"more impressive astronomer porn at five times the resolution!" As long as we keep in mind that the filled in regions of space or background match the assumed

"more impressive astronomer porn at five times the resolution!" As long as we keep in mind that the filled in regions of space or background match the assumed entropy, I think it's worth doing. The applications of L1 appear to be limited to cases where the structure and assumptions match the dataset. Maybe there's an underlying pattern we're recreating? Something inherent to physical laws, fractal replication, etc. Your comment helped me to see that L1 isn't just a matter of interpolation but an assumed underlying pattern that is used to fill in or estimate the unknown data. Thanks.