3 ms·
Computational complexity was one of 6 reasons the author listed at why the optimization problem was infeasible. I read the article in its entirety, and I saw a
by jkingsbery 6y ago
Computational complexity was one of 6 reasons the author listed at why the optimization problem was infeasible. I read the article in its entirety, and I saw a claim that the number of calculations scaled in polynomial time, but I didn't see any claims of problems that are verifiable in polynomial time with no known polynomial time solutions.
Then there's the other 5 criteria, which show that this is not a computer science problem. Coming up with an objective function is entirely a subjective matter. It depends entirely on what the ruling party wants. Communist countries don't act to the benefit of all, they act to the benefit of the chosen few, just those few are chosen differently.
- masijo 6y agoSo, if the planification was on the hands of the people via direct democracy, like Cockshott proposes, then problem solved, kind of?
- sudosysgen 6y agoAs long as they could settle on a good cost function. I think my proposal for how the cost function should be found is better than Cockshott's in the short term.
- yters 6y agoIf you go to Shalizi's article, which the author links to, then you'll see that polynomial time is a based on very generous assumptions about the problem space of the economy to make it amenable to convex optimization. Under realistic assumptions, convexity disappears, which pushes economic planning into NP Complete and harder territory.