bug 258246: RFC: Accept language header: 'en-gb' and matching

[Gervase Markham]

David Miller wrote:
> In otherwords, we can't depend on the presence of a generic "xx" to be 
> compatible with "xx-yy"

I can see why they say that, to cover their backs - but surely it is in 
fact true in 99% of cases?

If, as Tobias asserts, there are a bunch of misconfigured clients out 
there sending only de-de and not de, surely this trumps the possibility 
that someone sending pt-BR would rather have en than a generic pt?

Gerv