Last Updated: 2009-11-02 01:11:04 UTC
by Daniel Wesemann (Version: 1)
Two days ago, the ICANN authorized the introduction of country code top level domains (ccTLDs) using character sets other than the latin a-z alphabet. This is no earth shattering change - we had Internationalized Domain Names (IDNs) using greek, cyrillic, chinese, etc character sets for several years. The only change is that now also the top level domain (the rightmost portion of a domain name) can be written in characters other than A-Z.
From a security point of view, things might still get "interesting". Back when the IDNs were originally introduced, look-alike domain names and even SSL connections could be credibly faked. Some web servers, firewalls and IDS products also had huge gaping holes as a result of applying their security checks only in ASCII-Land, and ignoring Unicode completely. The past ten years of experience with IDNs have brought the problem reasonably under control, and expanding the IDNs to include top level domains shouldn't be a big deal. But since we all know how software gets "fixed", chances are still that history will repeat itself, and we will soon read of a web server that readily divulges application source code when hit with a TLD in cyrillic...