prohibiting previously mapped and unmapped characters

[Erik van der Poel]

One correction: 0.0188% of all the URLs in the sample contained
character sequences in their domain names that were mapped to
something else in the IDNA and Nameprep processes, but not the
Punycode process. This includes the various versions of the dot (CJK,
full-width, etc), characters mapped to nothing and any sequences
affected by normalization and case-mapping, excluding ASCII
case-mapping.

Erik

On 12/1/06, Erik van der Poel <erikv at google.com> wrote:
> 0.0188% of the domain names
> are mapped to different strings by the IDNA process, from the links
> found in HTML to the domain names passed to DNS.