Doug Ewell dewell at
Sat Jul 1 19:05:36 CEST 2006

Addison Phillips <addison at yahoo dash inc dot com> wrote:

> I agree that the order in the registry is not significant. Consider 
> the many names of Old Church Slavonic.

Randy and Addison are correct.  If anyone is taking the order to be 
significant, it is outside the boundaries of the RFC.

In my tag-generating application, the user can either select the name of 
a language (etc.) from a drop-down list box or can type the subtag 
directly.  Changing one field automatically changes the other, so if you 
type "en" the drop-down list will display "English."  For the languages 
(etc.) that have multiple Description fields, I had to choose one to be 
displayed in this scenario.  So you can select either "Spanish" or 
"Castilian" to generate subtag "es", but if you type "es" in the edit 
box you will see "Spanish" and not "Castilian" in the drop-down.  This 
is intentional, but not meant to be prescriptive, and I understand it is 
outside the purview of the spec.  (The validation screen does show 
"Spanish; Castilian.")

> However, in this case it might be useful to choose a particular order.

The current document doesn't specify anything about this.  We should 
review this during the 3066ter cycle, to see whether it would be 
desirable (I think it would be, but am willing to be proven wrong). 
Debbie points out it would be a simple, low-cost change that might 
improve the relationship with ISO 11179.

> Also: has anyone considered my proposal of using the Comments field 
> for option (C)?
> Description: Ethiopic (Ge&#x2BB;ez)
> Comments: in ASCII "Ge'ez"

I don't like this in principle -- I think comments should provide 
additional information about subtags, not workaround for structural 
problems in the Registry-- but I won't fight to the death over it.

Doug Ewell
Fullerton, California, USA

More information about the Ietf-languages mailing list