It's no more far-fetched than they spending astronomical amount of money to attack a PoS system (and without guaranteed success, like attacking a PoW).
Okay - am coming around to your point of view now - but one point still bothers me which is the *hacking point*. With PoW you can't *hack the proof* you have to "do the work" (unless a fundamental flaw is found with SHA256 which would probably screw up all cryptos that exist at the moment). It may be much easier to *attack the wallets* of stake holders (via keyloggers, etc.) in order to gain the stake without having to spend "astronomic amounts" of money. I'd argue PoS miners are much more security minded, since they have to pretty much keep the wallet hot (online), even if briefly, in order to mine with it (though I believe in DPoS system you don't have to, since mining is delegated to the delegates, you can put your funds in cold storage). Also they are very much de-centralized, instead of concentrated in a pool, can you imagine how hard it is to obtain the identity of 10,000 PoS miner, their IP address, and then some how access their mining PC etc... I'd be much more worried that discus fish and ghash.io getting hacked simultaneously (imho a much more likely scenario) and the attacker instantly gain 51% in Bitcoin. Keeping your wallet "hot" by definition is going to make your funds vulnerable. If any PoS coin ever had any meaningful market cap then the miners would be under constant attack, likely to the point they would choose to no longer mine and exit ownership of the coin. Not really, you will only get constantly attacked if you are running a website with the hot wallet. If your PC is not running any service, not install any trojan/virus, and the network is properly firewalled, the chances of getting hacked, is practically zero. If it were that easy to hack someone, no one would be using Bitcoin anymore, since majority of the Bitcoin users are not using cold wallets neither, and the average Bitcoin user is much less security minded than a PoS miner, who probably has taken the correct security measures to keep his PoS mining PC safe. Plus the hacker would only be able to obtain an encrypted wallet, that he then have to try to crack, probably a fruitless effort. Usually a Bitcoin user gets hacked due to 1. unencrypted wallet, 2. backs up wallet + password in the sameplace (ie dropbox), and account gets hacked. 3. installed wallet stealing malware, and password got keylogged. Non of these will happen to a correctly set up PoS miner. So you need a dedicated computer and internet connection to mine in a PoS schema? Isn't that trending back towards the whole "wasteful resources" thing? Certainly not using gobs of electricity, but it does dictate a hardware requirement of sorts. Not really a dedicated computer, it can be run in a VM if you want. Just a properly secured computer. Of course PoS still needs some hardware, nothing operates out of a vacuum. The difference is efficiency. PoS does still use some resource, but it's maybe 0.00001% of what Bitcoin PoW uses. Also the resource usage doesn't have to increase when the eco-system grows in value. It's basically a static amount of resource needed. These resources are needed to have full nodes anyway, so they are not really wasted. |
|
|
|
|
betsofbitco.in removed from gambling category due to owner inactive and website domain expired.
|
|
|
|
It's no more far-fetched than they spending astronomical amount of money to attack a PoS system (and without guaranteed success, like attacking a PoW).
Okay - am coming around to your point of view now - but one point still bothers me which is the *hacking point*. With PoW you can't *hack the proof* you have to "do the work" (unless a fundamental flaw is found with SHA256 which would probably screw up all cryptos that exist at the moment). It may be much easier to *attack the wallets* of stake holders (via keyloggers, etc.) in order to gain the stake without having to spend "astronomic amounts" of money. I'd argue PoS miners are much more security minded, since they have to pretty much keep the wallet hot (online), even if briefly, in order to mine with it (though I believe in DPoS system you don't have to, since mining is delegated to the delegates, you can put your funds in cold storage). Also they are very much de-centralized, instead of concentrated in a pool, can you imagine how hard it is to obtain the identity of 10,000 PoS miner, their IP address, and then some how access their mining PC etc... I'd be much more worried that discus fish and ghash.io getting hacked simultaneously (imho a much more likely scenario) and the attacker instantly gain 51% in Bitcoin. Keeping your wallet "hot" by definition is going to make your funds vulnerable. If any PoS coin ever had any meaningful market cap then the miners would be under constant attack, likely to the point they would choose to no longer mine and exit ownership of the coin. Not really, you will only get constantly attacked if you are running a website with the hot wallet. If your PC is not running any service, not install any trojan/virus, and the network is properly firewalled, the chances of getting hacked, is practically zero. If it were that easy to hack someone, no one would be using Bitcoin anymore, since majority of the Bitcoin users are not using cold wallets neither, and the average Bitcoin user is much less security minded than a PoS miner, who probably has taken the correct security measures to keep his PoS mining PC safe. Plus the hacker would only be able to obtain an encrypted wallet, that he then have to try to crack, probably a fruitless effort. Usually a Bitcoin user gets hacked due to 1. unencrypted wallet, 2. backs up wallet + password in the sameplace (ie dropbox), and account gets hacked. 3. installed wallet stealing malware, and password got keylogged. Non of these will happen to a correctly set up PoS miner. |
|
|
|
I'd argue PoS miners are much more security minded, since they have to pretty much keep the wallet hot (online), even if briefly, in order to mine with it (though I believe in DPoS system you don't have to, since mining is delegated to the delegates, you can put your funds in cold storage). Also they are very much de-centralized, instead of concentrated in a pool, can you imagine how hard it is to obtain the identity of 10,000 PoS miner, their IP address, and then some how access their mining PC etc...
Being more security minded is not good enough. Good security is both difficult to achieve and hard to maintain even with security experts. What we need is a protocol that is secure intrinsically by several different methods by its innate design and not dependent upon how careful the users are. I'd be much more worried that discus fish and ghash.io getting hacked simultaneously (imho a much more likely scenario) and the attacker instantly gain 51% in Bitcoin.
This is a valid concern and why I suggested that your proposals are 1 step forward and 2 steps back. Mining centralization should be addresses by incentivinzing p2p pools and gettig more involved in mining, and a greater adoption of full nodes. P2pool is a good alternative, but hacking is actually far from my biggest concern. My biggest concern with PoW is that continuous non-stop transfer of wealth from the Bitcoin community into the pockets of mining hardware vendors and electric company. This I believe is the biggest reason we are seeing the strange price action this year, with massive amount of good news and adoption, the price keeps dropping. I think this is due to the inflow of capital can not keep up with the insane mining cost (and wealth transfer rate out of the Bitcoin eco-system into the pockets of hardware/electric companies). Can you imagine, if these hundreds of millions of dollars, instead of spent on pointless PoW mining, profiting the hardware/electric companies, if these money are directly invested in the Bitcoin eco-system, what it would do for Bitcoin? |
|
|
|
I'd argue PoS miners are much more security minded, since they have to pretty much keep the wallet hot (online), even if briefly, in order to mine with it (though I believe in DPoS system you don't have to, since mining is delegated to the delegates, you can put your funds in cold storage). Also they are very much de-centralized, instead of concentrated in a pool, can you imagine how hard it is to obtain the identity of 10,000 PoS miner, their IP address, and then some how access their mining PC etc...
I'd be much more worried that discus fish and ghash.io getting hacked simultaneously (imho a much more likely scenario) and the attacker instantly gain 51% in Bitcoin.
You put forward some good arguments and I think that it is inevitable that we'll have to move from PoW to something more efficient down the track. I guess the next thing is to work out "which kind of low-power proof is best" (btw - I have a new one that is neither PoW nor PoS but is as energy efficient as PoS - you'll have to PM me to discuss that). That sounds great, but you'll need to first implement your concept in an altcoin, so that it becomes a "proven" concept, otherwise I doubt it can be considered as a choice. |
|
|
|
It's no more far-fetched than they spending astronomical amount of money to attack a PoS system (and without guaranteed success, like attacking a PoW).
Okay - am coming around to your point of view now - but one point still bothers me which is the *hacking point*. With PoW you can't *hack the proof* you have to "do the work" (unless a fundamental flaw is found with SHA256 which would probably screw up all cryptos that exist at the moment). It may be much easier to *attack the wallets* of stake holders (via keyloggers, etc.) in order to gain the stake without having to spend "astronomic amounts" of money. I'd argue PoS miners are much more security minded, since they have to pretty much keep the wallet hot (online), even if briefly, in order to mine with it (though I believe in DPoS system you don't have to, since mining is delegated to the delegates, you can put your funds in cold storage). Also they are very much de-centralized, instead of concentrated in a pool, can you imagine how hard it is to obtain the identity of 10,000 PoS miner, their IP address, and then some how access their mining PC etc... I'd be much more worried that discus fish and ghash.io getting hacked simultaneously (imho a much more likely scenario) and the attacker instantly gain 51% in Bitcoin. |
|
|
|
Because it's cheap, not because the US or Europe don't know how to manufacture ASIC.
So they are going to somehow set up a "major manufacturing plant" to create ASIC that *no-one knows about* in order to attack Bitcoin? Seems a little far-fetched to me (western governments are not so good at hiding stuff like eastern ones are). It's no more far-fetched than they spending astronomical amount of money to attack a PoS system (and without guaranteed success, like attacking a PoW). |
|
|
|
Yes, they can just buy from existing miners, with a bit of a premium if they want the equipment fast. Also any first world government can easily mass produce ASIC equipment if they wanted to do it, there's no technical hurdles.
Hmm... then why are the vast majority of ASIC manufactured in China? (the hurdles are perhaps more economic as well as social) I am pretty sure Australia (the first world country that I come from) could *not* do this as they don't manufacture nearly *any electronics* nowadays. Australia could do it if they could do it economically and with a competitive price. They can't make money from it is the only reason they are not doing it. But if like you said, when the government don't care about printing more money, then they could easily do it. China is not some technologically advanced country, in fact all their chip maker machines are imported from ASML, which is an European company. China have cheap labor, that's about it. |
|
|
|
Yes, they can just buy from existing miners, with a bit of a premium if they want the equipment fast. Also any first world government can easily mass produce ASIC equipment if they wanted to do it, there's no technical hurdles.
Hmm... the why are the vast majority of ASIC manufactured in China? (the hurdles are perhaps more economic or social) Because it's cheap, not because the US or Europe don't know how to manufacture ASIC. |
|
|
|
In that case, it will cost the "government" much less to destroy a PoW chain. Also it might not even be possible to buy up 51% in a PoS chain if the coin is simply not for sale. But it's ALWAYS possible to get 51% in a PoW chain.
Too quick on the reply - the "resource needed to gain >50% is ASIC" and they are in short supply - so a government would have to set up manufacturing at least as good as the world's best (and keep it secret) to do so. Of course they could just try and "purchase" existing ASIC so that would depend upon the pools as to how easily they can be *bought out*. Yes, they can just buy from existing miners, with a bit of a premium if they want the equipment fast. Also any first world government can easily mass produce ASIC equipment if they wanted to do it, there's no technical hurdles. |
|
|
|
I feel this "nothing at stake" argument is being used without proof of its existence. Even though theoretically it's possible, but it also goes against human nature and logic. Why would you attack and destroy the reputation/value of the currency, that you hold an extremely large stake in? it makes no sense to attack yourself and destroy your own wealth.
From a "normal person's" point of view I'd agree - but from a "government's point of view" things change (as they can just print their own money as long as they wish). So if a government wants to "destroy" a PoS chain then they can just buy up (paying more and more as they print more and more) and they will succeed. In that case, it will cost the "government" much less to destroy a PoW chain. Also it might not even be possible to buy up 51% in a PoS chain if the coin is simply not for sale. But it's ALWAYS possible to get 51% in a PoW chain. "In PoW even if a government wants to spend billions to kill of Bitcoin they just don't have the ASIC equipment to do so." Why wouldn't they have the ASIC equipment? what's keeping them from obtaining ASIC equipment? I can see a variety of ways for them to obtain the equipment. If they want them fast, they can just offer to buy the equipment from current miners at a premium. If they have time, they can mass manufacture the equipment, easily. |
|
|
|
2 miner (discus fish and ghash.io) controls 40%-60% of the Bitcoin network, how is it easier to collude with 51 delegates?
Delegates are usually highly reputable members of the community, they also earn a small amount of fees for being a delegate, and at the first sign of dishonesty, they will be voted out. If they can't be trusted, why would you ever trust an anonymous miner who possibly have 0 stake in the eco-system?
Those aren't miners but mining pools. Mining pools have historically grown and faded over time and miners can switch pools and often do instantly. The beauty with bitcoin is I don't need to directly trust the intentions of people or delegates. Trust is built upon the fact that there are great incentives to play honest and it costs real money to make an attack. With DPOS or POS their is only the first and not the second, thus Bitcoins weakness is also its strength as an added layer of security. Whether or not one person can be trusted as a respectful member of the community or not should should not be the concern. We should focus on decentralizing bitcoin to a point where trusting a small group of humans isn't needed as all. Delegates can have the best of intentions and be corrupted over time, by special interests(lobbyists), misguided decisions, changes in character, or simply being a target by a hacker who holds them hostage or gains access to there vote through compromising their systems. Except Ghash mostly owns the hardware and can do as they like with them, cloud mining is getting more and more popular in PoW mining. I've already explained in the previous pages, for an attacker that is starting from scratch, it's MUCH cheaper to attack PoW system. It's nearly impossible to 51% attack a PoS system unless you spend an astronomical amount of "real money". Now if you are constructing a scenario where the attacker already owns a substantial amount of the PoS currency, then it's no different from a PoW attacker that already owns a substantial amount of the network mining power, of course it's much easier for them to attack, PoW has no advantage here neither. On the other hand, in the PoS system, the attacker would be also attacking himself, since he has an extremely large stake in the eco-system, costing the attacker large amount of "real money". So PoS system would again win in this scenario. I feel this "nothing at stake" argument is being used without proof of its existence. Even though theoretically it's possible, but it also goes against human nature and logic. Why would you attack and destroy the reputation/value of the currency, that you hold an extremely large stake in? it makes no sense to attack yourself and destroy your own wealth. |
|
|
|
Um no, for other coins, that makes no sense. Using PoW is not equal to snapshotting Bitcoin blockchain.
It seems we are arguing in circles now - so how do we get the "fair distribution" for your PoS without using the Bitcoin blockchain nor an IPO nor anything anonymous? You can make another thread for fair distribution discussion, it has no relevance to this discussion. Also I already told you, you CAN use PoW to distribute in a PoS system, so unless you are calling PoW distribution unfair, otherwise this discussion is over. |
|
|
|
There's a variety of choices in terms of solving "nothing at stake". Bitshare's DPoS system being one of them.
http://bitshares.org/delegated-proof-of-stake/Delegating voting to representatives does not address the "Nothing at stake" weakness. It simply makes it easier to collude with fewer delegates than gaining a consensus of miners. There is no electrical cost and/or equipment cost with an attack in a POS or DPOS framework. Your fears with PoW and its current pool centralization, limited ASIC manufactures, and limited miners is valid, but what we should be addressing are these problems directly and not codifying centralization through delegates into the framework as that represents one step forward and two steps backward in time. What we should focus on to improve bitcoin is 1) Encourage the use of decentralized p2p mining pools 2) crowdfund a company to create ASIC appliances. Would be better if the ownership of this ASIC manufacturer was controlled and owned by a large pool of bitcoin users. 2 miner (discus fish and ghash.io) controls 40%-60% of the Bitcoin network, how is it easier to collude with 51 delegates? Delegates are usually highly reputable members of the community, they also earn a small amount of fees for being a delegate, and at the first sign of dishonesty, they will be voted out. If they can't be trusted, why would you ever trust an anonymous miner who possibly have 0 stake in the eco-system? |
|
|
|
What issue? Bitcoin has already been distributed. Also PoW is a valid distribution method for a PoS system. You CAN use PoW to distribute for all I care.
Okay - so you agree with the Peter R model then (distribute according to a snapshot of the Bitcoin blockchain)? Um no, for other coins, that makes no sense. Using PoW is not equal to snapshotting Bitcoin blockchain. |
|
|
|
But initial distribution is irrelevant to the topic at hand, I'm discussing fading out PoW for Bitcoin, it does not need initial distribution or re-distribution.
Wrong - if you are advocating PoS to replace PoW then *you can't omit the issue of initial distribution*. What issue? Bitcoin has already been distributed. Also PoW is a valid initial distribution method for a PoS system. You CAN use PoW to distribute for all I care. You are trying to create a non-existent issue here and going off topic. |
|
|
|
Bitcoin's distribution is far worse than, say, Nxt's. It makes no sense to adopt it for a new coin.
This is very wrong as > 50% of NXT is owned by anonymous accounts all created at the time of the IPO (so most likely either one individual or a small group). And in PoS it is not *hash power* but ownership % that *controls*. You can't *overlook the importance of the initial distribution* if you are seriously going to suggest PoS and why would anyone who has bought, mined or earned BTC want to be "left out" because they didn't jump into some IPO (the reason why I think Peter R's idea is better than any other so far suggested). But initial distribution is irrelevant to the topic at hand, I'm discussing fading out PoW for Bitcoin, it does not need initial distribution or re-distribution. |
|
|
|
There's a variety of choices in terms of solving "nothing at stake". Bitshare's DPoS system being one of them.
If Bitshares are going to give a fair % (> 50%) for all holders of BTC from a given snapshot then I'd support it but I somehow doubt they are going to do that. Why would they do that? how is that even fair? Why are we discussing distribution at all here, since I'm trying to gain support for fading out PoW in Bitcoin, and Bitcoin does not need re-distribution. |
|
|
|
Please clarify exactly what alt coin are you supporting. Besides the initial distribution problem, you aren't addressing the "nothing at stake" flaw in POS. With POW , not only do you have to purchase or collude with a large mining company but spend the electricity to create the attacks. Bitcoins weakness("wasted" electricity) is also its strength in that it adds real costs to attacking the system. I'm not supporting any alt coin, I'm supporting fading out PoW for Bitcoin. There's a variety of choices in terms of solving "nothing at stake". Bitshare's DPoS system being one of them. The initial cost to attack a PoS system is actually pretty much astronomical, it's so high, that it hasn't even happened yet, and unlikely to ever happen. There has been zero 51% attack on PoS altcoin. While many PoW altcoin has been attacked to death. So apparently "wasted" electricity hasn't really deterred PoW attackers that much. |
|
|
|
Usually only about 10-20% of the coin actively engage in mining PoS.
That in itself is "worrisome" as presumably it means that forks could appear more often. The other problem with PoS is that those not mining could be in fact colluding to present a new better chain (the cost of doing so being very little as the "Nothing At Stake" point made). Again it comes down to "distribution" with PoS that basically *cannot be anonymous* in order to be trusted. I'm not sure where this PoS "collude" concern come from. Who would collude to destroy their own wealth? I hold some peercoin, if you tell me to collude with you to attack Peercoin, I would tell you to gfys. Do you think it's easy to collude with 10,000 peercoin holders? or is it easier for discus fish and ghash.io to collude and gain 51%? I think the answer is obvious. The fact is forks has NOT appeared more often in PoS, in fact PoW system has had much more forks, including Bitcoin. Sorry to speak the truth. It is easy to collude with a 10k Peercoin holder than colluding discus fish or ghash.io to gain 51%. Investment made by discus fish or ghash.io is far more than the investment made by 10k Peercoin holder. Since discus fish and ghash.io's only stake is in their mining equipment, the attacker just has to spend a bit more than that amount of money to buy their loyalty, or if they don't agree, the attacker could just start their own mining operation, yes? that amount is not even 5% of Bitcoin's marketcap. Now with Peercoin, if you want 51%, you have to buy them, you can't just "collude", since no one will listen you. Now if Peercoin's marketcap is same as Bitcoin, tell me which way is easier? Well, please tell me 1 BTC = ? Peercoin. Sorry, I'm not into any Alt except Stellar. You don't need the exchange rate, you just need to know under the same marketcap condition, getting 5% or 51%, which one is easier? actually 51% is much harder than it appears, nearly impossible unless you spend astronomical amount of money. |
|
|
|
|
Show Posts
