[UA-discuss] 答复: The Open Dot as a label delimiter in Chinese and Japanese

Mon Nov 6 23:27:30 UTC 2017

Agreed.  I think we can agree that UASG can/should recommend that linkifiers treat open dot and closed dot as equivalent when attempting to detect label separators.  But linkification itself is inherently hard in many instances and getting adoption of that reco is app/service specific.

From: ua-discuss-bounces at icann.org [mailto:ua-discuss-bounces at icann.org] On Behalf Of Andrew Sullivan
Sent: Thursday, November 2, 2017 8:58 PM
To: Peter Green <seekcommunications at hotmail.com>; Don Hollander <don.hollander at icann.org>
Cc: ua-discuss at icann.org
Subject: Re: [UA-discuss] 答复: The Open Dot as a label delimiter in Chinese and Japanese

Hi,

I understand why people want these things, but it is not possible and will never be possible to treat label separators in quite the way people want. This is because the label separators don't actually appear in the wire protocol, but the presentation format of domain names is also a kind of protocol.

First, upper and lower case _are not_ equivalent in the DNS.  The protocol makes them match, but the case is supposed to be preserved. You can observe this on the Internet.

Second, mapping of actually different names (like the different spellings of China in Han characters) is quite different to the in-protocol match that the DNS unfortunately did for upper and lower case ASCII.

But most importantly, in the wire format there is no separator character. Instead, the length of the label indicates the separation. But in zone files and the like, the separator is there for humans. This is also part of the protocol, so we can't arbitrarily change it.  Applications can map it, however. That is not quite the same but it is what idna does generally.

A

Sent with AquaMail for Android<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fplay.google.com%2Fstore%2Fapps%2Fdetails%3Fid%3Dorg.kman.AquaMail&data=02%7C01%7Cmarksv%40microsoft.com%7C8fddc726734a455c719108d5226f4626%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C636452783767202795&sdata=6SN6raq3UdxMm5IifXGtUxkCyae18femiSuZcuYVl%2Bg%3D&reserved=0>
http://www.aqua-mail.com<https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.aqua-mail.com&data=02%7C01%7Cmarksv%40microsoft.com%7C8fddc726734a455c719108d5226f4626%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C636452783767202795&sdata=jBeBULyA57H4TIcS7ueQlmv7g3n%2FbfXaXGASKE2z9eA%3D&reserved=0>

On November 2, 2017 23:16:44 Peter Green <seekcommunications at hotmail.com<mailto:seekcommunications at hotmail.com>> wrote:

Hi Don,

To add to Jiankang Yao.

Simply put,  what we aim to achieve is "。” should be or even must be equivalent to “.” in domain names, as is the case with Upper case and Lower case in domain names, e.g. "abc" is equivalent to "ABC" when they are used in domain names; as is the case with Simplified Chinese and Traditional Chinese in domain names, e.g. "中国”(China) is equivalent to "中國"  in domian names.

I am not a technical guy.

If anything wrong, please do correct me. @Jiankang Yao<mailto:yaojk at cnnic.cn>

Best Regards

Zuan Zhang

________________________________
发件人: ua-discuss-bounces at icann.org<mailto:ua-discuss-bounces at icann.org> <ua-discuss-bounces at icann.org<mailto:ua-discuss-bounces at icann.org>> 代表 Jiankang Yao <yaojk at cnnic.cn<mailto:yaojk at cnnic.cn>>
发送时间: 2017年11月3日 10:09
收件人: Don Hollander; ua-discuss
主题: Re: [UA-discuss] The Open Dot as a label delimiter in Chinese and Japanese

Hello,

In Chinese version of IE,  when the full stop is inputed into IE, it will be automatically turned into ASCII dot.
In Chinese Input Method which helps to input Chinese Character into computer, when you input Chinese character, the  full stop is immediately
followed if you want to finish a sentence.  The Chinese Input Method can not know whether you want to input Chinese sentence or Chinese domain name. If it is Chinese domain name, it should be ASCII dot. If it is Chinese sentence, it should be the full stop. Usually, Chinese Input Method will choose the full stop for chinese characters. If the user want to  input the ASCII dot, he needs to switch to English Input Method. In order to be convenient to users, CNNIC talked with many browsers to push them to support "Chinsed full stop should be automatically turned into ASCII dot in chinese domain name" in address bar of browser. Now almost all browsers with Chinese version support this function.

The browser with English version may not support this function "Chinsed full stop should be automatically turned into ASCII dot in chinese domain name".

ASCII dot between Chinese character is only useful in Chinse Domain Names. Otherwise, the Chinese full dot should be used.

Best Regards.
________________________________
Jiankang Yao

From: Don Hollander<mailto:don.hollander at icann.org>
Date: 2017-11-03 08:10
To: Universal Acceptance<mailto:ua-discuss at icann.org>
Subject: [UA-discuss] The Open Dot as a label delimiter in Chinese and Japanese
G’day:

The UASG has in the past indicated that good practice is to treat the Open Dot as a label delimiter, just like the traditional full-stop.

The ideographic full stop (U+3002 [。]) is used in languages such as Chinese or Japanese to mark the end of a sentence. UASG004 states “We expect software to transform the ‘open dot’ to a standard ASCII dot “.”, thus making use of the already registered domain name.”
We found that some browsers do this.

As we go through the Linkification review, we’re not seeing this happen for social media communications apps.

Does anyone have reference or even perception to how widely used the Open Dot is in Chinese, Japanese and/or other script?

Don

Don Hollander
Universal Acceptance Steering Group
Skype: don_hollander

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mm.icann.org/pipermail/ua-discuss/attachments/20171106/4985b9f1/attachment.html>