Re: [idn] IDNRA comments

[Keith Moore <moore@cs.utk.edu>]

there is a twist in the twist as well.

I could see providing an alternate "raw" interface to DNS for 
queries for non-ascii-encoded IDNs.  but even if that is done
and resolver software is upgraded to use it, the ascii-encoded 
versions of IDNs will still need to be used in other contexts,
such as email headers.

so it's not clear what is to be gained by having a raw query interface,
other than the possibility of slightly different results depending
on which interface is used, due to implementation differences (presumably
bugs).

it sounds nice to provide a raw interface, but I'm not sure it's a win.

Keith