I have just spent the whole morning battling with a bug in an RTF encoding class in one of our applications. For some reason, Unicode characters were causing Microsoft Word to either consider that the file was corrupted or else to drop characters seemingly at random.
It turned out that the reason for the bug was that I was using some incorrect information on Wikipedia. Now I must confess to being a bit of a Wikipedia fan, but it turns out that this particular article omitted some important information, as I discovered on a closer reading of the MSDN documentation.
Since anyone can edit Wikipedia, I have corrected it, but it just goes to show that if you rely too much on the information that you get there, you can run into trouble.