Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Add in the word 'tweet' and 'retweet' and you're sitting at around 3% of all tweets (give or take). For reference, this is about 5-10 times higher than any other common term I checked (facebook, google, internet, car, obama, etc.) For futher reference, the words 'work' and 'new' both come in around the same amount (3-4%)


Yes, but look at how the word "tweet" is used. Calling it self-referential is a bit of a stretch, it just means "message" within twitter.


Better, it's like saying "FWD" in email. It's just meta-data.


It's a silly test to begin with. It doesn't take into account shortened URLs which are a big part of the Twitter discussion. It can't take into account phrases like "Democratization of News" or "New Media". It doesn't even take into account stemming ("tweeet" has a dramatically different graph than "tweeting")

It just doesn't prove a thing




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: