Bitmex - Complete guide to Proof of Stake

An interesting article on proof of stake.

In the article they conduct an ‘analysis’ on Peercoin and come to the following conclusing:

At the time Peercoin was an interesting early novel approach, however the proposal resulted in a centralised system, not able to match the properties of PoW.

What are peoples’ thoughts on this?

My takeaways from just reading this article and nothing else:

Peercoin: (1) 2 of the 4 perceived PoS weaknesses/deficits remain: Nothing at Stake and Long Range Attack. And I believe our developers suggest that “Defence 3” is most applicable to the Nothing at Stake criticism and resolves the concern. And the Long Range Attack criticism is probably rebutted by the ability to opt out of central checkpoints and perhaps a misunderstanding of coin age (the 30-90 day probability window of minting a block is not mentioned). (2) PoW still very valuable to have.

Summary: other than Nothing at Stake (which they suggest has no clear, easy to execute fix), the Peercoin criticisms seem overdone.

Ethereum, Casper 2018: PoS seems overly complex and heavily reliant on PoW to make it work. PoS is not integrated as greatly into Ethereum’s protocol as it is for Peercoin, and likely never to be. It’s more of just a check on PoW.

Summary: if Peercoin’s PoW mining added to chain convergence (making PoW a check on PoS), I read this article to imply Peercoin’s hybrid solution would be simpler, used less resources, less centralized, better incentivized, and therefore superior.

This old thread from October 15, 2015 was along the same lines:

Which the myths wiki addresses: