Re: Canonical URIs for regional variants in LOD datasets

Hi Ghislain,

There is a trick about two letter acronyms that US Civil Servants know from being US Civil Servants.In this case, they are also top drawer Librarians and the LOC is an ISO Maintenance Organization, but that's a bit of misdirection.
Any two letter Acronym can be translated to a leading zero filled three digit integer. The formula including provision for case is:Upper Case : (26 * (ord("$letters[0]") - 64)) + (ord("$letters[1]") - 64) - 26"CA" = ("C","A") = California = 053"MX" = ("M","X") = Mexico =  336
Lower Case   : (26 * (ord("$letters[0]") - 96)) + (ord("$letters[1]") - 96) - 26
"en" = ("e","n") = English = 118
"es" = ("e","s") = Spanish = 123

I was going to ask Phil Archer to consider bar coding the resulting URI's. (en-CA = 118-053, etc.) This would cut way down on misclassification of the records.  Sadly, I got busy and never followed up.  Phil cc'd.
(please tell no one in Silicon Valley about this trick.  The inventors of addition, subtraction and multiplication who dwell there will sue me silly)

--Gannon


      From: Ghislain Atemezing <auguste.atemezing@eurecom.fr>
 To: Linking Open Data <public-lod@w3.org> 
 Sent: Thursday, September 14, 2017 4:58 AM
 Subject: Canonical URIs for regional variants in LOD datasets
   
Hi all,
I am looking for canonical URIs to represent regional variants in a multilingual dataset. Fo example, I want to be able to use all the variants of Spanish say es-XX(e.g., es-MX for Mexican variant for Spanish), or English (e.g., en-CA).I was looking around, specially in the Library of Congress endpoint [1], but I am not able to have such URIs for geographical/regional variants. 
Any thoughts or pointers very much appreciated!
TIA.
Best,
Ghislain 
[1] http://id.loc.gov/download/ 
---------------------------------------Ghislain A. Atemezing, Ph.DMail: ghislain.atemezing@gmail.comWeb: https://w3id.org/people/gatemezingTwitter: @gatemezingAbout Me: https://about.me/ghislain.atemezing