[gnso-contactinfo-pdp-wg] A virtual continuation of yesterday's meeting

["Dillon, Chris" <c.dillon@xxxxxxxxx>]

Dear colleagues,

Thank you for yesterday's meeting. I would like to continue it on the list, as 
time did not permit us to get to the end of the draft initial document.

Justine - I went through the points you asked me to raise, but may to some 
extent have garbled them.

Anyway, starting where I left off:
p.11 bullet 1: "transparent, accessible" - the question here is to whom. If the 
recommendations in the straw man are implemented in fact there is only 
definitely English in one field and so perhaps the answer is "those familiar 
with the Latin script".

bullet 2: "lose its clarity" - again, clarity to whom. (Clarity in the sense of 
quality of information is not in field for T&T, although obviously it is to be 
hoped that the next generation gTLD directory service will have better data 
quality.) Whatever script/languages addresses are in, it would be useful to 
have confirmation that they work in a field.

bullet 3: This one requires considerable thought. Is it actually an argument 
for mandatory transformation?
As I have written before now, there would be immediate pressure for the 
replacement of a next generation gTLD directory service without IRD with a 
service with the functionality.

bullet 4: Would it not be better to use data similar to those at the top of 
p.13? Is original script not more likely to facilitate contact?

Speaking of the data on p.13 and the useful note just before them - what are 
our opinions on them? How functional would such a format be?

p.12 bullets 2 and 3: Objectively it does make sense to use the Latin script if 
a common script is to be used. It has a low number of letters (admittedly not 
as low as Greek) and is used by several other large (number of users) and major 
world languages (by spread). The use of English and to a lesser extent French 
and Spanish as languages of business means that the Latin script is taught in 
many parts of the world not using English or the Latin script.

bullet 3 again: The cost of consistent, large scale transformation is likely to 
be large even if carried out centrally, for example at a country level. If 
transformation is carried out by many organizations, consistency will be an 
issue. (My first job was as a Japanese cataloguer at the British Library - I'm 
very familiar with such consistency issues. Different people will Romanize a 
long Japanese o as o, ō, ô , oo, ou or even oh.) Perhaps guidelines per 
language would improve consistency. These issues call into question the 
practicality and feasibility of mandatory transformation.

I won't repeat comments above which also apply to the recommendations.

p.14 #1 bullet 4: In the real world registries and registrars will pass down 
transformation costs to registrants, increasing the cost of domain names in the 
non-Latin script world. To what extent in each country would governmental 
organizations be prepared to subsidize registries and registrars, so that the 
costs would not be handed down?

#2 bullet 3
The idea here should be in line with the EWG's recommendations. Perhaps the 
best we can do is to use the most recent wording.

As usual, I welcome your ideas on any of the above, on things missed out and on 
things we discussed yesterday.

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>