Dear Jim, Thank you for your email. I was just about to circulate the latest version (6a). Please find it attached. 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> From: James Galvin [mailto:jgalvin@afilias.info] Sent: 03 December 2014 15:10 To: Dillon, Chris; gnso-contactinfo-pdp-wg@icann.org Subject: Re: [gnso-contactinfo-pdp-wg] Virtual continuation of Thursday 20's T&T meeting Chris, Is it possible to distribute the DOC version of the file? It's much easier to comment inline than to write on paper or transcribe comments into an email message. Thanks, Jim On 11/26/14, 6:50 AM, Dillon, Chris wrote: 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>