Re: [idn] How should labels be encoded?

["Adam M. Costello" <amc@cs.berkeley.edu>]

Dan Oscarsson <Dan.Oscarsson@trab.se> wrote:

> Which labels are "host names"?
> The SOA record contains one that is an e-mail address.

Not a problem.  RFC 1035 tells how to represent an email address as
a domain name:  Prepend the local-part as a single label to the mail
domain.  IDNA affects the mail domain part, but not the local-part.  If
and when a working group forms to specify internationalized local-parts,
they will decide how to handle it here.  Until then, local-parts follow
the same rules as they do today, so there's nothing new to worry about
here.

> The SRV record has an owner name that represents a protocol.

Sorry, I'm not familiar with SRV, but perhaps the same argument would
apply.  Until some other working group specifies internationalized
owner-names (whatever those are), there's nothing new to be done here.

> Also not every thing which has an IP-address is a "host".

I'm not sure what you have in mind.  What's an example of something with
an IP address that's not a host?

> As DNS is more than "host names", how are labels that are not "host
> names" to be encoded when non-ASCII is used?
>
> IDNA does not answer this and it does not define which labels are
> "host names" and must be ACE+nameprep encoded.  It can happen that
> one application thinks a label in a DNS record is a host names, and
> another application uses it as something else (using some other
> encoding that ACE+nameprep).

IDNA says how to represent the labels of domain names.  It has nothing
to say about how to represent labels that are not labels of domain
names (that would be out-of-scope for IDN).  Not every sequence of
labels is a domain name; a domain name is the name of a domain.  When
amc.cs.berkeley.edu appears in an SOA record, amc.cs.berkeley.edu
is *not* a domain name, because it does not name a domain.  It is a
sequence of labels, the first of which is the local-part of an email
address.  The remaining labels form a domain name, and IDNA would apply
to them.

Maybe this is not clear enough in the IDNA draft, but I have no doubt
that this is the intention.

Maybe part of the confusion is that RFC 1035 uses the notation
<domain-name> to mean any sequence of labels, even when it's not really
a domain name.

AMC