Unfortunately time is one-dimensional, which creates a demand for ordering (I wouldn't be surprised if all demand for order actually came from that fact). I may not need ordering in 3D space, up until the moment I need to grab a few things - then I have to decide on an order, because grabbing is sequential.
This has fundamental implications, as many activities do have this cyclical parts (see also such decomposition for vector fields, https://en.wikipedia.org/wiki/Helmholtz_decomposition).
Vide Reinforcement Learning for StarCraft 2: https://deepmind.com/blog/article/AlphaStar-Grandmaster-leve...