[db-wg] Puny code or UTF-8 (or both)?

ripedenis--- via db-wg wrote on 13/07/2020 13:12:
> There has been support shown to introduce Puny code as a first step 
> towards internationalisation of the data, which can be done quickly by 
> the RIPE NCC.
> 
> Do we therefore have support to introduce Puny code now and then 
> consider how to move forward with UTF-8?

re: punycode, there's nothing to stop anyone using the ascii form of 
punycode for email addresses today.  So we need to ask what we're 
actually supporting here.  Is this a simply that the whois server will 
automatically normalise all email addresses to a specific format, or are 
we talking about something more extensive than this?

Either way, we also need to keep an eye on the road ahead.  I.e. if 
there is work planned for future native utf8 support, do we aim for 
using utf8 as the normal form now, and ask the DB people to put in a 
utf->punycode translator for whois presentation.  Does this also apply 
to RDAP / REST API?  There are a bunch of open questions here.

It would be great to get input from the NCC staff about their thoughts 
on this, because I'm sure it's something that they've given a good deal 
of consideration to.

Nick