Via Telepolis ¹:
Eine studentische Projektgruppe an der Universität der Künste Berlin im Studiengang Gesellschafts- und Wirtschaftskommunikation hat jetzt im Rahmen ihrer Diplomarbeit einen Fragebogen entwickelt, der Wissen und Einstellungen zum neuen Urheberrecht testen soll.
Due to Perl’s unicode issues (and Oddmuse’s peculiar charset-agnostic-wanna-be approach), I’ve had a round of problems. Most important point of failure was RSS inclusion, because the XML::RSS module “does it right” (and is *not* charset agnostic). Thus, I’ve had to jump through a few hoops, and I’ve decided to *deprecate* non-UTF8 charsets for Oddmuse. Yes. Painful.
But even then, a good solution escapes me!
If I use the utf8 layer on STDOUT, umlauts will be double encoded. If I use the bytes pragma, RSS feeds will contain Latin-1 characters instead of UTF-8 characters.
Well, actually after some more playing around, it seems as if the bytes pragma does it! Amazing. And my unit tests are being passed as well.
Ok, forget all of that. Hours wasted, going to bed at 6:00 in the morning on my second work-day after a two month holiday... This has to change. :D