Thx! It's really meant for subseq. execution of the same file but you can also compare different versions if the chunk name and it's variable names remain identical (so somewhat limited). We chose JSON because it's perfect for git. If you rename the file, you would have to rename the meta data files