> Has this been thought about? I'm not aware of artificial examples but this is essentially how biology works at the protein level - so it does work at least a

> Has this been thought about? I'm not aware of artificial examples but this is essentially how biology works at the protein level - so it does work at least at the molecular scale. In principle it should also work at larger scales although it would probably be much slower, molecules jiggle very quickly. Programmable magnets (aka "polymagnets") may be applicable to doing it at a larger scale.