I understand the attraction of using a podcast episode about Jovo, and using a real podcast to demo the final, complete solution makes sense, but while we're learning it would be easier to work with smaller files that can be conveniently run as a functional test.
(Yes, I know, we can drag the progress indicator to skip to the end of the episode, but FT shouldn't require manual intervention... and if we overshoot the drag we may generate "finished" events before we intend to; I'm not sure whether that will cause issues or not.)