They changed from the previous version because they figured out how to have ten times as many shards, lower the minimum stake from 1500 ETH to 32 ETH, and remove code duplication that the previous version would have required.

It's incorrect to say they "can't figure out how to make it work." They figured out one way, then a much better way. Right now they have a near-complete spec; the remaining changes are small optimizations. Eight independent teams are working on implementations in various languages.

That's incorrect. There are plenty of chains with pure PoS, and the reason it's taking Ethereum so long is because they have an established PoW chain that they are trying to seamlessly transition to PoS. At the same time they are also working on taking into consideration more aspects like how it works with their platform/smart contracts and sharding. I'm sure there's many other considerations I'm not aware of as well. All other pure PoS implementations are much simpler as they're just simple payment blockchains which greatly reduces the difficulties and moving parts.

I haven't seen a pure PoS chain and I'm pretty heavy in the space. I HAVE seen a LOT of dPoS chains, which isn't the same.

Have u seen Qtum?

Off the top of my head there's PIVX and Blocknet.

Neither of which are pure PoS.

I think the point is that all cryptocurrencies that we have currently are experimental in that sense, and so it's a good time to experiment with unproven concepts. Maybe that means ETH will crash and burn, but so what? It's a minuscule part of overall economy, so aside from people who decided to seriously invest into it (i.e. alpha testers), it's not a big deal. But, on the other hand, we do want to figure out what works before we start using these things at scale, when it's much harder to change.

This news is good for Bitcoin.

Is Tezos not pure PoS?

Tezos uses delegated PoS, so no, I don't think it qualifies. D-PoS is sort of antithetical to the idea of decentralized blockchain due to a small number of nodes and very questionable process of node (witness) selection.

Incorrect.

1. The number of nodes is not fixed.

2. The process for node selection is random, based on the amount staked. This happens continuously.

3. The D in DPoS is entirely optional. Anyone holding Tezos can directly participate by running a node. The only cost is the cost of a VPS.

The only real restriction* is a 10000 XTZ minimum for running a node. At current / ICO prices that's around $5k. While that's not cheap it's a lot more attractive than the CAPEX/OPEX of mining hardware.

Also, that $5k will grow by 5.5%++ per year, i.e., no depreciation of hardware and only a trivial amount of overhead.

Those not wanting to run a node (or with less than 10k XTZ) can delegate to a baker. There are a wealth of them available already.

*This restriction may be lowered in the future -- by way of on-chain voting (as opposed to a hard fork).

Good points, though I'll just clarify that 10000 XTZ is only required if you want your node to produce blocks.

Some numbers regarding decentralisation: The last couple of (3-day) cycles the number of unique blockproducing nodes (bakers) has averaged just above 200 pr cycle, and is steadily increasing. The Tezos foundation nodes are now down to producing ~25% of the blocks.

I recommend this post for anyone interested in Tezos 'Liquid Proof of Stake'-model: https://medium.com/tezos/liquid-proof-of-stake-aec2f7ef1da7

Won't any pure PoS system naturally evolve stake pooling that looks like DPoS?

In Ethereum there's one disincentive to stake pooling: if a staker misbehaves, perhaps because it got hacked, the penalty depends on how much stake misbehaves at once. A large amount of misbehaving stake could be completely destroyed, while a small amount will lose only a small portion.

This is mainly because a small amount does no damage and is presumed to be accidental, but also helps decentralize since you're safest not using a large staking pool, the most popular client, the most popular hosting service, etc.

Running as a small staker is feasible because the minimum stake is only 32 ETH, you're profitable if you're online at least 2/3 of the time, and the load on an individual node is feasible for a laptop. The computational load increases fairly linearly with the amount of stake, so there's little economy of scale for a large staker.

There are two attractors, it could push people to do it themselves but I think it's more likely that it evolves in stake pooling. Stake pooling in pure PoS involving trusting the pool operator to a great deal and I'm concerned it degenerates to a very centralized system. Adding optional delegation is a safety valve that takes this into consideration.

What Tezos does is merely let you lose a different key to custody funds and to create blocks. By itself this creates the possibility of delegation.

Last but not least, with trusted hardware the difference blurs even more.

Just my personal opinion, but designing a system that doesn’t allow for secure pooling will only enable insecure pooling. So making the choice to force people to share private keys as the only way to delegate capital will make the overall system a lot less secure, and just seems really irresponsible.

Delegation does not require sharing private keys. Nor does it require transferring XTZ to anywhere other than your own wallet. Ever. Period.

The only risk associated with delegation is that the baker will run off with the baking rewards instead of distributing them. Of course, the incentive is low since this can only be done once. The community is active is discussing the trustworthiness of bakers (which, to date, has been pretty much all of them).

The staking risk (risk of losing a bond) is entirely on the baker. This incentivizes them to make sure they're not double baking or otherwise running misbehaving nodes.

My point exactly. Without delegation the only way to pool capital is to give up ownership.

You need to control 2/3rds of the stake and be able to coordinate decisions with each staking node before the block time is over to fuck up the network. Considering that PoS can enable more tunable block times as validators no longer will churn out random numbers on GPUs/ASICs to mine a new block, I think this will be a non-issue.

Unless they somehow integrate PoS in the wallets, yes.

Looked more into it, this seems to put Tezos somewhere in the middle. https://medium.com/tezos/liquid-proof-of-stake-aec2f7ef1da7

However after reading that I am pretty sure the current Ethereum Serenity spec allows for delegation as well, so it seems to be on the same level Proof of Stake

Did you just describe Bitcoin miners?

If you mean mining pools, then no. Yes, they have consolidated power, but their selection process still relies on external resource (i.e. energy).

XTZ, RDD, there are numerous other POS coins.

The ETH network doesn't want to make it work.

Blackcoin? That was years ago.