language subtag registration request: Region EU

Tex Texin tex at
Tue Jan 17 13:05:58 CET 2006

Peter writes:

> There's no question that the EU is expanding over time. It's 
> not obvious to me that this would be a problem with using EU 
> as a region subtag in language tags. (Use of *-EU in matching 
> has different concerns, though -- see below.)
> Consider a hypothetical country Fenestria, the official 
> language of which is Fenestrian; for sake of discussion, 
> let's suppose the 3166-1 ID "FE" and the 639-2 ID "fen". At 
> present, content can be tagged "fen" or "fen-FE" (perhaps 
> it's used with different spelling in a neighbouring country). 
> Let's also suppose that in 2012 Fenestria becomes a member of 
> the EU. At that point, content can be tagged "fen-EU".
> While there may be ambiguity regarding the meaning of "EU" 
> used in isolation, there's no ambiguity whatsoever regarding 
> "fen-EU": it's Fenestrian as used within the EU.

Peter, whether it is ambiguous depends on how the language is used in the
countries that are added.
For example, if we use a subtag for South America in conjunction with
Spanish, and then we add central America and
in particular Mexico to the "S. America Union", then the Spanish language
for this region will need to adapt to the fact that Mexico uses different
Spanish words from South America.

Consider also defining English as spoken in the U.S. and Mexico, and lets
imagine that Canada is then added to the mix. Now we have to accommodate
spelling and wording changes if we are going to define the language for the
Add the U.K. to the "region" for additional fun. When you "table" an agenda
item in this region, do you add it or remove it? ;-)

I am presuming in these examples, that language for a "union" is handled
similar to Spanish for Latin America, where localizers try to find the
common denominator and the most acceptable terms and spellings for the

An alternative is to define a superset of the spellings and terms used in
the region, treating them all as part of the language. 

The former approach is good for localization. The latter is good if you are
creating a dictionary representing the "language" of the region (and all its


More information about the Ietf-languages mailing list