The video DSL looks like quite a nice idea. I could see myself using it to keep videos in git: add all the raw source clips once as immutable blobs, then do all

The video DSL looks like quite a nice idea. I could see myself using it to keep videos in git: add all the raw source clips once as immutable blobs, then do all of the editing with text. (The same could be done with shell scripts and ffmpeg, but urgh). I wonder how it fits with the ideas of FRAN (functional reactive animation)?