Re: [Ltru] Preferred Values for Irregular Tags

[Karen_Broome at spe.sony.com]

>From what I'm reading on the Web -- I'm no expert in this area -- it 
appears that Brythonic includes Welsh, Cornish, and Breton. Gaulish seems 
to be a more specific tag. 

ISO 639-3 offers:

Identifier: xcg 
Name: Cisalpine Gaulish 

Identifier: xtg
Name: Transalpine Gaulish

Would either of these suffice?  (Hmm. And would extlang be useful here 
too?)

"oeddict" -- I might prefer "oxford" as the tag if there's not another 
linguistic use for that term. 

"zhmin" -- Seeing the discussion this week on this list, I'd suggest that 
any tag name using "zh" is asking for a semantic mess. *In practice*, this 
would be understood as Mandarin-Hokkien under current assumptions, which 
doesn't make much sense.  If this is not the same as "nan", would a name 
like "hokkien" be preferable? 

Regards,

Karen Broome





"Mark Davis" <mark.davis at icu-project.org> 
Sent by: ltru-bounces at ietf.org
05/07/2008 03:35 PM

To
"LTRU Working Group" <ltru at ietf.org>
cc

Subject
Re: [Ltru] Preferred Values for Irregular Tags






BTW:

http://www.iana.org/assignments/lang-tags/i-enochian
http://www.iana.org/assignments/lang-tags/i-mingo
http://www.iana.org/assignments/lang-tags/zh-min
http://www.iana.org/assignments/lang-tags/cel-gaulish

For zh-min, nan might be the best choice:

%%

Type: grandfathered
Tag: zh-min
Description: Min, Fuzhou, Hokkien, Amoy, or Taiwanese
Added: 1999-12-18
Deprecated: 2029-09-09

%%
Type: grandfathered
Tag: zh-min-nan
Description: Minnan, Hokkien, Amoy, Taiwanese, Southern Min, Southern 
Fujian, Hoklo, Southern Fukien, Ho-lo
Added: 2001-03-26
Preferred-Value: nan
Deprecated: 2029-09-09

cel-gaulish

it looks like this was intended to be what the ethnologue calls Brythonic:

http://www.ethnologue.com/show_family.asp?subid=92090

So we would need a new code, or ask ISO to add one. (Having codes for the 
families in the Ethnologue would be pretty handy; many exist, but many 
don't.)


Mark


On Wed, May 7, 2008 at 1:43 PM, Mark Davis <mark.davis at icu-project.org> 
wrote:
> With RFC4645bis, there are only a handful of cases (listed below) where 
one cannot canonicalize on input and then not ever have to deal with 
irregular tags. That would allow us to only have to deal with irregular 
codes on input, and then never again have to handle them -- which would be 
a very useful simplification.
> 
> I was thinking of proposing to the languages group that we make the 
following registrations:
> 
> 
> languages: gaulish, default, enochian, mingo, and zhmin
> variant: oeddict
> 
> I wanted to run this by this group first, especially to see if there is 
nothing on the horizon in 639 that we could use instead of any in the 
first group (since that would of course be preferable), and if anyone else 
had any thoughts on the matter.
> 
> %%
> Type: grandfathered
> Tag: cel-gaulish
> Description: Gaulish
> Added: 2001-05-25
> %%
> 
> %%
> Type: grandfathered
> Tag: en-GB-oed
> Description: English, Oxford English Dictionary spelling
> Added: 2003-07-09
> %%
> 
> %%
> Type: grandfathered
> Tag: i-default
> Description: Default Language
> Added: 1998-03-10
> %%
> 
> %%
> Type: grandfathered
> Tag: i-enochian
> Description: Enochian
> Added: 2002-07-03
> %%
> 
> %%
> Type: grandfathered
> Tag: i-mingo
> Description: Mingo
> Added: 1997-09-19
> %%
> 
> %%
> Type: grandfathered
> Tag: zh-min
> Description: Min, Fuzhou, Hokkien, Amoy, or Taiwanese
> Added: 1999-12-18
> Deprecated: 2029-09-09
> %%
> 
> -- 
> Mark 



-- 
Mark _______________________________________________
Ltru mailing list
Ltru at ietf.org
https://www.ietf.org/mailman/listinfo/ltru



_______________________________________________
Ltru mailing list
Ltru at ietf.org
https://www.ietf.org/mailman/listinfo/ltru