Deprecating grandfathered tags redux

Doug Ewell doug at
Sat Mar 7 18:16:30 CET 2015

Mark Davis ☕️ wrote:

> On Sat, Mar 7, 2015 at 5:26 PM, John Cowan <cowan at> 
> wrote:
>> But, y'know, other than satisfying Mark Davis's rage for order :-),
> ​It is not particularly a rage for order; these few old irregular
> codes complicate our processing so we just toss them (mapping to
> "und").
> We'd like to have a canonical, syntactically-regular way to represent
> all of them, even though they are obscure.​ Mapping to "und" or simply
> rejecting them is certainly workable, but it would be nice to be able
> to represent them with a regular code, one that can preserve the
> original semantic. Anyone who cared could always then map them back to
> the old semantics for old programs; and those old problems have no
> need to change their use of, say, "i-default".

I'm confused, because Mark co-wrote 4646 and 5646 and I know he knows 

Irregular grandfathered tags without a P-V are part of BCP 47, just as 
much as "en". They are even included in the ABNF in 5646 to help drive 
the point home. There are only a small handful of them, and they are 
supposed to be processed as what they are, and under no circumstances 
does BCP 47 support or endorse simply throwing them away.

I say this as a software developer who has spent decades dealing with 
problems that are hard to solve and problems that are easy to solve. 
Grandfathered tags are not all that hard.

Doug Ewell | | Thornton, CO 🇺🇸 

More information about the Ietf-languages mailing list