I might be biased because I recently spent a couple days wrestling with it, but I think the Turkish case is more interesting. There are two vowels that use the letter I forms:<p>İ/i and I/ı<p>So the lowercase and uppercase are essentially split into two letters and then a new uppercase and lowercase form is created based on growing or shrinking the originals. I'm sure it seemed quite the elegant solution in the 1920s when they were working on the latinized Turkish alphabet.<p>The implication of this is that unlike the Dutch case, this affects all text-transform actions (uppercase, lowercase, and capitalize).<p>When we were doing our Turkish localized site I started digging into this, and I was horrified that no browser actually supports proper Turkish capitalization rules. Firefox in particular had a bug open since 2004 (now finally fixed as well). Asking around with Turkish web developers I heard of some crazy hacks (custom fonts!), but I got the feeling that Turkish web designers just avoid text-transform. This was not an option for us as we rely heavily on text-transform in our design (<a href="http://tr.mubi.com" rel="nofollow">http://tr.mubi.com</a>).<p>In the end I was able to piece together a surprisingly robust javascript replacement method with some help from Stack Overflow:<p><a href="http://stackoverflow.com/a/8743095/8376" rel="nofollow">http://stackoverflow.com/a/8743095/8376</a>