Re: [idn] Debunking the ACE myth

["Eric A. Hall" <ehall@ehsco.com>]

> That's not what most of the proposals at hand require. They require
> that any name server that might do IDN queries be upgraded.

This is true only if the operators of that domain want to provide UTF-8
IDNs to their users. If they are happy with the level of service provided
by ACE, they can stick with it.

NB: the choice is ACE or ACE-with-UTF-8. The operators do not have to
upgrade if they don't want to. No upgrades are *required* (the word you
used in your comment) by either the operators or end-users.

I'm not sure I understand where the confusion on this point comes from,
nor why it is proving to be so hard to dispel.

> This probably also includes every DNS server that gets hit a
> recursive query (even if that server has no IDNs itself), such as the
> root servers and all the TLD servers.

This is accurate. Every 3LD has the authority to create their own IDN
labels, so every TLD must be able to process the resulting queries.

However, it is important to point out that most of these servers are
upgraded frequently, so the threat of lag is not very high. The only
question would be if the TLD operators want to support UTF-8 IDNs or not
(very probable, IMO). We can certainly ask them if they would refuse to
implement such a technology if it were the standard.

-- 
Eric A. Hall                                        http://www.ehsco.com/
Internet Core Protocols          http://www.oreilly.com/catalog/coreprot/