PoS is far inferior to PoW - why are so many people advocating switching to PoS

[visual111]

you know, if somebody would just attack a PoS coin like NXT then we could end this debate. seems like it's all just masturbation at this point

[Bill White]

Interesting discussion. I have 2 questions about PoS (as it's implemented in NXT):

1. If transaction fees are the only reward for forging a block, why are blocks with no transactions forged? It would seem sensible to wait until someone makes a transaction before bothering to forge a block.

2. In NXT apparently reorganization beyond 720 blocks is not allowed. Does this make it possible to make an attack that would cause a permanent fork? I can imagine an attacker forming a secret chain with 720 blocks competing with the latest "real" 719 blocks. Suppose the attacker releases his 720 blocks at the same time the 720th "real" block is forged. (The exact second this "real" 720th block will be forged can be reasonably predicted, right?) Half the network might see the "real" 720th block first and the other half might see the "attack" 720 blocks first. It seems like the "real" half would reject the attack chain and the "attacked" half would throw away the latest 719 "real" blocks and refuse to look at the 720th "real" block. This would be a permanent fork.

[BombaUcigasa]

With  PoS coin, on the other hand you can potentially sell your "investment" in the coin prior to your attack being discovered by the rest of the network, which would result in you having a zero net cost to an attacker

Who told you that non-sense?

Me.

Or anyone else with an IQ over 100 whom can infer this algorithmic conclusion.

Assume my IQ being 97ish, could you provide a concrete but simple step by step guide?
I would like to try such a zero-cost attack myself.

Thanks in advance.

Go here: https://bitcointalk.org/index.php?board=159.0
Pick one new coin with PoS that hasn't been launched yet. Find the best options to acquire as much as possible of the coins cheaply (mining, IPO, market). In 2-3 months build your own little network corner with 2+ nodes, keep collecting the coins. Monitor the network for active nodes count (there should be pools and exchanges at least). Scale up your network before the attack to have more nodes than the network (even on the same IP). Power down clients, edit peer list, ban seed peers (maybe edit the source to exclude them). Start staking across your nodes and increase your chain's difficulty while separated from the network. DDoS the fair nodes and release your network to most peers (except seed peers maybe). Be sure to read the rules on chain reorgs and stake limits.

Here are some fancy PoS coins:
https://bitcointalk.org/index.php?topic=852050
https://bitcointalk.org/index.php?topic=843495.0

Here is a calendar:
https://www.altcoincalendar.info/calendar

Here is another guide:
https://bitcointalk.org/index.php?topic=604716.0

[instagibbs]

you know, if somebody would just attack a PoS coin like NXT then we could end this debate. seems like it's all just masturbation at this point

The argument that it hasn't been attacked yet is not a crypto-system argument. (it sorta starts to work after 5+ years of operation as the foremost cryptocurrency, but just barely, and weakly)

I could make a coin today, and make a challenge, and no one would attack, because no one cares. It's up to the creator and their sycophants to demonstrate security through analysis, etc.

The fact that these proposed systems just keep getting more and more complicated to patch known vulnerabilities should tell you all you need to know. PoS may be fixable. It hasn't been fixed yet.

[LiQio]

With  PoS coin, on the other hand you can potentially sell your "investment" in the coin prior to your attack being discovered by the rest of the network, which would result in you having a zero net cost to an attacker

Who told you that non-sense?

Me.

Or anyone else with an IQ over 100 whom can infer this algorithmic conclusion.

Assume my IQ being 97ish, could you provide a concrete but simple step by step guide?
I would like to try such a zero-cost attack myself.

Thanks in advance.

Go here: https://bitcointalk.org/index.php?board=159.0
Pick one new coin with PoS that hasn't been launched yet. Find the best options to acquire as much as possible of the coins cheaply (mining, IPO, market). In 2-3 months build your own little network corner with 2+ nodes, keep collecting the coins. Monitor the network for active nodes count (there should be pools and exchanges at least). Scale up your network before the attack to have more nodes than the network (even on the same IP). Power down clients, edit peer list, ban seed peers (maybe edit the source to exclude them). Start staking across your nodes and increase your chain's difficulty while separated from the network. DDoS the fair nodes and release your network to most peers (except seed peers maybe). Be sure to read the rules on chain reorgs and stake limits.

[...]

Thanks for elaborating.

For a coin not yet launched the initial financial risk (acquire a lot of coins of a coin with maybe no future value plus building a substantial personal network) seems like a no-go to me. What about an "established" PoS coin? Costs of the way described could be astronomical?!

[blackbird307]

The problem with PoS is that someone can easily 50% attack it. Large holders get interest for staking.

[BombaUcigasa]

For a coin not yet launched the initial financial risk (acquire a lot of coins of a coin with maybe no future value plus building a substantial personal network) seems like a no-go to me. What about an "established" PoS coin? Costs of the way described could be astronomical?!

You can reduce costs by trading in wait time. As long as you have some coins and sit on your wallet and stake them, as time goes by major stakers will "trip over" or get out. You can attack the coin.

Now you might think that it costs a lot to buy a whole coin. But in reality, after the top 300 altcoins, it costs less than 1000 USD to buy one completely: http://coinmarketcap.com/4

[bluemountain]

For a coin not yet launched the initial financial risk (acquire a lot of coins of a coin with maybe no future value plus building a substantial personal network) seems like a no-go to me. What about an "established" PoS coin? Costs of the way described could be astronomical?!

You can reduce costs by trading in wait time. As long as you have some coins and sit on your wallet and stake them, as time goes by major stakers will "trip over" or get out. You can attack the coin.

Now you might think that it costs a lot to buy a whole coin. But in reality, after the top 300 altcoins, it costs less than 1000 USD to buy one completely: http://coinmarketcap.com/4

You really should not measure the cost to attack a coin in terms of dollars, but should measure it in terms of the percentage of the value of the coin. You also fail to remember that there is nothing that would stop an attack from selling his coins on an exchange after he has attacked the network

[johnyj]

Why should I buy PoS coin when I can mine them cheaply? Everyone will just mine PoS coin while no one will buy them, the price will be as low as PoS stake holder can hold on an exchange 

you dont seem to understand the first thing about pos?

legendary member they said... knows his stuff they said...

Just give me some motivation that why shouldn't I use the cheapest way to get the coin?

Another problem is, PoS coin consume much less energy, so the cheapest way to get the coin in fact is to start another PoS coin, it is much easier to clone the PoS coin, while you can't clone the bitcoin network hash power

[devphp]

Another problem is, PoS coin consume much less energy, so the cheapest way to get the coin in fact is to start another PoS coin, it is much easier to clone the PoS coin, while you can't clone the bitcoin network hash power

Starting another PoS coin works just as great as starting another PoW coin trying to replace Bitcoin has worked (Litecoin and other failures, maybe only Doge has a small chance to make Bitcoin share some of its success).

Just take a look at all the NXT wannabes (pure NXT clones and those with modifications and even original PoS code cryptos), none has been able to reach the activity and market cap of NXT and that even with all the criticism on NXT's poor initial distribution - people tend to buy NXT regardless (granted all cryptos are in a bear market), because you can't easily clone core developers, infrastructure, network of nodes, vested interest, the spirit of innovation.

[cbeast]

Another problem is, PoS coin consume much less energy, so the cheapest way to get the coin in fact is to start another PoS coin, it is much easier to clone the PoS coin, while you can't clone the bitcoin network hash power

Starting another PoS coin works just as great as starting another PoW coin trying to replace Bitcoin has worked (Litecoin and other failures, maybe only Doge has a small chance to make Bitcoin share some of its success).

Just take a look at all the NXT wannabes (pure NXT clones and those with modifications and even original PoS code cryptos), none has been able to reach the activity and market cap of NXT and that even with all the criticism on NXT's poor initial distribution - people tend to buy NXT regardless (granted all cryptos are in a bear market), because you can't easily clone core developers, infrastructure, network of nodes, vested interest, the spirit of innovation.

Distribution can also be accomplished by marketing. All you need to do is clone a successful PoS coin and market it well. That's why Doge was marginally successful. Doge also benefitted from the discarded mining gear. With PoS you only need marketing. I'll just wait for a major brand backed PoS.

[phillipsjk]

You can't mathematically prove it, because the other stake-holders can roll back to any point of time of their choosing.

That's why I said that PoS 1 != PoS 2 != PoS 3. There is at least one PoS implementation where what you just said is not possible.

The NXT prohibition on deep re-orgs just makes such re-orgs a manual process if/when they they happen.

Edit: Bitcoin proper now implements something similar called "safe mode". This was implemented after the last major Blockchain fork

[romerun]

because mining is bullshit

[devphp]

You can't mathematically prove it, because the other stake-holders can roll back to any point of time of their choosing.

That's why I said that PoS 1 != PoS 2 != PoS 3. There is at least one PoS implementation where what you just said is not possible.

The NXT prohibition on deep re-orgs just makes such re-orgs a manual process if/when they they happen.

Can you explain exactly what you mean by deep manual re-orgs? There is no such thing in NXT. As far as I know the re-orgs are fully automatic, based on the rolling checkpoints 720 blocks deep maintained by each node's software as a distributed consensus.

[phillipsjk]

Can you explain exactly what you mean by deep manual re-orgs? There is no such thing in NXT. As far as I know the re-orgs are fully automatic, based on the rolling checkpoints 720 blocks deep maintained by each node's software as a distributed consensus.

Pretend you are in China and the "great firewall" cuts Chinese "forgers" from the rest of the network for 3 days. What happens when they reconnect with the rest of the world? Obviously, somebody has to force the re-org manually.

[devphp]

Can you explain exactly what you mean by deep manual re-orgs? There is no such thing in NXT. As far as I know the re-orgs are fully automatic, based on the rolling checkpoints 720 blocks deep maintained by each node's software as a distributed consensus.

Pretend you are in China and the "great firewall" cuts Chinese "forgers" from the rest of the network for 3 days. What happens when they reconnect with the rest of the world? Obviously, somebody has to force the re-org manually.

My guess is this would be each individual user's decision to join the other fork after 3 days or stay on your own, fully decentralized. How would this work in Bitcoin if Chinese miners were cut from the rest of the world?

[achimsmile]

Can you explain exactly what you mean by deep manual re-orgs? There is no such thing in NXT. As far as I know the re-orgs are fully automatic, based on the rolling checkpoints 720 blocks deep maintained by each node's software as a distributed consensus.

Pretend you are in China and the "great firewall" cuts Chinese "forgers" from the rest of the network for 3 days. What happens when they reconnect with the rest of the world? Obviously, somebody has to force the re-org manually.

The chinese nodes would be on a permanent fork. Each node owner could still delete his copy of the blockchain individually and the node would eventually download the "right" chain. There is not a single "somebody" that triggers the re-org though.

[instagibbs]

The chinese nodes would be on a permanent fork. Each node owner could still delete his copy of the blockchain individually and the node would eventually download the "right" chain. There is not a single "somebody" that triggers the re-org though.

Pretty sure that's the definition of consensus failure. I guess they think this is ok, and people will naturally(somehow?) decide which fork is "right".

OTOH I call that "broken".

[devphp]

The chinese nodes would be on a permanent fork. Each node owner could still delete his copy of the blockchain individually and the node would eventually download the "right" chain. There is not a single "somebody" that triggers the re-org though.

Pretty sure that's the definition of consensus failure. I guess they think this is ok, and people will naturally(somehow?) decide which fork is "right".

OTOH I call that "broken".

What would happen in Bitcoin in the same situation when you have Chinese miners cut off from the rest of the world and mining their own fork with their PH/s for 3 days?

I guess the difference here is not in Bitcoin's favor. If Bitcoin miners are cut off for 3 days, they can't magically transfer their hardware to some other country and keep on mining there. While in NXT they can rent a VPS for $5/month in some other country (provided it's not a full cutoff of internet access which would be a major disruption of all economic activity) and keep on forging with the rest of the world on that VPS, as well as use it to send transactions. Bitcoin miners would be at a loss in that case. That makes PoS superior as it doesn't have a physical location limitation.

[phillipsjk]

What would happen in Bitcoin in the same situation when you have Chinese miners cut off from the rest of the world and mining their own fork with their PH/s for 3 days?

Since the Chinese miners would still have less hashing power than the rest of the world, their chain will get "orphaned". When their nodes rejoin the network, they should automatically choose the fork with more "work" in it. Many transactions would have to be rebroadcast, and such nodes will likely enter "safe mode" if that happened.

Edit: Somewhat relevant