[Intermernet]

No problem.

Out of interest, do the Firefox team and the Chromium team compare notes on decisions like this?

Purely in this one area (IDN homograph attacks), it might be an idea to look at the Chromium Unicode vetting rules (Which characters and combos get "punycoded") as they seem to be more conservative from a "Latin" perspective.

I'm not sure if a "blacklist" (mentioned in the bug report) is the best way of handling this. Perhaps only direct-encoding the "exemplar characters" for the language setting, and punycoding everything else? I'm pretty sure it would have eliminated the mid-dot issue, but perhaps this "whitelist" is too prohibitive.