Maybe I'm a sucker for this kind of thing, but I like the Times Word Nerd – it is a very simple tool that scans your tweets and gives you a score.

For me, it is:

BUT, more importantly, Word Nerd is also helping push users to Times writers – see right side of the page . . .


There is a science behind Word Nerd, according to the Times . . .

‘Word Nerd’ tests the distinctiveness of the words you use on Twitter. It works by scanning your last 100 tweets, and comparing the words you have used with those stored in Google’s immense NGram database – a collection of several hundreds of thousands of words Google has gathered from its Books project.

Foreign words, non-words (like ‘Lolz!’), and ‘stop words’ – words like ‘the’ and ‘and’ that are used so frequently by everyone as to be irrelevant – are first removed from the list.

A custom algorithm then calculates the distinctiveness of the words you’ve used by examining how frequently each appears across the books from which Google’s NGram database is derived. (A word that appears more frequently scores lower, and vice versa.)

You are given an ‘overall score’ based on the distinctiveness of each individual word in your tweets, and ranked in one of four ‘performance quartiles’.

It seems to be a clever idea to attract readers to the Times and to its writers/columnists, as Jennifer Whitehead points out on the Wall blog.

Seems like it is taking off . . . 20,000-plus have taken it so far . . . or maybe a few hundred people keep trying to raise their score. Watch out for wordy tweets!

Word nerd 1