Topic: PoS is far inferior to PoW - why are so many people advocating switching to PoS - page 6. (Read 12865 times)

With PoW you need to invest to "get into the money flow", the same holds true for PoS.
PoW and PoS have a tendency to favor early adopters assuming that a coin is successful.
No real difference, is there?

One of the advantages of PoW over any PoS variant is that it encourages innovation.

New generations would be welcome to come up with advancements in science, new technologies and even attempt to crack SHA256 or Scrypt in order to gain an advantage with PoW systems and get into the money flow.
PoW encourages hacker mentality (in a positive sense), with PoS however, if you don't own a stake in an entrenched system, you are out of luck.

Would you like old, filthy rich psychopaths to run the society in the future?
Choose PoW, protect the children!

Agree, EC involves some level of trust (economic laws), not only math.

Not that big a deal though - NXT-philosophy is not a one-to-one copy of Bitcoin-philosophy.

Bob has a client that only downloads valid chains.
He sees 99 nodes with invalid chain and 1 node with valid chain (more stake involved).
Which one does he download?

My node knows which chain is valid, the one with more stake involved.

You need to build a valid longer chain without majority of stake. Which, as we are discussing, requires extreme amounts of computing power (calculations pending).

Please expand.

Of course..

But since I now have bots running half the network, any NEW user would download MY CHAIN.

And here's the kicker, they would have NO (trustless) WAY of knowing which of us is pushing the VALID chain.

As for Economic Clustering.. Just 'Trust' wrapped up in a more complex way..

 

Nxt will have a concept known as economic clustering, you join the chain with the largest economic cluster when you first join the network. But it won't be included before the Christmas so we can discuss in future just prior to release. Or there is an Economic Clustering sub forum in the Whitepaper section of nxtforum.org.


As discussed above, the history attack (have an old account(s) with a majority of stake and rebuild a better chain) doesn't work (in Nxt). As I said before, Nxt security is layered and interlocking (not all elements active at the moment) and I am sourcing and checking a complete answer for Dumbfruit above. The approach is too simplistic to work.


Come-from-Beyond is happy for everyone to sell their historic passphrases as this is not a threat. N.B there are no anti-clone traps in Nxt anymore, the post is from May.

then my node would ignore your chain because it's invalid.

I'm sorry - It's not.

My INVALID chain has just as much stake put into it as the VALID one.

I bought accounts that are empty now but were full of coins sometime in the past.. (Or hacked an exchange to get the keys to their now empty wallets etc..)

Well - in a POS chain, you see which chain has the most STAKE put into it. This is independent of anyone. You just sum it up and choose the one with the most.

simple as that.

Let's try a different attack..

When you first connect to a POS network you need to download the current valid chain.

If 2 Chains are presented to you what happens ?

Well - in a POW chain, you see which chain has the most WORK put into it. This is independent of anyone. You just sum it up and choose the one with the most.

In a POS chain, there is no way of knowing which chain is MORE valid than the other. You need to TRUST someone.

If I run a bot-net, that behaves absolutely correctly, so that it takes up 'Half' of the valid nodes in a network (EASY - even bitcoin is only 7000 nodes), for months, I can then launch a nice attack.

My assumption is that either I HAD - years/months/days back - an account with a large stake in the POS network but now do not or I buy the key off someone who had a large balance in the past - maybe from one of the 77 initial NXT stake holders.

I push out a chain, that I generate from way back when (Splitting it up etc etc to mask the hack), where everyone has the same balance (So you won't mind) EXCEPT for 2 or 3 large accounts where all the money has been given to me.

My bot net accepts it as valid and pushes it out as the correct chain.

The solution, for normal users, is to check with HEAD OFFICE - say the main NXT nodes - which is the valid chain, sure, but this requires something distributed consensus is supposed to remove. Trust.

Again - I like POS - it's just not ready for primetime yet, and worse, we still have no idea it ever will be.. (trustless)

If the value and transaction volume is low, not much is lost if somebody attacks it. That is why I am not too concerned that it "only" costs about $300 Million to attack the Bitcoin network. If I want to be perfectly honest; Bitcoin has very small adoption at the moment. Bitcoin is not that important as a payment processing network in the scheme of things.


I think Dogecoin was simply a fad (that lasted a surprisingly long time). I will, however, be following Monero.

I understand it is not an easy thing to do as it must be hard to figure out but easy to verify. However, I think Primecoin has proven that PoW processing can benefit society in some way by advancing mathematical research by finding long chains of prime numbers. I have faith that a "better" use case will pop up sooner or later.. surely finding chains of prime numbers is not the only use case of useful PoW.

I agree, however PoS coins can be diluted as well via interest with serves a similar purpose. The only truly deflationary coins I have seen are PoS, as with PoW the miners need to be paid. Personally for a store of value, I would prefer a truly deflationary coin.

I'm not sure I agree with this. It is possibly a deterrent from those thing happening, but it would not affect someone that was determined enough to do so and had deep pockets.

Not necessarily. That's why it's called supply and demand, and not simply supply.

In some versions of PoS I agree with this point.. when stake is generated proportional to the amount of coins you own. However, in DPoS (and possibly others) this is a non issue. All stakeholders benefit equally from securing the block chain as no interest is paid. Fees are destroyed in the PoS process, thus increasing everyone's buying power equally. It is one of the few truly deflationary coins I have seen. Bitshares just implemented dilution.. but it is so the block chain can hire employees and improve upon itself through funding development, marketing, etc.. otherwise DPoS has the potential to be truly deflationary as it has been since its launch.

Yes, it may not even happen in our lifetimes if at all, but I see it as being a possibiliy. There are some ALT coins that have quicker emission curves, so we will see how they are able to deal with it. We all know how Dogecoin went.. Monero is another one with a fast emission curve. If the crypto markets crashed really hard it could happen sooner than you'd think. I realize a PoW coin can switch later, but I think it is important that PoS be tweaked and improved upon now rather than later. That way if Bitcoin does have to switch it can switch to the best possible solution that has been "battle tested".

I'm not sure what you're getting at here. I think it depends on the value of the PoW coin on the free market and the amount of transactions that are occurring on its block chain. If the value isn't high enough and there aren't enough transactions then the network will possibly not be very secure as people will shut down their equipment. Someone could probably buy up that equipment for pennies on the dollar and mount a 51% attack on the less secure network. There are a lot of entities that would have incentive to do this.. banks, central banks, competing cryptocoins, governments pissed at dark markets and tax evaders, etc..

PoW coins can not use computations that are otherwise useful. If the computation is otherwise usedful, it can be pre-computed: weakening the security PoW is supposed to provide.
This dilution helps spread coins around to new users. It lowers the barriers to entry if some organization manages to take control of the majority of the coins/hash-power for some reason. Put another way, scarcity will make the price rise, which will encourage somebody, somewhere to plug in a miner. In a PoS coin, I am not sure why anyone would sell enough stake to lose control over the "forging". The whole point is that stake costs almost nothing to maintain.

I agree the question of miner fees for securing the network is an open one. As somebody in another thread pointed out: we can switch to PoS if it becomes clear that fees are not sufficient to secure the network. That won't happen for decades.

Edit: Ideally, once coin creation ceases, mining fees will settle on an amount sufficient to secure the network. It is not clear that there is any stable feed-back mechanism to ensure this. Limited block-size may have to act as an approximation.

kokojie nails it on the head.  A POW 51% attacker WILL attack a coin to steal and double spend  destroying the coins value and move on to another coin to attack, steal from and destroy.  But this will practically never happen to a POS coin because the attacker would be attacking his own money that he already owns and would be destroying the value of his own wealth in the process.

Add in the fact that POW is so obscenely wasteful for zero benefit when compared to POS and one begins to realize the evolutionary advantage of POS mining vs POW mining and why more and more Bitcoin devotees have been seeing the benefits of POS.

You are missing the point of PoS. The point of PoS is not to be better at achieving decentralized consensus than Proof Of Waste. The point of PoS is to achieve sufficiently decentralized consensus without wasting a ton of electricity, therefore harming the environment, and/or wasting a ton of processing power that could be used to benefit society in one way or another. If you insist on PoW, then make the PoW beneficial to society like Primecoin or curing cancer... otherwise it is a waste of energy and processing power.

Furthermore, it is unclear whether Proof Of Waste will work once the block reward halvings stop and no more currency will be printed. Currently PoW coins are diluting their shareholders to pay for miners to secure their block chains. The only coin that is far enough along in its emission curve (Dogecoin) needed to switch to a merge mine coin because no one was mining it. If it is not profitable to secure the block chain then no one will and it leaves the coin wide open for attacks. If the biggest PoW coin's values don't rise enough so that the transaction fees alone will pay for miners to secure the block chains, then they will need to eventually change their consensus algorithms or be left wide open to attacks. In other words, it may be inevitable that all coins eventually have to switch from Proof Of Waste, so that is another reason why it is important that PoS exists and people are altering and improving upon it.

What I wrote about Bitshares was obviously not some security analysis. I was demonstrating the clear overwhelming complexity that PoS brings in order to "solve" a problem that is solved very well and very simply with PoW.

Millions of dollars in electricity is wasted every day to support all PoW coins. The main culprits being Bitcoin and Litecoin, obviously smaller chains use less energy and processing power. If all PoW coins were to switch to a Myriadcoin-like approach to PoW, it would not change this fact and the problem still exists.

BTW- although I have touted Myriadcoin as being an improvement to PoW in general, I have never touted it as a good investment. There are a lot more dynamics than simply what consensus algorithm it uses as to the price realization of a cryptocurrency which is traded on a free market. Features, inflation, community, developers, development goals, innovations, marketing, security, adoption, etc... all plays into price discovery. All factors need to be considered when deciding which coins to invest in.

This is not accurate. All decentralized consensus algos are vulnerable to 51% attacks whether it be PoS or PoW. Someone with enough hash power to 51% a PoW network can do so forever, or until the PoW network forks and changes PoW algos. Someone with 51% stake of a PoS coin can 51% the coin forever, or until the PoS network forks and burns the attacker's stake.


It is waste though in that it wastes electricity and processing power unnecessarily to achieve decentralized consensus. As to your price examples... purely PoS coins havent even been out a full year so it is too early to make those comparisons, but I speculate that it will be no different than PoW coins. Bitshares for instance has increased in value this year, whereas Bitcoin has gone down in value. You can cherry pick data points that support your argument, but it doesn't necessarily mean what you are saying is true.

PoW or not, it doesn't use that much electricity now, it's basically useless... You can relax knowing that less than 200 USD worth of electricity is wasted daily to sustain this coin.