Mapping song-intensity over a single song does give some pretty interesting data points.
Now, I guess, the question is how you handle that over multiple mp3s in a playlist.
Like, the code basically takes max intensity and gives it a ten, but that's one song at a time. One song's MAX might be another song's 1/3rd, etc
I guess the answer here is to analyse all the songs in the playlist and then scale the intensity over all of them?
But then you can't reuse individual data for mp3s.