Re: [idn] What's wrong with skwan-utf8?

["Edmon" <edmon@neteka.com>]

Does anyone have a good list of "what happens" if an full 8-bit name is
received by the different applications?
The reason we should strive to compile this list is that once IDN is
deployed, there will be people trying to enter multilingual domain names
from non-compliant applications.  There is simply no way of avoiding this,
even if we choose to go with ACE.

If something is bound to break or go down, we should know about it
beforehand so that we can prepare for it.

Edmon


> At 16.44 -0800 00-12-24, Rick H Wesson wrote:
> >There are also a lot of of protocols that
> >expect name parts to be in 7bit ASCII, SMTP and SNMP come to mind, the
> >latter being one that is often found in embedded systems running critical
> >network infractstructure.
>
> SMTP, SNMP, IMAP, POP, HTTP just to mention a few.
>
> They can be updated of course (and I think that is your point). The
> Applications Area have created a small task-force(!) which is to
> review the protocols in question to see what can be done with them to
> see that they are internationalized. I.e. not only the domainname is
> to be internationalized but most certainly other protocol elements
> which are displayed to the user aswell. One example is the localpart
> in email addresses.
>
> We have to differ between implementations and protocol specifications
> here. Many implementations of the above protocols happen to be able
> to handle UTF-8, while others can not. Regardless of this, the
> protocol spec have to be updated.
>
> Also, there is a question whether UTF-8 is really what we should use.
> UTF-8 is not much more different than any ACE encoding because it is
> just yet another encoding of the 32 bit characters which is what many
> people feel we should start to use instead of the 7-bit which we use
> today. So, the transision is not from 7bit to 8bit, it is from 7bit
> to 32bit clean transport, and that is a completely different issue.
>
> The problem with skwan-utf8 is that it doesn't talk enough about the
> implications on the application layer protocols which use domainname
> protocol elements. I.e. IDN is not a DNS problem. It is an
> application layer problem. DNS is already a binary transport, so if
> this was only DNS, we could say asap that we should send 32bit
> characters in the DNS protocol. DNS doesn't have to be changed.
>
>     paf
>
>
> --
> Patrik Fältström <paf@cisco.com>       Internet Engineering Task Force
> Area Director, Applications Area                   http://www.ietf.org
> Phone: (Stockholm) +46-8-4494212            (San Jose) +1-408-525-0940
>         PGP: 2DFC AAF6 16F0 F276 7843  2DC1 BC79 51D9 7D25 B8DC