Assuming GP is talking about Von-Mises style criticism of command economies, it's not computational tractability, it's information tractability.

The problem isn't that you don't know whether action A results in N units of X, it's that you don't know if the aggregate preferences are such that it's optimal to produce N units of X or M units of Y.

> I am wondering to what extent the central planning is much more tractable now with the amount of data and computerization. It is not like free market is perfect in adjusting either (i.e. situation with timber or computer chips or containers in covid times). I'm not convinced that central planning is the best way to organize the state, but I'm wondering whether the intractability argument against is still valid or not.

I am personally very grateful to have come across the work of Lynn Foster and Bob Haugen (and a few other parties, e.g. Sensorica). They are working on exactly this problem. Their project is called http://valueflo.ws (their software org is: http://mikorizal.org). One of the projects I'm most excited about is the Valueflo.ws vocabulary on the Holochain distributed app framework: https://github.com/holo-rea. It can basically replace today's centralized Enterprise Resource Planning model with an ecology of fully distributed economic resource protocols.

Lynn did a fantastic presentation; which includes a good intro to their accounting model (Bill McCarthy's Resource Event Agent (REA)): https://www.youtube.com/watch?v=vymAHXGSM14

There is also a great story (using the example of making a pie) in pdf format on the Valueflo.ws website: http://mikorizal.org/ValueFlows-Story.pdf

NY Textile Lab is a project that is working to integrate Valueflows: https://www.newyorktextilelab.com/holochain-app-development

It's going to be highly non-trivial when most of your supply chain (any item maker in the world) is not on the same plan, but instead offers wares at market rate.

The value computation problem that socialist economists attempted remains intractable. Dogma has them clinging to assumptions are fundamentally wrong like such as the very idea of a universal fixed value, or their notion of exploitation.

Let’s say with data central planning is more tractable. What is the benefit? To make more items people should want? I mean capitalism does a pretty good job making things people do want.

And capitalism is highly distributed decision making. As we all know, top down decision making pretty much sucks no matter how much information they might have.

> capitalism does a pretty good job making things people do want

This is a two-edged sword, because it also makes people want stuff that it can make.

Have a good example? I mean, the demand for relatively vacuous consumer goods was pretty strong in communist countries as well.

There was a semi-successful attempt at this in the 70s in Chile, but it was stopped by the CIA: https://en.m.wikipedia.org/wiki/Project_Cybersyn

What an interesting article, thanks for sharing. The photo looks like it could have been taken on the set of Star Trek.

There is a great documentary that came out just a few weeks ago.

Cybersocialism: Project Cybersyn & The CIA Coup in Chile: https://www.youtube.com/watch?v=RJLA2_Ho7X0