The proof of stake you are describing is of your own invention and doesn't address how systems like Peercoin or algorand actually function.
Starting from scratch with a new genesis block? Sure go ahead.
Rewriting any number of blocks is not straightforward due to the use of digital signatures and the cryptographic guarantees that go along with generating and choosing who gets to sign and generate the next block.
>An honest mining majority is a hard requirement of Bitcoin. But proof-of-stake doesn't solve the problem of collusion -- it's even more susceptible to it since it's so easy to recreate valid-looking chains. With Bitcoin, two or more competing chains will at least be visible (two distinct chains with a lot of work), whereas they're everywhere in PoS and the main chain is qualitatively no different from any other valid chain (of which we can create thousands quickly).
I'm not sure what math you are using to base this assumption on. A formal argument by algorand is that the probability of a fork would be the age of the universe. It is NOT straightforward or cheap to recreate valid looking chains and generate the next block creating a fork.
>> In the worst case scenario -- where an exploit spreads through a security hole in the reference implementation, and replaces the blockchain with its own copy, silently assigning the balance of an old, rarely used account to the attacker -- PoS falls on the floor, and proves unsuitable as a base for digital money.
Replace exploit with malicious miner and it's not clear why proof of work is better than proof of stake.
>> Because it doesn't make sense. If a chain, on which a token exists, needs securing, you can't use that chain's token to secure itself.
I mean I'm not entirely sure what this argument is based on? In addition to existing proof of stake systems which have been running for years now, there exist several papers with rigorous formal proofs. Including Turing Award winner Silvio Micali's algorand.
Refute his paper. You will became very famous overnight.
> Mining has huge energy requirements as shown by the article.
>> Mining has no requirements at all. People just follow the best chain, and miners can sell the bitcoins they earn by extending the best chain. It's entirely voluntary.
I assume this is trolling or extreme nitpicking. Securing the network in proof of work requires that the majority of computing power is held by honest miners.
Why not move to proof of stake? Proof of Work mining used in bitcon to secure the network as originally proposed in Nakomoto's paper has huge energy requirements. The result is large energy hungry mines as shown by the article. In addition, there is a centralization of governance by the miners.
One such example of a long running proof of stake is Peercoin. Peercoin is a legit energy-efficient contender to Bitcoin. It's one of the first publicly known implementations of proof of stake has been around for close to 5 years now.
Peercoin is energy-efficient and runs on raspberry pis. It's been running successfully for years now.
The proof of stake you are describing is of your own invention and doesn't address how systems like Peercoin or algorand actually function.
Starting from scratch with a new genesis block? Sure go ahead.
Rewriting any number of blocks is not straightforward due to the use of digital signatures and the cryptographic guarantees that go along with generating and choosing who gets to sign and generate the next block.