Re: Click letters in ISO 639-3 and Registry (long)

With respect, that’s one subtag. Can we make similarly quick and accurate decisions about the other 100 or so? If you can, that’d be great; post your findings to the list and we can discuss them.

We know that usage by Ethnologue and 639-3 of apostrophes and other non–Latin-1 characters doesn’t necessary match the spelling we would think, or each other. It’s the same problem as with the click letters. We know this because Ethnologue and 639-3 consider these to be “display issues” and not character or name identity issues. It may simply have come down to what somebody’s font did and didn’t support N years ago. (That may be where the perception came from that Unicode didn’t support click letters until 2004.)

Fortunately, we have an opportunity to fix the 639-3 issue with click letters. I’d have to check, though, whether they would be similarly amenable to changing 100 names with apostrophes.

--
Doug Ewell | Thornton, CO, US | ewellic.org



From: Hugh Paterson 
Sent: Tuesday, February 27, 2018 17:54
To: Doug Ewell 
Cc: ietf-languages ; ISO639-3 
Subject: Re: [Ietf-languages] Click letters in ISO 639-3 and Registry (long)

The apostrophe like character in ISO 639-3 [gel] should not be word-forming http://unicode.org/reports/tr29/#Word_Boundaries (causing double click behavior to fall short of selecting the whole word).


in the case of [gel], The ISO 639-3 website uses U+0027, while the Ethnologue website uses U+2019. I brought this to the attention of the ISO 639-3 registrar some years ago (maybe 3). My point was that the published names are not the same with respect to the encoding of characters outside of the Latin charters [a-zA-Z]. And neither seem to respect the usage in the script by speakers.

So based on the encoded character's behavior (Unicode attributes) I would suggest that U+02BC be used here, not U+2019 or U+0027. I was told at the time that U+0027 was used in ISO 639-3 to be ASCII compliant, while Ethnologue could make its own choices.


- Hugh


On Tue, Feb 27, 2018 at 3:01 PM, Doug Ewell <[email protected]> wrote:

  Hugh Paterson wrote:

  > Are there other language names that are not using U+02BC MODIFIER
  > LETTER APOSTROPHE which should be?

  I count roughly 132 language subtag Description fields which include the
  ASCII apostrophe. That includes the click languages already discussed,
  as well as some cases like "Coeur d'Alene" and "Brooke's Point Palawano"
  which appear to be real apostrophes and not apostrophe-letters. (that's
  not to say they shouldn't be changed to U+2019, but.)

  I don't know if I or anyone else is prepared for the challenge of
  researching and updating 100 or more Description fields. Equally, I
  don't know if it would be a mistake to fix U+02BC for the click
  languages and not for the others.


  --
  Doug Ewell | Thornton, CO, US | ewellic.org




--------------------------------------------------------------------------------
_______________________________________________
Ietf-languages mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ietf-languages

_______________________________________________
Ietf-languages mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ietf-languages