[Ltru] extlang and stability of macrolanguage

[Nicolas Krebs <[e-mail]>]

Hello, 

The meaning of rm has changed in january 2007
http://www.alvestrand.no/pipermail/ietf-languages/2007-January/005615.html

http://tools.ietf.org/html/draft-ietf-ltru-initial-06#page-24  
Type: language
Subtag: rm
Description: Raeto-Romance
Added: 2005-10-16

http://tools.ietf.org/html/draft-ietf-ltru-4645bis-05#page-27 
Type: language
Subtag: rm
Description: Romansh
Added: 2005-10-16

I would had prefer that the maining of rm had not been changed, 
or at least with a change notification in the registry, but this is another 
subject wich has been discuted previously 
( http://www.ietf.org/mail-archive/web/ltru/current/msg06987.html ). 

Lets suppose that later is added a new language code abc for 
Rhaeto-Romance, added as Macrolanguage: for Rhaeto-Romance 
languages. 

The rm ISO 639 language tag would become 
Type: language
Subtag: rm
Description: Romansh
Macrolanguage: abc 

Without extlang ( urn:ietf:id:draft-ietf-ltru-4646bis-14 
http://tools.ietf.org/html/draft-ietf-ltru-4646bis-14#section-2.2.2 )
the bcp 47 ietf language tags would become 

rm Romansh; macrolanguage abc 
lld Ladin; macrolanguage abc 
fur Friulian; macrolanguage abc 
abc Rhaeto-Romance

With extlang only ( urn:ietf:id:draft-ietf-ltru-4646bis-09 
http://tools.ietf.org/html/draft-ietf-ltru-4646bis-09#section-2.2.2 )
the bcp 47 ietf language tags would become 

rm Romansh; deprecated, use abc-roh
lld Ladin; deprecated, use abc-lld
fur Friulian; deprecated, use abc-fur
abc Rhaeto-Romance
abc-roh Romansh; macrolanguage abc 
abc-lld Ladin; macrolanguage abc 
abc-fur Friulian; macrolanguage abc 

With and without extlang (both allowed, 
http://www.ietf.org/mail-archive/web/ltru/current/msg10366.html )
the bcp 47 ietf language tags would become 

rm Romansh; macrolanguage abc 
lld Ladin; macrolanguage abc 
fur Friulian; macrolanguage abc 
abc Rhaeto-Romance
abc-roh Romansh; macrolanguage abc
abc-lld Ladin; macrolanguage abc
abc-fur Friulian; macrolanguage abc



Other example: Current situation for "Comorian" in 
urn:ietf:id:draft-ietf-ltru-4645bis-05 : 
swb Comorian
wlc Mwali Comorian
wni Ndzwani Comorian
zdj Ngazidja Comorian

Lets suppose that later swb Comorian change to 
macrolanguage of other "Comorian" 
( http://www.ietf.org/mail-archive/web/ltru/current/msg06987.html ). 

Without extlang ( urn:ietf:id:draft-ietf-ltru-4646bis-14 
http://tools.ietf.org/html/draft-ietf-ltru-4646bis-14#section-2.2.2 )
the bcp 47 ietf language tags would become 

swb Comorian
wlc Mwali Comorian; macrolanguage swb
wni Ndzwani Comorian; macrolanguage swb
zdj Ngazidja Comorian; macrolanguage swb

With extlang only ( urn:ietf:id:draft-ietf-ltru-4646bis-09 
http://tools.ietf.org/html/draft-ietf-ltru-4646bis-09#section-2.2.2 )
the bcp 47 ietf language tags would become 

swb Comorian
wlc Mwali Comorian; deprecated, use swb-wlc 
wni Ndzwani Comorian; deprecated, use swb-wni 
zdj Ngazidja Comorian; deprecated, use swb-zdj 
swb-wlc Mwali Comorian; macrolanguage swb
swb-wni Ndzwani Comorian; macrolanguage swb
swb-zdj Ngazidja Comorian; macrolanguage swb




It seem there would be more stability of language code (less change) 
without extlang in such case. Of course extlang have the advantage of 
easier fall back into macrolanguage, by only delete-from-right, instead 
of looking in database if there is a macrolanguage for. 


_______________________________________________
Ltru mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/ltru