Casefolding Sigma (was: Re: IDNAbis Preprocessing Draft)

Kenneth Whistler kenw at sybase.com
Tue Jan 22 21:27:44 CET 2008


I find myself in complete agreement with John on this topic:

> Nothing prevents their "using final sigma".  ...
> 
> What they are prevented from doing is to have a label, say,
> αβςψω and expect to distinguish it from αβσψω.  ...
 
> If you are convinced that it is really a different character, no
> more related to sigma than alpha is to omega, then the IDNA2003
> case-mapping action was a mistake. 

And of course it was not a mistake. The treatment of sigmas
is perfectly normal casefolding behavior for Greek.

What the "prohibition" in IDNA is all about is prohibiting
distinctions in domain names that depend only on case
differences from getting registered (and causing attendant
mischief). And because of the way casing works in Greek,
allowing *distinct* domain names that depend only on
choice of one of two lowercase sigmas would definitely result in
confusion and mischief.

That doesn't prevent Greeks from *displaying* domain names
or IRI's with preferred positional variants if they happen
to be in lowercase, any more than it prevents them from
*displaying* Greek domain names or IRI's in uppercase -- even
though all the uppercase Greek letters are *also* NEVER
characters in the IDNAbis tables we are reviewing now.

--Ken



More information about the Idna-update mailing list