Re: [Ltru] Eliminating the preposterous ASCII ordering in lookup

I'm guessing, but only a guess, that this is related to the following text:

> For example, an implementation could return the matching content that 
> is first in ASCII-order. For example, if the language range were 
> "*-CH" and the set of content included "de-CH", "fr-CH", and "it-CH", 
> then the content labeled "de-CH" would be returned.
"preposterous" is a overblown language. And I strongly disagree with 
your proposed change. In the example there clearly *exists* content 
matching *-CH. So returning the default content (which could be, say, 
Japanese) is clearly *not* what I would have expected!

There could, of course, be other strategies for picking a single tag to 
return from lookup when there are multiple matches for a wildcard. But 
whatever example we choose should not return something so clearly 
disconnected from the user's desired outcome. And ASCII is simple to 
explain.

Mark

John Cowan wrote:
> It's really laughable to suggest that implementations might use ASCII
> tag ordering to make fallback decisions on lookup.  IMHO, the behavior
> of extended ranges on lookup should be:
>
> 	If the first subtag of a language range is '*' and it is
> 	followed by other ranges in a priority list, skip it.
> 	If the first subtag is '*' and there are no following
> 	ranges, return the default content.  All other '*' subtags
> 	should be removed before lookup processing is done.
>
> That is simple, clear, to the point, straightforward, and doesn't provide
> preposterous results.
>
>   

_______________________________________________
Ltru mailing list
Ltru@ietf.org
https://www1.ietf.org/mailman/listinfo/ltru