December 2014 - Gnso-contactinfo-pdp-wg

MP3 Translation and Transliteration of Contact Information PDP WG meeting - 04 December 2014
by Terri Agnew Dec. 4, 2014

Dec. 4, 2014

Dear All, Please find the MP3 recording for the Translation and Transliteration of Contact Information PDP Working Group call held on Thursday 04 December at 14:00 UTC at: <http://audio.icann.org/gnso/gnso-transliteration-contact-20141204-en.mp3> http://audio.icann.org/gnso/gnso-transliteration-contact-20141204-en.mp3 On page: <http://gnso.icann.org/en/group-activities/calendar#dec> http://gnso.icann.org/en/group-activities/calendar#dec The recordings and transcriptions of the calls are posted on the GNSO Master Calendar page: <http://gnso.icann.org/calendar/> http://gnso.icann.org/calendar/ Attendees: Chris Dillon – NCSG Rudi Vansnick – NPOC Peter Green (Zhang Zuan)-NCUC Jim Galvin - RySG Justine Chew – Individual Petter Rindforth – IPC Amr Elsadr – NCUC Emily Taylor – RrSG Pitinan Kooarmornpatana – GAC Mae Suchayapim Siriwat – GAC Wanawit Ahkuputra – GAC Apologies: Jennifer Chung – RySG ICANN staff: Julie Hedlund Lars Hoffmann Terri Agnew ** Please let me know if your name has been left off the list ** Wiki page: <http://tinyurl.com/mpwxstx> http://tinyurl.com/mpwxstx Thank you. Kind regards, Terri Agnew GNSO Secretariat Adobe Chat Transcript for Thursday 04 December 2014: Terri Agnew:Dear all, Welcome to the Translation and Transliteration of Contact Information PDP WG call on 04 December 2014 Pitinan Kooarmornpatana:hello Chris Dillon:Hello everyone Pitinan Kooarmornpatana:Hi Chris Petter Rindforth:Hi there, all six of you! Rudi Vansnick:hi everybody Chris Dillon:Hi Rudi Rudi Vansnick:still fighting with many deadlines Rudi Vansnick:strikes are not really very helpful Lars Hoffmann:Chris you are a host now - so scrolling should be fine. Terri Agnew:Mae Suchayapim Siriwat has joined Pitinan Kooarmornpatana:just some editorial comment , the GAC full name is missing from the footnote on page 8 Amr Elsadr:Hi. Apologies for being late. Terri Agnew:Welcome Amr Elsadr Amr Elsadr:Thanks Terri. Amr Elsadr:I haven't had a chance to go over the last version sent (6). If possible, I'd like to send comments over the list. Amr Elsadr:I was referring to 6a in my last comment. Emily Taylor:Agree with Jim's suggestion about ordering of bullet point - expense first Amr Elsadr:+1 Emily Taylor:Did you mean ccTLDs, Chris? Rudi Vansnick:ccTLD's have other relations to the data in the whois than the GTLD's Amr Elsadr:@Chris: Which point are you referring to? Amr Elsadr:Thanks. Justine Chew:pg 13? Amr Elsadr:My understanding of this bullet is similar to Jim's, in that this bullet addresses labels rather than the actual contact data. Amr Elsadr:@Chris: +1 Amr Elsadr:Would we want to specify that the labels should be translated vs. transformed? Jim Galvin:i'm sorry I just had a conflict arise and I need to leave this call. Amr Elsadr:The Japanese use two local character sets; Hiragana for native words and Katakana for foreign words adopted locally + Kanji. :) Terri Agnew:goodbye Jim Pitinan Kooarmornpatana:ok Pitinan Kooarmornpatana:i got disconnected for a while Pitinan Kooarmornpatana:just want to add that i'm agree with the label transformation Pitinan Kooarmornpatana:and we might want to refer to the list of languages here Pitinan Kooarmornpatana:ISO_639-1 Rudi Vansnick:http://en.wikipedia.org/wiki/ISO_639-1 Pitinan Kooarmornpatana:and i might want someone helping explain more on how it will display Pitinan Kooarmornpatana:like.. on the interface of the registrar? Rudi Vansnick:Codes for the representation of names of languages — Part 1: Alpha-2 code, is the first part of the ISO 639 series of international standards for language codes. Part 1 covers the registration of two-letter codes. There are 136 two-letter codes registered. The registered codes cover the world's major languages. Petter Rindforth:Can thar be done in a way that is easily searchable for anyone looking for the contyact information - "A quick search"? Rudi Vansnick:See also IETF language tag. Emily Taylor:I'd suggest that we try to think of an "on the fly" type of translation of the labels both for people entering data and for people searching data Emily Taylor:Otherwise it could become very cumbersome and expensive Rudi Vansnick:http://en.wikipedia.org/wiki/IETF_language_tag Rudi Vansnick:if we could stick to the IETF than we respect the Internet standards Emily Taylor:We could suggest a minimalist approach (6 UN languages) Pitinan Kooarmornpatana:or at least the language that has ccTLDs Rudi Vansnick:question : which limit do we foresee (number of languages) ? could exclude important ones Emily Taylor:I understand that over 90% of the world's population speak about 70 languages Emily Taylor:That might miss first language for many. Emily Taylor:But it's a bit more tractable than the 6,000 that UNESCO believe in existence Rudi Vansnick:is it our duty to define the number of languages ? Emily Taylor:The UN languages may be unsatisfactory, but there is a certain consistency in that it kind of works in the UN environment and people are used to it Rudi Vansnick:let us not forget the target market is GTLDs and there could be GLTD's that only address specific region Amr Elsadr:To be clear, are we suggesting that the data labels are available in local languages/scripts, but only translated to (for example) the 6 UN languages? Emily Taylor:Good point, Amr Petter Rindforth:Agree Pitinan Kooarmornpatana:agree Peter Green （CONAC）:Hi everybody，maybe the point I wanna mention is not the point being discussed by you guys now.I would like to pick up with an issue that has always been haunting me since joining this group. Peter Green （CONAC）:In Page 10 we have the agreement that "the main purpose of transformed data is to allow those not familiar with the original script of a contact information entry, to contact the registrant", then following "[t]his means that the accuracy of contact information data that are entered and displayed is paramount." Peter Green （CONAC）:I think there is a logical issue here, which we may have missed. Peter Green （CONAC）:My point is, starting from the need of "contact", how is it possible or likely for those unfamiliar with the originalscript of a contact information entry to contact the registrant"? Peter Green （CONAC）:How is it possible for an English speaker to contact the registrant who registers a domain name in Arabic?How is it likely for a Chinese to contact the registrant who registers a domain name in Russian? Peter Green （CONAC）:I think we should do some discussion work on the "possibility", other than "accuracy" mentioned above. Peter Green （CONAC）:It is the "possibility" that matters in regards to whether transformation is "mandatory". Peter Green （CONAC）:thanks Chris Pitinan Kooarmornpatana:just to be clear, have we narrowed down the scope to trasforming label only ? Peter Green （CONAC）:currently，i think it is Pitinan Kooarmornpatana:oh... i c Amr Elsadr:@Chris: Yes. That's the conundrum. :) Emily Taylor:Agree with point about registrars should be able to offer services in whatever language is convenient. It's an area we do need to unpack a bit more Pitinan Kooarmornpatana:ok. Thanks Chris Emily Taylor:Amr's earlier point was useful. My understanding is that - going with a UN language parallel - my understanding is that people can speak in their own language, and it will be translated *into* those 6 languages. So, in a WHOIS context, if UTF8 is supported data labels and input can be in local languages Emily Taylor:And then maybe we could explore "what if the data labels were available in a limited number of languages" Amr Elsadr:@Emily: Thanks. That's the direction I was trying to head us into. :) Pitinan Kooarmornpatana:+1 Emily Taylor:Chris on the marketing point, I'm thinking, leave the marketing up to the channel, and provide tools *after the fact* for others to understand what the labels mean Amr Elsadr:@Emily: +1 Rudi Vansnick:@Amr : +1 Pitinan Kooarmornpatana:+1 with Amr Pitinan Kooarmornpatana:we just point then how to translate correctly. like if you have the code TH in ISO691 then you can look up to use the word "ไทย" if you plan to serve Thai Customer Emily Taylor:@Chris on your consistency point, it seems do-able to have a consistent data set in a limited number of languages Amr Elsadr:Yes. I would think it is far more doable than consistency with actual contact data. :) Emily Taylor::) Pitinan Kooarmornpatana:Thanks all :) Emily Taylor:Thank you Chris. Great chairing Lars Hoffmann:thanks everybody, thanks chris! Amr Elsadr:Thanks all. Bye. Apologies for being late again. Rudi Vansnick:thanks Chris and to all of you for this fruitful discussion Chris Dillon:Many thanks, all Peter Green （CONAC）:bye Chris， bye all Chris Dillon:Bye all

1 0

Virtual continuation of Thursday 20's T&T meeting
by Dillon, Chris Dec. 3, 2014

Dec. 3, 2014

Dear colleagues, As I mentioned in my last meeting I would like to continue Thursday's call virtually, encouraging you to comment on version 5 of the Draft initial report (attached). Here are the comments I would have made if we had had more time: p.12 Mike asks us [MZ14 and MZ31] whether "increase in users that are not familiar with the Latin script" should be replaced with "increase in users whose languages are not based on the Latin script". Both aspects are true, but the latter wording hints at the former and so is a good replacement. As a statement the latter version would also be less Anglocentric. In MZ15, Mike suggests that some statements about law enforcement are actually broader. That seems true and I can at least add "for example". However, does anyone have concrete examples of organizations apart from law enforcement for which transformation to a Latin script would be useful? p.13 In ER16 and ER17 Erika highlights an apparent contradiction. The bullet point at the top of p.12 says that transformation would need to take place at a later stage (than entry by registered name holders) and that this would be detrimental to accuracy and consistency. The bullet point above the ccTLD graphic argues that only the data fields should be transformed by the registrar or registry. I will make clear the distinction between transformation (of data - how we have been using the term "transformation" on its own) and transformation of field names. Moreover, accuracy (at least senses 1 and 2 in the footnote) and consistency are likely to be worse the greater number of players involved i.e. if registrants were to do the transformation. CD18 Is anyone aware of reasons why the ccTLD approach exemplified wouldn't work with gTLDs? I reckon MZ21 is addressed by "not justified by benefits to others", the last line of p.12. MZ22 suggests the text "if no consensus is reached the status quo will be maintained". The key thing here is whether we're talking about the current Whois status quo where the system cannot accept non-Latin script data (answer "no" as this does not encourage the development of the Internet in wide areas of the world not using the Latin script), or a new DNRD with no Latin script (answer "no" as such a system would be very expensive, as it would need to be replaced soon) or a new DNRD with non-Latin script functionality (answer: possibly "yes" as the status quo would not involve transformation, except possibly of field names). ER23 picks up how we would handle a situation where we move from no clear consensus to a greater level of consensus. The short answer is to use the GNSO procedure. ER27 Automated transformation cannot occur if data are not marked as being in a language. ER29 "it" refers to "contact information data" and so should be "them". India-based companies are an interesting case, as in many cases there will be three or more possible languages - Hindi, English and a local language. If the language is not stipulated, there will be consistency issues in the event of transformation. Lars suggests the language the registrar operates in, but again there could be several and bad actors could deliberately apply in different languages to different registrars. As usual I welcome your views on any of these issues, or issues not in this list. I shall circulated a new version of the draft initial report shortly before our meeting on Thurs. 4 December. Regards, Chris. -- Research Associate in Linguistic Computing, Centre for Digital Humanities, UCL, Gower St, London WC1E 6BT Tel +44 20 7679 1599 (int 31599) www.ucl.ac.uk/dis/people/chrisdillon<http://www.ucl.ac.uk/dis/people/chrisdillon>

2 2

Straw poll on number of options
by Dillon, Chris Dec. 1, 2014

Dec. 1, 2014

Dear colleagues, During Thursday's call, we had a straw poll: Are you in favor of having only one option in the initial report? As you know, in the versions of the draft initial report until now there have been two options (recommendations for and against mandatory transformation), but if it is possible to have only one, the report will likely have a stronger effect. Whatever the result of the report, the arguments for and against will remain in it; this poll only concerns the options. Please send your vote to the list if you did not vote on Thursday. The options are: Yes, No and Abstain. Please vote by 14:00 UTC on Thursday 27 November. (Note that there is no meeting on that day; the next one is 4 December.) In summary - This is not a consensus call on the options. - This is to decide whether the initial report should have one set of recommendations or two sets of recommendations. - If a majority believes it should be only one set, the WG, at a later stage (probably during our next meeting, on 4 December) will decide which set it will be. - Please bear in mind that this is the initial report and following public comments on it we will be able to modify/amend/change/reverse our draft recommendations. Incidentally, I shall email soon asking for your comments on version 5 of the report and including the rest of mine. Regards, Chris. -- Research Associate in Linguistic Computing, Centre for Digital Humanities, UCL, Gower St, London WC1E 6BT Tel +44 20 7679 1599 (int 31599) www.ucl.ac.uk/dis/people/chrisdillon<http://www.ucl.ac.uk/dis/people/chrisdillon>

8 16

Results of the straw poll on the number of options
by Dillon, Chris Dec. 1, 2014

Dec. 1, 2014

Dear colleagues, I have added the results of the straw poll on the number of options during our meeting on 20 Nov. to this wiki page: https://community.icann.org/display/tatcipdp/20+November+2014 Please check that I have recorded your vote correctly and get in contact if I haven't. Regards, Chris. -- Research Associate in Linguistic Computing, Centre for Digital Humanities, UCL, Gower St, London WC1E 6BT Tel +44 20 7679 1599 (int 31599) www.ucl.ac.uk/dis/people/chrisdillon<http://www.ucl.ac.uk/dis/people/chrisdillon>

2 2