No three-letter variants

CE Whitehead cewcathar at
Sat Feb 17 20:26:32 CET 2007

Hi, my comments are below:

>Please do not propose or offer support for three-letter variant subtags 
>such as "nta".  They are syntactically invalid (not well-formed) and will 
>not be considered for registration.
>The syntactic restrictions on variant subtags are described in Section 
>2.2.5, item 4.  There is a very sound reason for these restrictions: they 
>make it possible for humans and computers to tell the type of each subtag 
>without looking each one up in the Registry.  In the RFC 4646bis era there 
>will be a tag "az-azj" for North Azerbaijani.  If a tag "az-nta" were also 
>permitted, it would be impossible to know that "azj" was an extlang and 
>"nta" a variant without consulting the Registry.
>Doug Ewell  *  Fullerton, California, USA  *  RFC 4645  *  UTN #14

here's the RFC 4646 info:

      1.  Variant subtags that begin with a letter (a-z, A-Z) MUST be
           at least five characters long.

       2.  Variant subtags that begin with a digit (0-9) MUST be at
           least four characters long.
This leaves us with a couple possible subtags that have been proposed

first, for all of the variants there are:


and for mainly Tatar:


My objection to panturk is that its nomination has not yet been seconded by 
the person originally seeking the subtags;
in addition of course it's not as descriptive (to me) as is nta1926 as it 
does not indicate that the script is Latin.

But whatever everyone decides on.

--C. E. Whitehead
cewcathar at

With tax season right around the corner, make sure to follow these few 
simple tips.

More information about the Ietf-languages mailing list