There are few topics in technology that are as devious as text handling: Most of the time, we developers got away with «it just works», but as time has progressed, we run into more and more issues where stuff doesn't work and sometimes we don't understand why. In his talk, Philip is going to dig into the depths of character encodings and Unicode to show how the models we have in our minds about text are often wrong to some surprising (and funny) effects. We'll talks about many contemporary languages and we'll show how only a small minority gets it right - if getting it right is even possible.