[ssac-gnso-irdwg] Fw: [comments needed] IRD Whois bundled output

[Steve Sheng <steve.sheng@xxxxxxxxx>]

FYI.

Steve

________________________________
From: Yao Jiankang <yaojk@xxxxxxxx>
To: Steve Sheng; edmon@xxxxxxxxxxxxx <edmon@xxxxxxxxxxxxx>
Cc: Julie Hedlund; Ram Mohan <rmohan@xxxxxxxxxxxx>
Sent: Tue Feb 02 18:03:02 2010
Subject: Re: [comments needed] IRD Whois bundled output

pls see in line.
----- Original Message -----
From: Steve Sheng<mailto:steve.sheng@xxxxxxxxx>
To: edmon@xxxxxxxxxxxxx<mailto:edmon@xxxxxxxxxxxxx> ; 
yaojk@xxxxxxxx<mailto:yaojk@xxxxxxxx>
Cc: Julie Hedlund<mailto:julie.hedlund@xxxxxxxxx> ; Ram 
Mohan<mailto:rmohan@xxxxxxxxxxxx>
Sent: Wednesday, February 03, 2010 12:35 AM
Subject: [comments needed] IRD Whois bundled output

Hi Edmon and Jiankang,

  As a follow up to the call on Monday, Ram asked the staff to consult with you 
on a question about bundled output in IDN whois for Chinese registries.

  Here is my understanding of the problem. When a use submit an A-label or 
U-label query, the WHOIS server might want to return bundled outputs. For 
example, a U-label query of 中国银行.CN (Bank of China) might return both the WHOIS 
information for 中国银行.CN  (simplified) and also 中國銀行.CN (traditional). Chinese 
is the most obvious example, but there are other examples such as Indian 
languages as well.

  We’d love to hear your thoughts on the following two questions:
it is my pleasure.


  1) Is this the current practice in .CN or .ASIA?

yes, it is true for .cn.


  2) do you think this is a good idea? Why or why not?

yes.  based on RFC3743, when a registrant registers a Chinese domain name, the 
same registrant will get all the bundled names.
Those bundled names are all in one package.  when one of the domain names in 
the package is transferred or updated or deleted, the whole package including 
all the names will be transfferred or updated or deleted.
Most CDNC members(www.cdnc.org<http://www.cdnc.org>) follow the practise in 
RFC3743.



Thanks,
Steve



On 2/2/10 10:49 AM, "Julie Hedlund" <julie.hedlund@xxxxxxxxx> wrote:

Dear IRD-WG members,

Below are the action items and main discussion points from the 01 February 2010 
meeting of the IRD-WG.  These also are on the wiki at: 
https://st.icann.org/int-reg-data-wg/index.cgi?internationalized_registration_data_working_group
 
<https://st.icann.org/int-reg-data-wg/index.cgi?internationalized_registration_data_working_group>
 .   Please let me know if you have any changes or questions.  Our next meeting 
is scheduled for Monday, 15 February at 1400 UTC, 06:00 PST, 09:00 EST, 14:00 
London, 15:00 CET, 22:00 Beijing; 16 February: 03:00 New Zealand. (Note that 
meetings will alternate between 1900 and 1400 UTC to accommodate various time 
zones.)

Best regards,

Julie

Julie Hedlund, Director, SSAC Support

Attendees: Jay Daley, Robert Hutchinson, Andrei Kolesnikov, Mark Kostas, Steve 
Metalitz, and Ram Mohan; From staff: Francisco Arias, Glen de Saint-Gery, 
Gisella Gruber-White, Julie Hedlund, and Steve Sheng

Action Items:

 1.  WG members will continue to provide comments on the list to the draft of a 
questions related to topic #1 -- “What do we require from internationalized 
registration data?”  See the email archive for the latest discussion at: 
http://forum.icann.org/lists/ssac-gnso-irdwg/ 
<http://forum.icann.org/lists/ssac-gnso-irdwg/> .  See also the attached 
summary provided by Steve Sheng of comments received by 25 January, as well as 
the discussion summary below.
 2.  Steve Sheng will follow up with Jiankang and Edmon concerning Question 1a 
and the recommendation to provide output in A-label and U-label format and how 
this would work for Chinese registries.

Main Discussion Points: WG members discussed questions 1a and 1b, continuing 
from comments provided on the email list and summarized by Steve Sheng.
-- Question 1a: 1) What do we require from internationalized registration data: 
that a user can submit or have a domain name displayed in the IDN A-label 
(xn--) format or U-label (local language readable) format?
WG members agreed that 1)  WHOIS must accept a "submit" in either U- or 
A-label; and 2) WHOIS must "display" both in U- and A-label. There is also an 
additional recommendation raised by Ram, which would be optional, that bundled 
representations (e.g. both the simplified and traditional Chinese) of a single 
A or U-label query should be returned. The WG members asked Steve Sheng to 
follow up with Jiankang and Edmon concerning their experience with Chinese 
registries and thoughts on this recommendation. In earlier emails, Jiankang 
mentioned that A-label input to Whois needs to be checked to confirm that it is 
a valid IDN. If it is not, error should be returned. This would require changes 
to the Whois client.

-- Question 1b: that registration data be extensible to accommodate users who 
would benefit from the ability to submit and have registration information 
displayed in "familiar" characters from local languages and scripts?   Much of 
the discussion focused on Jay’s suggestion that the various elements of 
registration data could separately internationalized.  WG members agreed that 
the sponsoring registrar name should be displayed in US-ASCII7 subset of the 
Latin-1 character set. (Note: although some working group members do not fully 
endorse this idea, they nevertheless think this is acceptable).  Ram and Andrei 
noted that in India and Russia, respectively, the registrar representation is 
provided in both local script and ASCII.  The WG members also agreed that email 
addresses can be internationalized using RFC 4952 and 5336. The WG members 
discussed using  UPU convention for postal address internationalization.  Bob 
noted that the UPU standard is not computer based and that it is difficult to 
write a web page to capture manual addresses.  Jay added that UPU is a standard 
that already exists and that the WG should not try to develop a separate 
standard. This needs to be taken up again in the next meeting.  The WG 
discussed using E123 format to internationalize phone numbers, Ram noted that 
currently there is no standard for phone numbers in Whois. This issue should be 
followed up in the next IRD meeting.  The rest of the discussion focused on the 
question of whether or not contact information (include registrant, admin 
contact name, tech contact name) should at least be displayed in ASCII.  WG 
members presented reasons both for and against:

Reasons for at least have ASCII contact information:
- Avri noted that it is important for registrant information to be accessible 
to others.
- Ram noted that law enforcement often requires both local and ASCII output and 
Andrei agreed that was also the case in Russia.

Reasons against:
- Jay noted that insisting all data is in both the local language and ASCII 
will not make it universally usable. It will present a barrier for those who 
does not use English to register, and would introduce errors.
- Jay added that a solution would be to offer the information in as many 
scripts as possible, but that this wasn’t likely to be feasible.

WG members also discussed whether their could be a standard for ccTLDs.  Jay 
noted that ccTLDs handle output very differently and that it isn’t in the scope 
of the IRD-WG’s work to recommend a standard for ccTLDs, although it might be 
useful to provide a tool kit for each data element.  Ram suggested that the WG 
should hold this topic for later.  Andrei noted that it might be helpful to do 
a survey among non-ASCII ccTLDs on this issue.  Ram suggested that the WG 
should focus on recommendations on output in a local script and some type of 
standard set, but should not address the issue of confusingly similar character 
sets.