Michael Everson everson at
Tue Sep 9 09:45:53 CEST 2008

On 8 Sep 2008, at 23:37, Frank Ellermann wrote:

> Michael Everson wrote:
>> I trust that this compromise will prove satisfactory.
> Actually this again shows that the possibility of "generic variants"  
> are a bug in RFC 4646.

Why? en-fonipa and bo-fonipa are both perfectly reasonable  
descriptions of text.

> What is really needed is an extension registry for this zoo (fonipa,  
> fonupa, pinyin, ...), but as long as nobody creates it, and as long  
> as this ugly hole in RFC 4646 permits such "generic variant"  
> kludges, go for it.

I don't follow you, but I see that you support registering the  
"pinyin" tag with

> I'd strip the remark about "omitting Latn", however.

I could live with that but I think in the real world people will omit  

> A proper extension registry could arrange that, but the generic  
> variant kludge asks for hardcoding this
> info in applications to get the matching right.

I don't follow. What's wrong with tagging bo-Latn-pinyin? It specifies  
a particular well-defined orthography for a particular language. I  
certainly do not think that bo-Latn-bopinyin would be "better".

Michael Everson *

-------------- next part --------------
An HTML attachment was scrubbed...

More information about the Ietf-languages mailing list