Bugzilla seems to change the 8-bit special characters to 7-bit. E.g. 'a' with two dots --> d. See bug #12250 for an example.
Why not make bugzilla UTF-8 compliant ?
Bugzilla is not 8bit clean. The problem is really, how does bugzilla know what encoding you're using to submit a report with? And how does it know what encoding it should use to display the report to another user? Both are probably solveable, but I don't think it would be an easy task per se. (or someone would have likely done it by now.) ;o) If bugzilla did use UTF-8, then how would someone using ISO8859-x to file a bug report end up with their bug being treated? Garbled as UTF-8? Not sure, but this sort of thing is more of a major bugzilla feature request that is probably best to as a request to bugzilla.org
*** Bug 15303 has been marked as a duplicate of this bug. ***
Discussion in bug 80349 seems to indicate that this is fixed now.