Actors, athletes and politicians aren’t the only people that should be careful with what they tweet.

IBM is taking every public tweet since Twitter’s inception in March 2006, mapping it to Wikipedia, tagging it with sentiment  (saying something good v. something negative), then digitally archiving it in the Library of Congress.

Which means they’ve seen the best and the worst of your personal tweets (along with those of Charlie Sheen), plus they have been gathering some interesting social data and insights.

So what are some of the nuggets of knowledge they have uncovered?

• 10 tweets per second mention Starbucks
• Lady Gaga gains followers faster than Twitter adds accounts
• Your tweets give away where you live by the words you use
• People are more likely to tweet something negative than positive

In mid 2011 they were saying there were 200 million tweets a day and I assume that number will keep growing. I just can’t imagine how IBM could keep up, nor what the Library of Congress will do with all of them when IBM is done. Sounds like a colossal waste of hard drives to me.

Hopefully IBM will avoid the banal and boring and concentrate on the amazing instant news source that Twitter has become — but who knows? Your tweets about what you did last Friday night may go down in history.

Of course IBM apparently isn’t doing this out of the goodness of their heart or just to add more useless data to the Library of Congress. According to this article, IBM researchers have apparently been using Twitter in order to develop new machines that are smarter than a computer and are able to understand the nuances of how humans use language. More importantly, it seems these new machines will be able to filter the news from the noise.

Now that’s worth tweeting about.