Again courtesy of Dare, a post about spelling that is deeper than it seems. He's looking at the use of Bayesian analysis by Google, specifically in the spellcheck department. The question it really raises is who defines normal, and if it's the "wisdom of the crowd," how short is the race to the bottom? My teeth are set on edge by the encroachment of test message spelling into email and other documents, and I now fear the same will be true in the use or misuse of the language. For the record, just because "everyone" spells something wrong, it's not right.
And now that I sound like my Dad, I'll log off for the night...