I try to base my view on IDNs on the lessons learned back in europe where I worked for a quite large domain registrar. First of all, I am sure that IDNs are the right way to integrate as many ethnic groups in the development of a digital world as possible. Or, to turn around the point of view, to lower the barrier for access as much as possible.
If we take a look at the developments in the ICANN we see that internationalization is a big topic. For ICANN the internationalization of the root (.test) is one of the biggest running project at the moment. This hopefully leads to a broader acceptance and better support in all kinds of software.
If we look at the current usage of IDNs however, the picture is diverse. We had a big demand back in 2004 when IDNs got introduced in Germany, Austria and Switzerland. Out of my head we registered around 25.000 .de IDNs on the first day. A quick review of the current figures show that there are just around 21.000 left. Normally we grew quite good but the IDNs went back due to lack of support in mail clients and in IE (biggest drop was before the second renewal).
I made a quick query over our database and came up with the following registered IDNs per TLD:
the figures are based on a total of about 900.000 domains. As one can see easily, the number of IDNs is really a small part. Nevertheless there are some IDNs in advertisements in europe and they gain acceptance in the public slowly.
The current figures for some european registries:
.de 11.335.201 (376.534 IDNs)
.ch 1'015'576 (~33.000 IDNs)
.at 785.536 (105.899 IDNs)
The nic.at had a promotion to push the IDNs where they gave away the IDNs for free the first year. Interesting here is the amount of domains that get renewed next year.
I think that IDNs are important even if in new zealand the number of registered domains will never be a significant one.