2 Why are we doing this - Is this particular feature necessary

Here is a summary of the main points raised on this question from the meeting on 09 January 2014:

The PDP WG noted that this is a key question to be considered along with the issue of costs.  In order to consider why to provide translation and/or transliteration of contact information the PDP WG could consider the following:

• The users of WHOIS
• Customer trust issues (part of the ICANN AoC)
• Legal and marketing needs for clear information – allowing a better answer in problematic cases; PDP WG could consider an approach to have the local script be the default script and the translation/transliteration would then be into Latin script/ASCII.

Volunteers to work on the legal aspect are:  Vinay Kumar Singh and Petter Rindforth

Benefits

In Linda Corugedo Steneberg's e-mail, 10 Feb. 2014, the following benefits were suggested:

What exactly the benefits to the community are of translating and/or transliterating contact information, especially in light of the costs that may be connected to translation and/or transliteration? // it facilitates registration for those Registrants who do not speak a common language or use a common script, while at the same time it makes easier the consultation of such data by parties (like Law Enforcement) who require a common language or who may face difficulties while dealing with non-Latin scripts. Registrants would have full rights when it comes to respect for multilingualism and Registrant data can be consulted/searched in a more homogeneous manner thanks to translation to a common language or script.

What are the relationships between the benefits?

Are there trade-offs?

Scenarios

See the purposes for data on p.12-14 of the Initial report from the EWG on gTLD Directory Services.

Features

6 March, 2014

Dear Chris,

Dear all,

When looking into the document Status update EWG, only page 53 and 80 mention the word „translation”. In annex A (page 61) I don’t see any indication of fields allowing the storage of translated or transliterated contact information. Perhaps this is something we should emphasize a bit more in our discussions. If none is preparing a protocol / storage for translated/transliterated information our whole process does not have any effect. I know this looks quite pessimistic but it is a fact in none of the actual documents I’ve seen any hint on this.

... 

Rudi Vansnick

Desirability

10 March, 2014

Hi Chris, 

Apologies for the slow response.

On Mar 7, 2014, at 10:23 AM, Dillon, Chris <c.dillon@ucl.ac.uk> wrote:

[SNIP]

That said, Amr’s point about our needing to stress more the desirability of translation and/or transliteration is very important:

For example, “Do we feel that all contact information in all gTLD registries for all registrants should be translated and/or transliterated for the purpose of WHOIS look-up?”

As Rudi mentioned, the consensus is basically that it should, although Kathy Kleiman's post on the NCSG-Discuss mailing list (5 March, 2014)

https://community.icann.org/pages/viewpage.action?pageId=44960127

highlights that translation and/or transliteration could add to the financial burden of registrants.

Is there consensus that translation/transliteration is desirable?? If so, I am in the minority opposing this consensus. So far, I have yet to see one convincing reason to make me think that it is (desirable). Could someone please point one or more out to me?? I’m starting to think that I’ve missed something.

Thanks.

Amr

---

 

Translation and Transliteration of Contact Information? Informing the WG Debate (work in process)

Please feel free to add to this debate and/or edit the document - you can always go back to earlier version by using the page history button above) 

 

One of the key threshold question the WG is charted to answer is whether or not to translate and/or transliterate Contact Information. To inform the debate on this initial yes/no question Staff has collated some information from related policy and advisory efforts in the wider ICANN community.

One issue related to the question of whether to translate/transliterate Contact Information might be the purpose of Contact Information: who needs access to this data and why?

The EWG in its Initial Report has listed a number of purposes and related use cases and rationales (p.11). It might be useful for the T&T WG to go through these and see whether the Group believes that translation/transliteration of Contact Info might be useful for these different purposes of WHOIS data:

1. Domain Name Control
2. Personal Data Protection
3. Technical Issue Resolution
4. Internet Service Provision
5. Individual Internet Use
6. Business Domain Name Purchase or Sale
7. Domain Name Research
8. Legal Actions
9. Regulatory and Contractual Enforcement
10. Abuse Mitigation
11. Malicious Internet Activities

 

It might be the case that for some issues translation/transliteration might les useful than for others. The WG could assess these purposes and see whether members believe that translation/transliteration would be helpful for any/all/none of these purposes.

It might also be useful for the Group to assess ccTLD practices for the Final Report of the ICANN IRD WG contains a summary of queries to ccTLD operators (p.11):

In 2010, ICANN staff informally contacted ccTLD operators whose communities

 

Staff solicited current WHOIS submission and display practices from 16 ccTLDs by asking the following questions:

1. Does your registry allow users to register domain names using characters from local scripts?

2. Does your registry collect and store registration data in US-ASCII in addition to characters from local scripts?

3. Can users of the web interface choose the display language? What languages does your web interface support?

4. Does your registry provide access to registration information via WHOIS/port 43? Can users choose the display language?

 

The responses from the 16 ccTLD operators are summarized as follows:

• 10 ccTLDs allow users to register domain names using characters from local scripts;

• 10 of 16 ccTLDs support “English” (US-ASCII) and a local language/script, including

Arabic, Chinese, German, Japanese, Lithuanian, Portuguese, Spanish, Swedish;

• All ccTLDs support WHOIS/Port 43; and

• Character set dependencies affect WHOIS client submission and Display (some uses UTF-8, others UTF-16, or ISO-8859).

The Final Report of the ICANN IRD WG provides some interesting points regarding the benefits and the potential problems relating to translating and transliterating Contact Data

 

Benefits of Translation/Transliteration:

• Contact Information is less useful to the WHOIS service users who are only familiar with languages that require character set support other than US-ASCII; and this community is growing faster than that capable of using US-ASCII

• Many registrants are monolingual, which is the expectation and motivation behind internationalized domain names[1]

• There are also official and unofficial transliteration standards, most notably those from ISO TC 4627 and the U.S. Library of Congress

 

Possible Arguements against the necessity to translate/transliterate:

• Domain registration data is encoded in US-ASCII. This legacy condition is convenient for WHOIS service users who are sufficiently familiar with languages that can be submitted and displayed in US-ASCII

• Many millions of domain name registrants and Internet users today submit their data in US-ASCII, even though their primary language is written using a different script.

• Using US-ASCII is convenient for registrants, registrars and registries, and the installed base of operational WHOIS services that display US-ASCII.

• Many language translation systems are inexact and cannot be applied repeatedly to translate from one language to another. Thus there will be problems with both consistency and accuracy

• Translation/transliteration may vary significantly across languages using the same  script.

Further Point of consideration from the IDN WG:

• The desirability to use scripts different than US-ASCII should be balanced against the desired uses of the data. While domain registrants may intend to only use their domain "locally" or interact with people in their native script, the nature of the Internet itself means that any domain provisioned on it is available globally.

• Registrars who accept registration data in local scripts (non US-ASCII) should make registration data in the local script available through DNRD-DS output, and should at least include tags to identify languages and scripts (e.g. RFC 5646).

The GAC notes in its 2007 GAC Principles Regarding gTLD WHOIS Services:

 

• gTLD WHOIS services could provide sufficient and accurate data about domain name registrations and registrants subject to national safeguards for individual’s privacy in a manner that (1) supports the stability, reliability, security, and global interoperability of the Internet, from both a technical and public trust perspective; and (2) facilitates continuous, timely and world-wide access.

• Stakeholders should work to improve the accuracy of WHOIS data, and in particular, to reduce the incidence of deliberately false WHOIS data.