Checking for Mixed Scripts in a domain name or email address - is this a UASG issue? - Group Discussion
We’ve had the following suggested for the Programming Language Criteria: Could we include one more test case in "Programming Language Evaluation Criteria ". 1. checking of multiple script in Email,Domain,Url. Ex: еріс.com[xn--e1awd7f.com] contain two language script (Cyrillic ,Latin ). deepak.भारत contain two language script (Latin, Devanagari) I understand the thinking behind this, but I’m not sure that a) it’s in our remit, b) it’s a good idea c) it will be rejecting perfectly valid domain names (CJK, Cyrillic+Latin Numerals, etc) I also don’t know what standard this would reference or what policy it would reference. Something from the Unicode group or M3WAAG? Your thoughts, please. Don
Hello Don, usually IDN cyrillic contains digits, so one of the examples is not correct (Cyrillic+Latin Numerals is not a mixed script). https://www.iana.org/domains/idn-tables/tables/xn--80adxhks_ru_1.0.txt <https://www.iana.org/domains/idn-tables/tables/xn--80adxhks_ru_1.0.txt> usually issues arise with the string containing chars , which do not fall into the same IDN table, for example something from IDN mistakes - tчху xn--t-cubfh from https://www.icann.org/sites/default/files/packages/reserved-names/ReservedNa... <https://www.icann.org/sites/default/files/packages/reserved-names/ReservedNa...> Or, which is important, IDN string from the table not allowed for the particular TLD for example moscow.москва or москва.moscow (both TLDs are strictly one script, only Cyrillic Russian in the first, and only allowed ASCII symbols in the second) P.s: formally each TLD has IDN policy or "no-IDN policy" and it describes allowed combinations, and what is important it might change over the time (for example some old TLDs decided to allow some of IDN tables). Sincerely Yours, Maxim Alzoba Special projects manager, International Relations Department, FAITID m. +7 916 6761580(+whatsapp) skype oldfrogger Current UTC offset: +3.00 (.Moscow)
On Aug 7, 2017, at 21:52, Don Hollander <don.hollander@icann.org> wrote:
We’ve had the following suggested for the Programming Language Criteria:
Could we include one more test case in "Programming Language Evaluation Criteria ".
1. checking of multiple script in Email,Domain,Url. Ex: еріс.com[xn--e1awd7f.com] <https://urldefense.proofpoint.com/v2/url?u=http-3A__xn-2D-2De1awd7f.com&d=Dw...> contain two language script (Cyrillic ,Latin ). deepak.भारत contain two language script (Latin, Devanagari)
I understand the thinking behind this, but I’m not sure that a) it’s in our remit, b) it’s a good idea c) it will be rejecting perfectly valid domain names (CJK, Cyrillic+Latin Numerals, etc)
I also don’t know what standard this would reference or what policy it would reference. Something from the Unicode group or M3WAAG?
Your thoughts, please.
Don
I believe you could use Unicode definitions of scripts as, in that system, numbers are NOT Latin. (They are ANSI and ASCII but Unicode clearly separates them and punctuation.) That said, the other concerns (about whether this is within remit and whether it is a good practice remain. From: ua-discuss-bounces@icann.org [mailto:ua-discuss-bounces@icann.org] On Behalf Of Maxim Alzoba Sent: Monday, August 7, 2017 12:47 PM To: Don Hollander <don.hollander@icann.org> Cc: ua-discuss@icann.org Subject: Re: [UA-discuss] Checking for Mixed Scripts in a domain name or email address - is this a UASG issue? - Group Discussion Hello Don, usually IDN cyrillic contains digits, so one of the examples is not correct (Cyrillic+Latin Numerals is not a mixed script). https://www.iana.org/domains/idn-tables/tables/xn--80adxhks_ru_1.0.txt<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.iana.org%2Fdomains%2Fidn-tables%2Ftables%2Fxn--80adxhks_ru_1.0.txt&data=02%7C01%7Cstuartst%40exchange.microsoft.com%7C782eb85495ba4d81404408d4ddcc9f27%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C636377318366514294&sdata=d4hXA4lp8vKiNeQIN6VHRJF2RGsk7m3PkBkmNapg3ew%3D&reserved=0> usually issues arise with the string containing chars , which do not fall into the same IDN table, for example something from IDN mistakes - tчху xn--t-cubfh from https://www.icann.org/sites/default/files/packages/reserved-names/ReservedNames.xml<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.icann.org%2Fsites%2Fdefault%2Ffiles%2Fpackages%2Freserved-names%2FReservedNames.xml&data=02%7C01%7Cstuartst%40exchange.microsoft.com%7C782eb85495ba4d81404408d4ddcc9f27%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C636377318366514294&sdata=iL4gpIGnwTnxvCAUFcH0791zWpk7lZdYQqAuBBX%2F3XM%3D&reserved=0> Or, which is important, IDN string from the table not allowed for the particular TLD for example moscow.москва or москва.moscow (both TLDs are strictly one script, only Cyrillic Russian in the first, and only allowed ASCII symbols in the second) P.s: formally each TLD has IDN policy or "no-IDN policy" and it describes allowed combinations, and what is important it might change over the time (for example some old TLDs decided to allow some of IDN tables). Sincerely Yours, Maxim Alzoba Special projects manager, International Relations Department, FAITID m. +7 916 6761580(+whatsapp) skype oldfrogger Current UTC offset: +3.00 (.Moscow) On Aug 7, 2017, at 21:52, Don Hollander <don.hollander@icann.org<mailto:don.hollander@icann.org>> wrote: We’ve had the following suggested for the Programming Language Criteria: Could we include one more test case in "Programming Language Evaluation Criteria ". 1. checking of multiple script in Email,Domain,Url. Ex: еріс.com[xn--e1awd7f.com]<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Furldefense....> contain two language script (Cyrillic ,Latin ). deepak.भारत contain two language script (Latin, Devanagari) I understand the thinking behind this, but I’m not sure that a) it’s in our remit, b) it’s a good idea c) it will be rejecting perfectly valid domain names (CJK, Cyrillic+Latin Numerals, etc) I also don’t know what standard this would reference or what policy it would reference. Something from the Unicode group or M3WAAG? Your thoughts, please. Don
The ASCII digits have script "common", not "Latin". The issue isn't numbers, it's mixing letters. Also, mixing scripts within a label seems somehow different from mixing scripts within a FQDN. A./ On 8/7/2017 12:46 PM, Maxim Alzoba wrote:
Hello Don,
usually IDN cyrillic contains digits, so one of the examples is not correct (Cyrillic+Latin Numerals is not a mixed script).
https://www.iana.org/domains/idn-tables/tables/xn--80adxhks_ru_1.0.txt
usually issues arise with the string containing chars , which do not fall into the same IDN table, for example something from IDN mistakes - tчху xn--t-cubfh
from https://www.icann.org/sites/default/files/packages/reserved-names/ReservedNa...
Or, which is important, IDN string from the table not allowed for the particular TLD
for example moscow.москва or москва.moscow (both TLDs are strictly one script, only Cyrillic Russian in the first, and only allowed ASCII symbols in the second)
P.s: formally each TLD has IDN policy or "no-IDN policy" and it describes allowed combinations, and what is important it might change over the time (for example some old TLDs decided to allow some of IDN tables).
Sincerely Yours,
Maxim Alzoba Special projects manager, International Relations Department, FAITID
m. +7 916 6761580(+whatsapp) skype oldfrogger
Current UTC offset: +3.00 (.Moscow)
On Aug 7, 2017, at 21:52, Don Hollander <don.hollander@icann.org <mailto:don.hollander@icann.org>> wrote:
We’ve had the following suggested for the Programming Language Criteria: Could we include one more test case in "Programming Language Evaluation Criteria ".
1. checking of multiple script in Email,Domain,Url. Ex: еріс.com[xn--e1awd7f.com] <https://urldefense.proofpoint.com/v2/url?u=http-3A__xn-2D-2De1awd7f.com&d=Dw...> contain two language script (Cyrillic ,Latin ). deepak.भारत contain two language script (Latin, Devanagari) I understand the thinking behind this, but I’m not sure that a) it’s in our remit, b) it’s a good idea c) it will be rejecting perfectly valid domain names (CJK, Cyrillic+Latin Numerals, etc) I also don’t know what standard this would reference or what policy it would reference. Something from the Unicode group or M3WAAG? Your thoughts, please. Don
My gut level response is that mixing scripts within a label is much worse than mixing scripts in FQDN, though I am hard-pressed to quantify that feeling. One could use confusable code points from Latin, Greek and/or Cyrillic, with only a single script in each label but a mixture overall within the FQDN. Is that really less dangerous? Hmm… From: ua-discuss-bounces@icann.org [mailto:ua-discuss-bounces@icann.org] On Behalf Of Asmus Freytag Sent: Monday, August 7, 2017 3:36 PM To: ua-discuss@icann.org Subject: Re: [UA-discuss] Checking for Mixed Scripts in a domain name oremail address - is this a UASG issue? - Group Discussion The ASCII digits have script "common", not "Latin". The issue isn't numbers, it's mixing letters. Also, mixing scripts within a label seems somehow different from mixing scripts within a FQDN. A./ On 8/7/2017 12:46 PM, Maxim Alzoba wrote: Hello Don, usually IDN cyrillic contains digits, so one of the examples is not correct (Cyrillic+Latin Numerals is not a mixed script). https://www.iana.org/domains/idn-tables/tables/xn--80adxhks_ru_1.0.txt<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.iana.org%2Fdomains%2Fidn-tables%2Ftables%2Fxn--80adxhks_ru_1.0.txt&data=02%7C01%7Cmarksv%40microsoft.com%7C7b44e9571246424d486f08d4dde4ac24%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C636377421751192468&sdata=YkDXN3bX4ThceauSipnSvZs1pUV5DBXjMlXP1PBDsRQ%3D&reserved=0> usually issues arise with the string containing chars , which do not fall into the same IDN table, for example something from IDN mistakes - tчху xn--t-cubfh from https://www.icann.org/sites/default/files/packages/reserved-names/ReservedNames.xml<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.icann.org%2Fsites%2Fdefault%2Ffiles%2Fpackages%2Freserved-names%2FReservedNames.xml&data=02%7C01%7Cmarksv%40microsoft.com%7C7b44e9571246424d486f08d4dde4ac24%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C636377421751192468&sdata=gT%2FLd4t3vy0SHpUjs%2Bzqa16T2SkgRRgOOCksPU6kQUg%3D&reserved=0> Or, which is important, IDN string from the table not allowed for the particular TLD for example moscow.москва or москва.moscow (both TLDs are strictly one script, only Cyrillic Russian in the first, and only allowed ASCII symbols in the second) P.s: formally each TLD has IDN policy or "no-IDN policy" and it describes allowed combinations, and what is important it might change over the time (for example some old TLDs decided to allow some of IDN tables). Sincerely Yours, Maxim Alzoba Special projects manager, International Relations Department, FAITID m. +7 916 6761580(+whatsapp) skype oldfrogger Current UTC offset: +3.00 (.Moscow) On Aug 7, 2017, at 21:52, Don Hollander <don.hollander@icann.org<mailto:don.hollander@icann.org>> wrote: We’ve had the following suggested for the Programming Language Criteria: Could we include one more test case in "Programming Language Evaluation Criteria ". 1. checking of multiple script in Email,Domain,Url. Ex: еріс.com[xn--e1awd7f.com]<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Furldefense....> contain two language script (Cyrillic ,Latin ). deepak.भारत contain two language script (Latin, Devanagari) I understand the thinking behind this, but I’m not sure that a) it’s in our remit, b) it’s a good idea c) it will be rejecting perfectly valid domain names (CJK, Cyrillic+Latin Numerals, etc) I also don’t know what standard this would reference or what policy it would reference. Something from the Unicode group or M3WAAG? Your thoughts, please. Don
participants (5)
-
Asmus Freytag -
Don Hollander -
Mark Svancarek -
Maxim Alzoba -
Stuart Stuple