The author has good points, bad points and badly explained stuff. The article is a bit confusing at best and disorientating at worst. But I'll try to explain h

The author has good points, bad points and badly explained stuff. The article is a bit confusing at best and disorientating at worst. But I'll try to explain here, why the author thinks that PoW is magical. It's still bound to the readers, or philosophers, to pull whatever they want from this. Proof of Work creates time. In a decentralized system, you don't have time. If time was provable, the double-spending problem would not happen. You would sign a transaction and broadcast it; a second transaction that you would sign later, will have a higher timestamp. Obviously, you can sign a transaction later and have a lower timestamp, there is nothing that prevents you from that. What Proof of Work does, is create an arrow of time. Using this arrow of time, the nodes create a ledger (the blockchain). The OP is arguing that PoS cannot create an arrow of time; and as a result, the PoS is still liable to the double-spending problem.