[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[idn] FW: IDN status
- To: IDN List <idn@ops.ietf.org>
- Subject: [idn] FW: IDN status
- From: Jonathan Rosenne <rosenne@qsm.co.il>
- Date: Fri, 21 Apr 2000 07:26:19 +0200
- Delivery-date: Thu, 20 Apr 2000 21:30:29 -0700
- Envelope-to: idn-data@psg.com
> -----Original Message-----
> From: Jonathan Rosenne [mailto:rosenne@qsm.co.il]
> Sent: Thursday, April 20, 2000 9:44 AM
> To: unicore@unicode.org
> Subject: RE: IDN status
>
>
> Maybe it would be simpler for a Chinese site to publish two equivalent names,
> traditional and simplified.
>
> In general, I think it is a good idea to allow equivalent names in different
> languages, for example to allow a Swiss site to have names in French, German
> and Italian.
>
> Anyhow, the name is just a way to access the IP number, and a many-to-one
> relationship should cause to much trouble.
>
> Jony
>
> > -----Original Message-----
> > From: Kenneth Whistler [mailto:kenw@sybase.com]
> > Sent: Wednesday, April 19, 2000 7:57 PM
> > To: Multiple Recipients of Unicore
> > Cc: kenw@sybase.com
> > Subject: Re: IDN status
> >
> >
> > Paul,
> >
> > Thanks for the update.
> >
> > > One thing that came up in the WG meeting and again
> > > afterwards is the desire to have "Han folding". If we can create good,
> > > solid tables with only a single, non-contextual transformation for things
> > > like Traditional-to-Simplified Chinese and some of the characters marked
> > > zVariant in the Unihan.txt file, this could possibly be part of the
> > > canonicalization step.
> >
> > Speaking of allocating the pain, this one may be rather painful. The
> > relationship of Traditional-to-Simplified is anything but simple. And
> > while you may be able to produce a good, non-contextual transformation for
> > the 90% case, I suspect you will end up with a good amount of untrimmed
> > hair that will be difficult to deal with.
> >
> > Folding the CJK compatibility characters and the zVariants is likely to
> > be easier, but even there could get you into trouble. Some of the
> > zVariants are distinct in the source sets because of required glyphic
> > distinctions. Whether these differences would extend to domain names,
> > however, I'm not sure.
> >
> > --Ken