Greek Casefolding sigma

Mark Davis mark.davis at
Sat Mar 29 16:47:56 CET 2008

Patrik, you misunderstand. I'm not saying that this should be part of the
protocol. What I'm saying is that the protocol, *combined with a
postprocessing step for UIs, *would handle the situation.

(In retrospect -- and water long under the bridge -- we would have been
better off to use one of the variants of Punycode, which has the ability to
encode case and other distinguishing information in the original Unicode
using case in the ASCII form. Had we gone that route, we could have
maintained the visual distinctions on output of DNS for sigma and similar
cases, because the DNS does a caseless compare for A-Z.)


On Sat, Mar 29, 2008 at 2:00 AM, Patrik Fältström <patrik at> wrote:

> On 28 mar 2008, at 18.52, Mark Davis wrote:
> > That would show
> > σωτήρης<>properly, and maintain compatibility.
> This is not about how to "show" things, as DNS is nothing else than a
> storage / lookup mechanism.
> You store a key, and together with that key some data. You can _NOT_
> send in the data and get the key back.
> The way IDNA2003 is designed is that the key does _NOT_ allow final
> sigma, and this because of design of Unicode. Regardless of if "the
> user typed" sigma or final sigma, the same key should match. Without
> changing the matching algorithm (that is out), the key have to be
> calculated on the client side. This key that is sent in is sigma,
> never final sigma.
> There is _no_ "reverse" mapping in DNS. Just lookup.
> That said, one can do a lookup of a PTR record where the data is a
> domain name. This domain name because of same rules as for the key can
> not include the sigma, but the final sigma. And if that is the problem
> people try to resolve, then that is what should be discussed, but that
> has not much to do with the creation of the key used in the lookup.
>    Patrik

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

More information about the Idna-update mailing list