Putting it another way

[Jon Hanna]

> B. iso_639_code ("-" iso_15942_code)? ("-" iso_3166_code)?
> or
> C. iso_639_code ("-" iso_3166_code)? ("-" iso_15942_code)?
> 
> Either one of these is a reasonable extension of the format that takes
> important differences in written language into account. I myself favor
> (B), since it is in big-endian format, but that does not have to be
> decided now. This could be generalized to

You know it was when you called the first format "big-endian" that I started believing we should consider them orthogonal and somehow reflect that, and hence that the rush to register these tags was cat.