Related: is there a surviving fork that only extracts non-copyrighted content (e.g metadata) ?