivity will kill you. Any variant of a code point needs 
to also be a variant of all the other variants for that code point 
(transitivity). If variants are merely similar, you quickly get 
"variants" between dissimilar code points. That's why, in the root zone, 
we are limiting variants to cases where code points are accepted as 
fully equivalent.

(5) if you base variant definition on a limited sample of languages, 
you'll be surprised

What works in a limited context may create issues if you plan on 
allowing labels in other scripts or other languages in the same zone. 
That's why, in the root zone, we investigated over 200 writing systems 
for the Latin script, for example.

(6) if you don't define certain allocatable variants, you'll run into 
usability issues

Typically the case where users of different regions share a script 
(and/or language) but regional conventions conflict. For example the 
case in Arabic, where local keyboards may have a look-alike character, 
so if you can't offer both versions of your label, you cannot reach some 
communities. That's why, in the root zone, we have allocatable variant 
definitions for Arabic, Chinese, and for the case of sharp-s/ss, for 
example.

(7) if you don't have blocked variants, you can't take obvious issues 
"off the table"

There are a surprising number of letters (or for the second level, also 
digits) that have exact (intentional) look-alikes in Unicode. (They are 
considered different for other reasons than appearance, for example they 
belong to different scripts, have different case pairings, etc). These 
cases are quite obvious equivalences and can, and should be taken off 
the table, up front and mechanically. That's why, in the root zone, we 
ensure that such cases become blocked variants.

Now, why were we discussing this again?

A./

>
> Sincerely Yours,
>
> Maxim Alzoba
> Special projects manager,
> International Relations Department,
> FAITID
>
> Current UTC offset: +3.00 (.Moscow)
>
>> On 25 Nov 2019, at 11:05, Mark Svancarek (CELA) via UA-discuss 
>> <ua-discuss at icann.org <mailto:ua-discuss at icann.org>> wrote:
>>
>> Rubens, what is your opinion regarding variants policy? Should 
>> registrants be obliged to register all variants at the 2^nd level to 
>> avoid confusion?
>> *From:*UA-discuss <ua-discuss-bounces at icann.org 
>> <mailto:ua-discuss-bounces at icann.org>>*On Behalf Of*Mark W. Datysgeld
>> *Sent:*Sunday, November 24, 2019 1:13 PM
>> *To:*ua-discuss at icann.org <mailto:ua-discuss at icann.org>; Rubens Kuhl 
>> <rubensk at nic.br <mailto:rubensk at nic.br>>; ua-discuss 
>> <ua-discuss at icann.org <mailto:ua-discuss at icann.org>>
>> *Subject:*[EXTERNAL] Re: [UA-discuss] Ideas for engagement with 
>> stakeholders in Germany
>>
>> I see Rubens might have a vested interest in this one. :D
>> --
>> Mark W. Datysgeld from Governance Primer [www.markwd.website 
>> <http://www.markwd.website/>]
>> In partnership with AR-TARC and the Brazilian Association of Software 
>> Companies (ABES)
>>
>> On November 24, 2019 5:19:51 AM GMT+01:00, Rubens Kuhl 
>> <rubensk at nic.br <mailto:rubensk at nic.br>> wrote:
>>
>>
>>
>>         On 24 Nov 2019, at 00:22, Mark W. Datysgeld
>>         <mark at governanceprimer.com
>>         <mailto:mark at governanceprimer.com>> wrote:
>>         Hello everyone,
>>         I exchanged a few messages with Lars earlier this year, and
>>         would like to present below some points he raised that should
>>         be useful both for the IGF next week and looking towards
>>         ICANN meeting C next year, being tailored towards German
>>         stakeholders. In my view, we can aim to always try gathering
>>         such regional insights moving forward to optimize our engagement.
>>         *Special characters:*Eszett (ß) and Umlaut (ü) have fallback
>>         options by default:
>>
>>           * Eszett becomes "ss".
>>           * Capital Eszett (ẞ) has only recently been introduced
>>             officially to the language, meaning that anything in
>>             capital letters used to be written using "SS" even on the
>>             press.
>>           * Umlaut characters (ä, ö, ü) become "ae", "oe", "ue".
>>
>>     Long live the Umlaut!
>>     Rubens Kühl
>>
>>
>> _______________________________________________
>> By submitting your personal data, you consent to the processing of 
>> your personal data for purposes of subscribing to this mailing list 
>> accordance with the ICANN Privacy Policy 
>> (https://www.icann.org/privacy/policy) and the website Terms of 
>> Service (https://www.icann.org/privacy/tos). You can visit the 
>> Mailman link above to change your membership status or configuration, 
>> including unsubscribing, setting digest-style delivery or disabling 
>> delivery altogether (e.g., for a vacation), and so on.
>
>
> _______________________________________________
> By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mm.icann.org/pipermail/ua-discuss/attachments/20191125/916545e3/attachment.html>