Re: [idn] Internationalized PTR draft submitted

[Hongbo Shi <shi@goto.info.waseda.ac.jp>]

Hi, I am Hongbo Shi, one of the authors. James Seng is a person who has 
given us many comments. :)

Now let me explain our IPTR draft.
 
At first, I think internationalized domain name also means how to have 
multiples names for a single machine/IP in different languages. 
Such as www."iDN-in-JP".waseda."iDN-in-JP".jp. 

Due to the thinking, I suggest a new RR TYPE to provide address-to-iDN, 
exactly I should say address-to-iDNs. 
                                   ^^
Second, about "language tag" 
I also think whether it is necessary or not. But my reason is different 
from yours. :) 

If we talk about "language" on DNS, we will encounter a problem that the 
language is per domain name? per label? within label? In our draft there 
are some open issues. The problem I raised before is one of them. 
Every comment is welcome. 

Best Regards.

Hongbo Shi

From: James Seng <James@Seng.cc>
Subject: Re: [idn] Internationalized PTR draft submitted
Date: Mon, 18 Sep 2000 10:38:55 +0800

> Rick,
> 
> You asking the wrong person. :-) I am not the author of the I-D altho Hongbo
> Shi keep me in the loop while she is writing the I-D.
> 
> I was originally confused about her approach too until she explain to me that
> it is possible you want to have multiples names for a single machine/IP but in
> different languages. 
> 
> It is like saying "okay, I am a Japanese OS and I want the Japanese name of
> this IP if possible, otherwise, I take the next best bet." Her original idea
> was to let the server choose which to return. I suggested that she return
> everything and let the client choose the best they want.
> 
> -James Seng
> 
> Rick H Wesson wrote:
> > 
> > James,
> > 
> > Why must we always talk about the origin "language?" If the domain names
> > have been RACE encoded then they were eventually converted to UTF-16 which
> > basicly covers all "languages."
> > 
> > My meager understanding of RACE and Unicode 3.0 concludes we SHOULD ban
> > the word "language" from our discussion/drafts and replace it with
> > charset.
> > 
> > Since we could have *any* string of UTF-16 encoded chars what point is it
> > to maintain the origional language? Infact if we had the following
> > hostname, which should be legal (U-0F3C U-0726 U-35A6) which "language"
> > would go in the "language" tag? The above are one Tibetan, one CJK and one
> > unknown....
> > 
> > Please educate me on the utility of maintaining the description "language"
> > or charset for that matter, once you're in UTF-* you should be able to go
> > to any other charset!
> > 
> > -rick
> > 
> > On Mon, 18 Sep 2000, James Seng wrote:
> > 
> > >   example of an IPTR RR:
> > >
> > >    1.2.3.4.IN-ADDR.ARPA.    IPTR  "language" "name-in-utf8"
> > >
> > >   [RFC1766] describes the ISO 639/ISO 3166 conventions.  A language name
> > >   is always written in lower case, while country codes are written in
> > >   upper case.  The "language" field in an IPTR RR MUST follow the con-
> > >   ventions defined in [RFC1766].
> > >
> > >   For Example:
> > >
> > >    4.3.2.1.IN-ADDR.ARPA.            IPTR     "zh-cn"   "name-in-utf8"
> > >    4.3.2.1.IN-ADDR.ARPA.            IPTR     "zh-tw"   "name-in-utf8"
> > >    4.3.2.1.IN-ADDR.ARPA.            IPTR     "ja-jp"   "name-in-utf8"
> > >    4.3.2.1.IN-ADDR.ARPA.            IPTR     "ko-kr"   "name-in-utf8"
> > >
> 
>