Quintin Pope comments on DL towards the unaligned Recursive Self-Optimization attractor