For those unfamiliar with #puredata, the equivalent in python would be creating a script which, in order to get user input, writes another python script containing a reference to the first, which, when run by the user, sends its output back to the first script. Which then launches another python process (based on the results of the first sub-script), and manages it’s status.

It’s a programming paradigm which makes absolutely no sense in any other environment, but in puredata is surprisingly elegant — a pleasant side-effect of the programming environment and end-user environment being effectively the same.

Btw, this is the first experimental step towards the most exciting project idea I’ve had in quite some time… watch this space!