AW: AW: AW: sharp s (Eszett)
duerst at it.aoyama.ac.jp
Thu Mar 13 09:02:05 CET 2008
Georg has already provided a lot of detail in his answer.
I'd like to add two points:
1) All-uppercase domain names are used very rarely, in particular
for input. If upper-case doesn't provide a good distinction,
I'm sure all-uppercase also won't be used for publishing these
names. On top of that, add the fact that there is a bias in
IDNA2003 for lower-case names, and the current proposal for
IDNAbis strengthens this bias.
2) For people who have never lived with anomalies in case
mapping, it is difficult to immagine how these would best
be handled by people who actually have daily experience.
When I started to learn some foreign languages, I was
very vary that something like not having a singular/plural
distinction could ever work. Now that I use such a language
daily, I don't even have to thing about it. The human mind
is indeed very flexible, and in the end, we are designing
IDNs for humans.
At 12:27 08/03/12, Harald Tveit Alvestrand wrote:
>--On Tuesday, March 11, 2008 22:44:46 +0100 Georg Ochsner <g.ochsner at revolistic.com> wrote:
>> Can you explain for which applications in particular the
>> case-insensitivity for domain names is essential, and why? Can such
>> applications make use of the new uppercase sharp S in Unicode 5.1?
>We have trained all users since the birth of the DNS to expect that "FreeNet.de", "freenet.de" and "FREENET.DE" all point to the same host.
>I (speaking from my Norwegian context) think it's essential that the same thing be true for "$B%F)M(Besund", "$B%F)-(BESUND" and "$B%F!&(Blesund" too; that's the general "application" of "humans type a domain name and expect it to behave like domain names they're used to".
>Idna-update mailing list
>Idna-update at alvestrand.no
#-#-# Martin J. Du"rst, Assoc. Professor, Aoyama Gakuin University
#-#-# http://www.sw.it.aoyama.ac.jp mailto:duerst at it.aoyama.ac.jp
More information about the Idna-update