Nothing at stake - debunked?

[kodtycoon]

https://blog.ethereum.org/2014/11/25/proof-stake-learned-love-weak-subjectivity/

[jonald_fyookball]

So basically he is proposing miners need to make security deposits so there is something at stake, which sounds like a clever idea.   I dont know if it will work or not.

What is unclear is what stops a miner from privately working on chains and only broadcast a winning chain with a deposit once the probability is high they won, or at least higher than the normal risk reward ratio that all miners would have to pay with these deposits.

Another question I would have would be:   Wouldn't the miners need to deposit a lot of money constantly?  If the range is 3000 blocks, and you're trying to
vote for every block, that's a lot of blocks that you would need to make deposits for, far into the future.  This discourages voting/mining and centralizes it,
which makes attacks easier.  What would the numbers look like as far as how much capital would be needed to partiicpate in the voting
on a regular basis and what would the risk rewards be?

If it does turn out this scheme makes proof of stake viable the question then becomes: is it fair, is it a good issuance/distribution mechabism?

[Bill White]

Vitalik has written a very interesting article. He distinguishes between preventing short-term forks (using security deposits) and long-term forks.

Preventing long-term forks seems to require giving up on having an objective way to determine the "real" blockchain. However, he considers a notion weaker than objectivity but stronger than subjectivity: weak subjectivity. If a criteria is weakly subjective, then long-term forks can be (more or less) ruled out. An example is NXT's rule of disallowing reorganizations after 720 blocks, but Vitalik describes other possibilities as well. The price one pays is that someone new coming into the network could, in principle, be fooled into believing in a "fake" blockchain. Vitalik's article makes it seem unlikely for this to happen in practice since new joiners would likely be able to find a reliable "checkpoint".

[jonald_fyookball]

Surprised more people aren't weighing in on this.
Perhaps they are digesting it.

Seems like it would be big news either way:
Either Vitalik has a breakthrough concept
that can solve the nothing at stake problem,
or no, it is another failed attempt and PoS
truly can't work.

[kodtycoon]

Surprised more people aren't weighing in on this.
Perhaps they are digesting it.

Seems like it would be big news either way:
Either Vitalik has a breakthrough concept
that can solve the nothing at stake problem,
or no, it is another failed attempt and PoS
truly can't work.

i think its a case where all the people who usual weigh in on this subject are usually the ones who are advocating the idea that nothing at stake is a serious threat to "badmouth" pos. now faced with a technical explanation of why it isnt a threat (in certain implementations) from a very public person who is accepted to be an academic/pioneer in the area they cannot dispute the fact that they are (for the most part) wrong. and more specifically, wrong about nxt which uses 720 block reorg limit as is explained in the paper vitalik has published (understandably he doesnt use nxt as an example seeing as nxt is competition) thus giving credit to nxt for actually doing things correctly as he has actually said on the nxt forum after discussion with the devs.

now no one is willing to admit they are wrong. silence... as someone in on the nxt forum said.. "its like someone farted and everyone knows but no one will talk about it".

interestingly enough, nem is working on a new algorithm based on the concept of pos but instead of just the amount of coins an account has, the transactions an account does, and with who it does the transactions, as well as coin age is used in the calculation for deciding what weight an account has. it takes 30 days for nem to be fully vested before the weight assigned to the received coins can be added to the "active" weight. eigeintrust++ algorithm decides whether a node is "trustworthy" or not based on length of time a node is online and a number of other factors which im not sure of.

it sounds like vitalik is explaining how pos should work, and how nem/proof of importance does work.

jabo from nemforum put it well so il quote him.

These things kind of go over my head, but the way I read Vitalik's article he is advocating our system.   

He states that just counting each coin as a little miner (word for word BCNext's explanation) was flawed because there was no timeout (NXT fixed this because it has a 720 block lock out).  But NXT stops there.  And then he also goes on to explain that is not all that is needed.  He says also that coins needed to be aged in, and also nodes needed a way to trust each other, i.e. the longer a node has been online the more trustworthy it is.  He proposes doing this by giving gravity to an account; that basically the longer it has been known to be online and the more the coins have been aged, then the heavier and more trustworthy the node is.  It seems to me that he makes the argument that PoS needs all three of these.  Or did I miss something?

What he calls "gravity" we call PoI if I am correct.  At NEM we check to see if Nodes are trustworthy through the Egientrust algo used to determine which nodes can be trusted, and we also give weight to coins by how much they have been aged in. 

I kept on reading this thinking, "Wow, he is basically advocating NEM." 

if nem ticks all the boxes that vitalik has put forward which it sounds like it does using only algorithms. i think we could be on to a serious winner.

[disclaimer]apologies for advocating the coin im involved with but it is relevant.[/disclaimer]

[jonald_fyookball]

you're probably right that a lot of people are speechless.

however, here's what's occurring to me:

These security deposits may provide something "at stake",
and so discourage voting on every chain...but they
still don't prevent people from trying to game the system
by solving a better chain in private and broadcasting that
to the network (selfish mining style).  So, it could very
well turn into a kind of hybridized proof of work,
so to speak.

What will happen is that more resources will be consumed
in competition, and as has been pointed out in
recent articles, we will revert back to spending x dollars
worth , same as before, but it will just be
more convoluted...so we may not be saving any energy.

[inBitweTrust]

Interesting paper, but misleading thread Subject .... NaS has not been debunked as Vitalik has repeatedly acknowledged the threat thus a hybrid approach he is suggesting to combat NaS with TaPoS, deposits, and under certain circumstances using a trusted hash.

There isn't much conversation about this as some of us are taking the time to read and study the paper before commenting upon it. Seems like others are prematurely rejoicing feeling vindicated from a long battle against PoW, when they are forgetting this isn't about PoW vs PoS , but merely using the best algorithm for the job that has been tested.

It will take some time to digest the material, discuss the implications, standardize the algorithms, test them either on an alt or side chain before even thinking about suggesting a Bitcoin hardfork.  If this is all that its cracked up to be there will probably be a transitional period where a Hybrid PoW / TaPoS/Slasher 2 gets hard forked over into the system. Vested mining interests may prevent this from happening for some time as well.

[inBitweTrust]

  

Why has this thread been moved into the altcoin section? Probably doesn't qualify as being in the dev section but certainly in the general discussion section as it has nothing to do with a specific alt but a general discussion about new algorithms.

As far as I can see, changes to the algorithm aren't prohibited but disputed:
https://en.bitcoin.it/wiki/Prohibited_changes

[jonald_fyookball]

I agree, it should be discussed and not relegated to the altcoin section

"Nothing at stake" is a big topic worthy of discussion.

[kodtycoon]

Interesting paper, but misleading thread Subject .... NaS has not been debunked as Vitalik has repeatedly acknowledged the threat thus a hybrid approach he is suggesting to combat NaS with TaPoS, deposits, and under certain circumstances using a trusted hash.

There isn't much conversation about this as some of us are taking the time to read and study the paper before commenting upon it. Seems like others are prematurely rejoicing feeling vindicated from a long battle against PoW, when they are forgetting this isn't about PoW vs PoS , but merely using the best algorithm for the job that has been tested.

It will take some time to digest the material, discuss the implications, standardize the algorithms, test them either on an alt or side chain before even thinking about suggesting a Bitcoin hardfork.  If this is all that its cracked up to be there will probably be a transitional period where a Hybrid PoW / TaPoS/Slasher 2 gets hard forked over into the system. Vested mining interests may prevent this from happening for some time as well.

i wasnt entirely sure if it did indeed debunk n@s, thats why i put the question mark. wanted to hear opinions on it from people from the bitcoin section as well as the alt section. anyone in the alt section looks through the bitcoin section but not vice versa.

[e-coinomist]

 

Why has this thread been moved into the altcoin section? Probably doesn't qualify as being in the dev section but certainly in the general discussion section as it has nothing to do with a specific alt but a general discussion about new algorithms.

As far as I can see, changes to the algorithm aren't prohibited but disputed:
https://en.bitcoin.it/wiki/Prohibited_changes

anything threatening precious ASICs...

[inBitweTrust]

i wasnt entirely sure if it did indeed debunk n@s, thats why i put the question mark. wanted to hear opinions on it from people from the bitcoin section as well as the alt section. anyone in the alt section looks through the bitcoin section but not vice versa.

That's all fine , but why is NaS being a threat to certain PoS coins even a question? You cited a paper that is presenting solutions from someone who has done a lot of work in an attempt to find solutions to solving NaS so of course it didn't debunk NaS..... perhaps you meant to say:

"Nothing at stake  - solved?"


BTW... I didn't agree your thread being moved so I created a new one here:

https://bitcointalk.org/index.php?topic=873646.0

[Sentinelrv]

Sigmike and Jordan Lee have already provided a solution to help disincentivize users from minting on multiple chains. It's already coded into NuBits and will be introduced into Peercoin when v0.5 is released. Sigmike is a core developer for both NuBits and Peercoin. Sunny King has already approved the change as well. Here are threads with some more information...

- http://www.peercointalk.org/index.php?topic=3459.msg33474#msg33474
- http://www.peercointalk.org/index.php?topic=3451.msg33402#msg33402
- http://www.peercointalk.org/index.php?topic=3334.msg32393#msg32393

[devphp]

"Nothing at stake  - solved?"

Who would've thunk

[Yurizhai]

Has one of these nothing at stake attacks ever been pulled off to begin with?

[thelonecrouton]

Never mind the technicalities, Proof of Stake is an economically stupid idea. The world already runs on a version of it - take a look outside at where that's got us. While I'm ranting, a lot of the 'features' of the 2.0 currencies that allow the same fuckery that exists in traditional markets are also misguided.

[Yurizhai]

Never mind the technicalities, Proof of Stake is an economically stupid idea. The world already runs on a version of it - take a look outside at where that's got us.

What?

[thelonecrouton]

Never mind the technicalities, Proof of Stake is an economically stupid idea. The world already runs on a version of it - take a look outside at where that's got us.

What?

You earn more money by... having more money. Instead of doing anything useful with it. Breeding more money out of existing money is an ultimately pointless exercise. No service is provided, no real wealth is ever created.

[kodtycoon]

i wasnt entirely sure if it did indeed debunk n@s, thats why i put the question mark. wanted to hear opinions or802n it from people from the bitcoin section as well as the alt section. anyone in the alt section looks through the bitcoin section but not vice versa.

That's all fine , but why is NaS being a threat to certain PoS coins even a question? You cited a paper that is presenting solutions from someone who has done a lot of work in an attempt to find solutions to solving NaS so of course it didn't debunk NaS..... perhaps you meant to say:

"Nothing at stake  - solved?"


BTW... I didn't agree your thread being moved so I created a new one here:

https://bitcointalk.org/index.php?topic=873646.0

that may have been a better title.. Either way, thanks for reposting it, the discission over there is exactly what I was looking for

[Yurizhai]

Never mind the technicalities, Proof of Stake is an economically stupid idea. The world already runs on a version of it - take a look outside at where that's got us.

What?

You earn more money by... having more money. Instead of doing anything useful with it. Breeding more money out of existing money is an ultimately pointless exercise. No service is provided, no real wealth is ever created.

You earn money for securing the network. In most cases it's a trivial amount. People with money can afford to purchase  expensive mining equipment, and the same thing happens, only Earth's resources were wasted to create that hardware and the electricity powering it.

[spartacusrex]

Vitalik has moved the goal posts..

We know you cannot get the same level of 'independent verifiablilty' from POS chains as POW, but he's saying 'Does that matter ?'

Valid point.

POS systems need a 'number' to initialise. That's it.

If you're happy with that, POS away.

[kodtycoon]

Its far easyer to control PoW then PoS all they need to do is stop the destribution of the Mining Hardware just like China is doing at the moment.
Also they can force Internet Provider to Block all Ports related to mining Pools. DNS Main Clusters can just stop accepting Mining Pools.

You can't do this with PoS. If you use a VPN or Hidden Proxy you can be almost invisible.

Pools need always a good connection because of the Dataflow which needs to be very smooth or you will lose Hashrate.
No problem with PoS here again.


You can deny how many times you like, but PoW is more Centralized then people would like to acknowledge this.
The argument of Centralized checkpoints will not really matter if you still check the code for changes and if they matter. Because once the Infrastructure is there it will be really hard to stop it.

Not even the change of checkpoint will ever matter.

rolling checkpoints(decentralised) can be used in place of dev enforced check points(centralised) removing that issue.

[kodtycoon]

Just theorie. Have never been done, even with Scam PoS coins, they had to do some tricks means completely remove the coin check code section to make 51% attack available.

not true. nxt uses rolling check points and vitaliks discussion with the nxt devs is how this weak subjectivity started. they were discussing long/short range attacks and he admitted that nxt has solved the issues he raised and the issues brought up in the weak subjectivity paper he has just posted. so it is by no means theory. its fact.

[solid12345]

The only thing that matters is money talks and bullshit walks.

Guys like BCX like to brag about how BTC and LTC are so secure compared to every other "scam coin" out there but that is because the value of the coins are so high the incentive is there to keep the miners flocking to it like flies on rice.

what really matters is the intrinsic value of a coin, if the coin has any real value it will bring a wide enough net of investors and miners/stakers that security is not as much an issue. And if problems do arise like a pool approaching dangerously close to 51% hash or 1 guy claiming over 50% of a coin, there is no financial incentive to wreck their own investment.

I personally like POS because it rewards holders by paying interest and lets the little guy have his small piece of securing the network compared to the massive mining farms but that's just my opinion. You can make a case for both either way.

Next year true_asset of Urocoin will be launching SPOW which combines staking and POW and to me that will be the ultimate method of both securing the network and also rewarding investors and holders of the coin.

[kodtycoon]

http://www.peercointalk.org/index.php?topic=2656.25;wap2

Rolling checkpoints are generated by every node. Actually it's not checkpoints, a node just ignores blockchain reorgs deeper than 720 blocks from the top block. We need this trick to get rid of a possible attack that was described in the original Nxt thread by Cunicula. Later, when we completely implement Transparent Forging, we'll remove this code coz TF provides much better protection.

As far i see this not the problem. You can't dictate anything if every node uses the build in checkpoints.

hang on i think we are getting mixed up..

you said "Just theorie. Have never been done, even with Scam PoS coin"..

do you mean n@s has never been done? and that the attack is just theory?

they must remove the checkpoint code to make way for TF which would be better security against 51% attacks.. removing the code alone would allow for 51% attacks but if they replace it with TF then it wont be possible.. i think we are in agreement. i thought you meant that rolling checkpoints were just theory.

[othe]

Then why did Vericoin hardfork after Mintpal got hacked...

Fact is, people store their coins at exchanges and exchanges can easily get hacked or coins can get confiscated by 3 letter agencies and with that coins its easy to control the PoS networks, not much to discuss there.

[jonald_fyookball]

But why bother with alt coins and PoS when we have bitcoin and PoW, which
is a proven winner?

[jabo38]

But why bother with alt coins and PoS when we have bitcoin and PoW, which
is a proven winner?

Bitcoin is pretty awesome and is definitely a proven winner.  But it is by no means perfect.  Altcoins are addressing some of the more lacking points of Bitcoin. 

Altcoins look to be faster, stronger, safer, lighter, and prettier, with an additional wide range of added on functions. 

Few are actually doing this, I will give you that, but that doesn't mean that in the future it won't happen.  Instead altcoins are evolving faster and faster and are on track to be just what I mentioned. 

[Daedelus]

Vitalik's claim that POS can be attacked with "1% of the stake even" is been claimed by researchers to be "nearly impossible".


Find the research into the the 'Nothing at Stake' assertions here >>> https://bitcointalk.org/index.php?topic=897488.0

Please comment in the thread >>> https://bitcointalk.org/index.php?topic=897488.0. It would be good if we could get heads together from all aspects of cyrpto in the same place to discuss.

[unent]

But why bother with alt coins and PoS when we have bitcoin and PoW, which
is a proven winner?

Bitcoin is pretty awesome and is definitely a proven winner.  But it is by no means perfect.  Altcoins are addressing some of the more lacking points of Bitcoin. 

Altcoins look to be faster, stronger, safer, lighter, and prettier, with an additional wide range of added on functions. 

Few are actually doing this, I will give you that, but that doesn't mean that in the future it won't happen.  Instead altcoins are evolving faster and faster and are on track to be just what I mentioned. 

If Satoshi reappeared with a new alt coin that he developed to improve on bitcoin I doubt anyone would be asking what's the point of alt coins.

[The Chainmaker]

But why bother with alt coins and PoS when we have bitcoin and PoW, which
is a proven winner?

Bitcoin is pretty awesome and is definitely a proven winner.  But it is by no means perfect.  Altcoins are addressing some of the more lacking points of Bitcoin. 

Altcoins look to be faster, stronger, safer, lighter, and prettier, with an additional wide range of added on functions. 

Few are actually doing this, I will give you that, but that doesn't mean that in the future it won't happen.  Instead altcoins are evolving faster and faster and are on track to be just what I mentioned. 

If Satoshi reappeared with a new alt coin that he developed to improve on bitcoin I doubt anyone would be asking what's the point of alt coins.

If Satoshi even came out and mentioned an alt that he thought was cool it would cause huge distributions, and if he made one.......

[HCLivess]

This type of attack does not work with HBN due to checkpoints according to Tranz 

[Daedelus]

POS Consensus Research results summarised: No POS attack devised to date has shown to be plausible.

https://bitcointalk.org/index.php?topic=897488.msg10152632#msg10152632

[jonald_fyookball]

POS Consensus Research results summarised: No POS attack devised to date has shown to be plausible.

https://bitcointalk.org/index.php?topic=897488.msg10152632#msg10152632

summarized by who?  A bunch of forum yackers?

[Daedelus]

POS Consensus Research results summarised: No POS attack devised to date has shown to be plausible.

https://bitcointalk.org/index.php?topic=897488.msg10152632#msg10152632

summarized by who?  A bunch of forum yackers?

Summarised by one of the authors of the research..

[jonald_fyookball]

POS Consensus Research results summarised: No POS attack devised to date has shown to be plausible.

https://bitcointalk.org/index.php?topic=897488.msg10152632#msg10152632

summarized by who?  A bunch of forum yackers?

Summarised by one of the authors of the research..

which poster is that?

[Daedelus]

Kushti, the link above goes to his post

[Daedelus]

New research paper and investigation into "tails switching" relating to Nothing at Stake.

We have updated our github repository https://github.com/ConsensusResearch/ForgingSimulation with a new version of the PoS simulation haskell code.
It now included two branches, master - for the single branch classical Nxt based code and "multibranch-experimental" - for the multibranch forging simulation. Recently
new algorithm for regulating tails switching effect is proposed and implemented. With it, a possibility of the N@S attack becomes also regulated as we now can introduce deducible  parameter of confirmations needed to stabilize recent blocks tails. The idea of regulating is straightforward - from time to time the node "forgets" almost all the branches and prolong only those whose cumulativeDifficulty measure is above some retargeting threshold. This threshold changes discretely, starting from 0. Unlike the Bitcoin difficulty param, the threshold always grows as the best block cumulativeDifficulty exceeds the previous threshold+delta. So nodes work as multibranch almost all the time, but sometimes becomes "single-branch" for a short time (one tick). This approach allows to have all the multibranch benefits and also get the network with regulating convergence. With a certain confirmation number  calculated, we can propose the strong resistance to the N@S as the long tails switching become very-very unlikely after the confirmations. We'll present the N@S simulation results ASAP.

There are more possible regulation procedures, for sure. Basing on the idea that sometimes nodes switch to the single-branch behavior one can introduce any verifiable quasi-random algorithm to do this. The proposed is the simple but efficient one, however more complicated algos (e.g. based on some nice hashes) could  secure the system more likely.

New paper on tails switching effect had been publicly released (https://github.com/ConsensusResearch/articles-papers/tree/master/switching). However the results of the simulations presented in the paper have been already renewed by the simulation software at https://github.com/ConsensusResearch/ForgingSimulation/tree/multibranch-experimental with the proposed threshold algorithm. As expected the algorithm allows to have confirmation number parameter deducible from the system constants and prevents the prolongation of similar branches. With it the resistance to the N@S becomes feasible and measurable! The results of N@S simulation + switching tails length distribution are coming.