[troels]

Cool. Some of the words are very exotic, but then some are surprisingly common words. I wonder if there exists an index with a "commonality rating" that could be used to sort/weight the randomiser with, so that one could get more common words up first?

[sehugg]

WordNet has this. You can see it in action here [1] but you have to enable "Show Frequency Counts" in the pulldown. It has frequency counts for each "sense" of a word, not just the word.

[1] http://wordnetweb.princeton.edu/perl/webwn

[impostervt]

Here's where I pimp my side-project, which has a frequency score for words.

https://www.wordsapi.com/docs#frequency

[troels]

Very nice. What did you base the data on? Some 3rd party source or did you compile stats from a corpus of random texts?