[wgj]

I answered myself on this other thread. http://news.ycombinator.com/item?id=913284

Every other part of the URL including the rest of the domain name can already be Unicode encoded. This is how spoofing attacks are formed by using non-Latin characters in domain names that look like the Latin characters they replace.

This latest ICANN decision doesn't seem to make anything less convenient for the English speaking world than it already was.

[thaumaturgy]

Yeah, this is the biggest problem with the proposed change to URLs. When you can have two different characters that are visually identical but have different meaning to computers, you have a serious security problem.