Any system of checks and balances will have its price. Look at what we do with gold: we dig it from a hole in Africa, we transport it in other countries, we dug holes or build vaults and throw it there. Like Warren Buffett said, it has no utility, if Martians were to look at the process they would scratch their heads.
Except for the guarantee of uniqueness. Which is one of the most powerful invariants against corruption and government-sponsored inflation, goals which are worth these effort.
There are a few more actually: gold is one of the least putrescible materials out there (it's one of the least reactive elements available and as one of the royal metals very few acids can attack it), it has a distinctive look, and it has a very high density (most denser metals were only discovered in or after the 19th century, and are generally more valuable) which makes it easy to differentiate pure and impure golds (such as fake coinage). It also used to strike a good balance between rarity (which ensures the market won't be flooded) and availability (which actually made it usable as money).
Of course, most of this broke down as new, very productive, mines were found (e.g. in South America) and mining techniques improved. And nowadays we have the opposite issue: gold mining output declines, and is unable to follow the value growth of the world itself. This is one of the reasons why most countries moved first to fractional gold standards and then to convertible currency.
"And nowadays we have the opposite issue: gold mining output declines, and is unable to follow the value growth of the world itself. This is one of the reasons why most countries moved first to fractional gold standards and then to convertible currency."
This is just a ruse used by governments as a pretext for inflation. It isn't necessary for money supply to 'keep up' with growth. As demand increases, the value of money goes up, and people simply use smaller units for exchange.
Saying it is just a ruse for inflation is not being honest. Many, though not all, economists believe that deflation causes money to be more valuable tomorrow than today, which leads to over saving which leads to unnaturally low demand. It's not helpful to act as though there is no debate on the subject. I'm sure there are other problems as well.
> This is just a ruse used by governments as a pretext for inflation. It isn't necessary for money supply to 'keep up' with growth. As demand increases, the value of money goes up, and people simply use smaller units for exchange.
The result is exactly the same, you're just making the whole process more painful by rebadging money.
In economic terms what you're describing is called deflation.
During deflation prices keep falling, which causes consumers to post-pone their purchases (tomorrow they'll be able to get that TV at a lower price so there's no point in buying it today) and investors sit on the money since they'll be more valuable tomorrow.
This is the exact opposite of inflation, which causes people to convert 'today' money into goods since tomorrow they'll buy less with it; it also causes investors to put money to good use or in interest-bearing accounts since otherwise they lose the value).
If deflation is so dangerous I wonder why anyone buys consumer electronics which, despite currency inflation, continue to decline in price due to advances in technology. If we experienced the full effect of deflation, today's iPad would cost a much tinier percentage of yesterday's Apple II. Would that have stopped people from buying either? I doubt it.
If we experienced the full effect of deflation, then average wages would be falling, as well as the average prices for everything (not just for consumer electronics, which make up a very small percentage of the average person’s expenditures). This would be particularly bad news for any person or business that carried, or expected to carry, debt. You wouldn’t want to take out a mortgage on a house (or have your company borrow money to build a factory) if you expected that the nominal resale value of the house would go down and your nominal income would go down while your nominal monthly payments would stay fixed.
Consumer electronics are unique right now in becoming orders of magnitude more capable before the old versions even begin failing. New products sell because nothing comparable was even available before and it's inconceivable that next year's version could be better somehow (even though it will). No other category has this going for it.
Gold does not have much intrinsic value (not actually correct anymore, in our modern world gold is used quite a bit in industrial contexts): apart from being shiny, before the 19th century there wasn't much it could be used for. The vast majority of its value was extrinsic and pretty much arbitrary, in its use as part of monetary exchange systems.
Before the 19th century, gold was used for a very important purpose: money. Money is a commodity just like any other, and some things make better monies than others.
What are some characteristics of a good money?
It's divisible. 1 oz. of gold is equal to two 1/2 oz. of gold. Two halves of a diamond are not equal to the intact whole.
It's fungible. Gold coins of the same weight are, for all intents and purposes, the same. The same cannot be said of diamonds or oil.
It resists deterioration/decomposition. (Self-explanatory).
Its rarity/weight/volume characteristics are manageable. You can buy an iPad with a reasonably sized gold coin. The same could not be said of bushels of wheat, lead, stone, oil, etc.
It is relatively easy to identify. This provides anti-counterfeit measures out of the box.
I'm not a gold bug (I own zero gold... not even jewelry), but I do appreciate that it makes complete and total sense that it was used for money for thousands of years. Should 1 oz. of gold be worth $1500? I don't know; I don't follow the market. I do know that gold is extremely valuable due to how well it serves as a medium of exchange, and due to that fact alone.
tl;dr Gold has intrinsic value: it makes a good medium of exchange.
The major difference, of course, is that diamonds are not rare, can be industrially counterfeited, and are easily damaged.
Diamonds, in some way, are the "anti-gold." There is a great game-theoretic reason for why something as completely worthless as diamonds make great engagement rings though: They are something expensive to the male (so he can't repeat the process of proposing very often) - but worthless to the female (so she can't be accused of being a gold digger) - and at the same time sparkly and easy to show off.
> The major difference, of course, is that diamonds are not rare, can be industrially counterfeited, and are easily damaged.
Gold is not rare either. Please re-read my comment, it's about intrinsic and extrinsic value, especially across history. Before modern industrial usages, diamond was all but useless (even worse than gold, if anything) and all of its valuation was extrinsic.
And much like gold's, diamond's intrinsic value has risen in modern times with industrial uses.
> Diamonds, in some way, are the "anti-gold."
Most definitely not. If you want anti-gold, you should look at water instead.
The value of something is how much you can get in return for the thing from a purchaser that you can find. Even "value is what we agree it is" isn't really right, it's what you can concretely get for the thing. (For instance, the fact that a certain Magic the Gathering card is listed in some book for $20 is meaningless if you can't find anyone who will give you more than $5.) There is no other definition of value that's actually useful, but there are plenty of others out there that will lead you astray.
But at least money exists for the purpose of arbitrarily representing "real value". Gold has some real value, it's just quite a bit less than the value we've given it.
Bitcoin is not quantum-hard. When quantum computing reaches a certain threshold, bitcoin will lose its uniqueness and collapse.
There are post-quantum algorithms that are computable on ordinary computers, but they're still under research and they'd make Bitcoin's data set considerably larger.
Bitcoin is not quantum-hard. When quantum computing reaches a certain threshold, bitcoin will lose its uniqueness and collapse.
Could you elaborate which aspect of bitcoin has efficient quantum algorithms? The proof-of-work system requires you to invert hashes and quantum algorithms only give you a quadratic speed up (brute force takes time O(\sqrt{size of range}) rather than O(size of range)). Right now there is no quantum attack on SHA-256 (which is what Bitcoin uses).
There are post-quantum algorithms that are computable on ordinary computers, ...
This sounds interesting, but could you elaborate? What does it mean for a post-quantum algorithm to be computable on ordinary computers?
Could you elaborate which aspect of bitcoin has efficient quantum algorithms?
BitCoin uses public key cryptography for its digital signatures. That in turn relies on the difficulty of factoring large numbers, for which there are efficient quantum algorithms.
Bitcoin actually uses ECDSA for its digital signatures, which relies on the difficulty of calculating discrete logarithms rather than factoring. However I believe that there are efficient quantum algorithms for that, too.
Bruce Schneier has written that quantum computers only halve the effective number of bits for symmetric encryption algorithms, which seem more closely analogous to hashing than factoring. That would be a giant leap in mining capability, but perhaps not so much as to cause collapse. Switching to 512-bit hashes would bring us back to where we are now.
Correct, however the Bitcoin website has an answer for that: upgrade. The cryptographic methods are not fixed for all time, we can switch to other methods (which are quantum-resistant) when it becomes necessary.
Assuming they are available and deployable at that time. It's like arguing against climate change mitigation because a technological solution will present itself: you might be right, but why gamble?
Scarcity of money is not the most desirable end, because the supply and demand for money threatens to change the value of money. Even if all the money in the world were fixed, you would see inflation and deflation, more inflation than you've ever seen on a properly managed fiat money system. This is not mere economic theory, it's a matter of history.
> more inflation than you've ever seen on a properly managed fiat money system
While this may be true, a scarce money system does produce better result than poorly managed fiat money systems. You won't see the kind of hyperinflation that we saw in Zimbabwe with the gold standard.
A poorly managed money system is one symptom of a government that is poorly managed, period, and living under a poorly managed government is always going to suck. If armed thugs loyal to the ruling party come to kick you off your land, steal your valuables, and discourage you from voting for the opposition, you’re not going to be able to pay them off with Bitcoin. (ETA: ...or if you can, the ability to do so is not going to make that country a pleasant place to live.)
The author misses the point. The security of the Bitcoin network (at least in terms of double-spending) is tied to the level of work required to generate a block. If everyone decided to use only 1% of their computing power, they'd use less electricity, but the Bitcoin network would then be 100 times less secure.
When people talk about "wasting resources", it's an indication they don't understand how Bitcoin works. One might as well complain about wasting resources on SSL when you could just send data in plaintext.
The trick lies in changing the bitcoin protocol. If the majority of clients can agree on a way to use less resources, then that's what will be enforced by the network. If a new bitcoin client enforced a mandatory delay of one hour and invalidated blocks that were made before that one hour, we would effectively have added one hour of idle time to the network. This is not a solution, but it is a demonstration of how the protocol can be altered, as long as the majority agrees on it.
If it's just a majority of clients, then an attacker could just start up a million new client processes and outvote the existing network.
If we're talking a majority of unique IP addresses, then anyone with access to a large IP address block could outvote the network. This approach also wouldn't be feasible for IPv6, where IP addresses are not a scarce resource.
The Bitcoin approach is to define "a majority" as the majority that has the most computing power. Because computing power is always going to cost money, this guarantees that anyone wishing to subvert the network needs to outspend all the honest miners.
If you can think of another way of defining "a majority" without needing a lot of CPU resources, a lot of people would be interested in hearing your solution.
> If you can think of another way of defining "a majority" without needing a lot of CPU resources, a lot of people would be interested in hearing your solution.
Let Bitcoins equal votes.
Set up a Bitcoin account for each option in the vote; send Bitcoins to the account associated with the option you prefer. Transaction verification acts as an electoral audit.
That sort of system would give power over the money supply to those invested in the currency, which seems to be exactly what the Bitcoin community wants.
The question was how to allow clients to form a new consensus. I did not suggest using Bitcoins to verify transactions; rather, I suggested using verified transactions to verify the will of the Bitcoin community.
Forming a consensus isn't the problem; the problem is enforcing honest application of the new rules. Assuming the majority of people in the network are honest, you need some way of determining what that majority decides, within the network itself.
You have completely missed the point. Because the network can observe transactions, using a Bitcoin account as a ballot box, and Bitcoins as the votes, is a transparent, verified mechanism for determining the community's aggregate will.
This approach negates the incentive to create sham clients under a one-client-one-vote system; it has the same benefit over IP-based voting systems, or what have you. Voting with Bitcoins means that that every participant in the economy has a voting power equal to his account balance (if the Bitcoins are returned after voting) or the number of Bitcoins he both has and is willing to give up to affect the vote's outcome (if they are not returned). So those who are more invested in the Bitcoin economy would have more (or potentially more) power to determine the rules of the economy.
I'm not saying that you can't use bitcoins as votes. You could; but any change to the rules needs to be also be enforced within the network, right? So you need some mechanism for figuring out the majority within the network, which obviously can't be done with bitcoins.
The problem is that it is easy to simulate that you are waiting.
1) Save current time
2) "Precalculate" the current block
3) Wait until the 3601th second
4) Publish the block that you "calculate" in just 1 second
5) Profit??
Isn't the waste also present even without bitcoin, too? Governments make complex laws and regulations in different industries. They do 'magic tricks' to make fiat currency appear and disappear. Banks do complex things to make a profit that largely do not correspond to creating real value for society. Competing companies do the same things, duplicating effort, mostly to make a profit. (e.g. Do we really need so many social gaming companies?) Companies also do other things that do not create real value (e.g. lobby politicians to gain political advantage).
A large part of human endeavor is not about increasing value in the world, but rather increasing value for oneself, possibly with the side-effect of decreasing value for others. It's just the way we are.
I don't think bitcoin's purpose is to aggravate or ameliorate this pervasive problem. To me, the point of bitcoin is to shift economic power from the circles of people who presume to rule the rest of us with complex economic policies and regulation that ostensibly keep the economy stable and growing.
With or without bitcoin, the 'waste' from competition still going to be there, and I have no reason to think that the adoption of bitcoin is going to increase the total amount of 'waste'.
"Competing companies do the same things, duplicating effort, mostly to make a profit. (e.g. Do we really need so many social gaming companies?)"
In theory, we don't "need" more than one of anything: one telecom provider, one computer maker, etc. And they could be more efficient if they didn't have to spend money on advertising, etc.
But the history of monopoly companies seems to show that without the need to compete, companies do not improve their product or its cost-efficiency. So it appears that the loss of efficiency created by having many companies reinvent the process to make a widget is outweighed by the efficiency gains each company makes in an effort to out-compete others. Hence, better, cheaper widgets for less money.
As to the value of social gaming companies, I think it's nearly nil, but apparently millions of other people disagree.
> A large part of human endeavor is not about increasing value in the world, but rather increasing value for oneself, possibly with the side-effect of decreasing value for others. It's just the way we are.
It's not just the way we are, it's the way we are under capitalism.
But I agree that the wastefulness of Bitcoin is nothing compared to the waste caused by capitalism as a whole.
"It's not just the way we are, it's the way we are under capitalism."
What line of reasoning makes you think that humans are naturally unselfish and are made so only by an external economic system (which is itself created by humans)?
Judeo-Christian thought says that humans are naturally sinful and selfish. Evolution says that the fittest survive at the expense of others (though this may involve cooperation with related creatures). Who says we're all basically nice?
"But I agree that the wastefulness of Bitcoin is nothing compared to the waste caused by capitalism as a whole."
What system would you have instead of capitalism? If socialism or the like, can you give an example of when that has ever worked well? There are many counter-examples in history, you know. Workers sitting down doing nothing because the government quota said to only make 250 pairs of shoes today, etc.
> There is a solution to this problem. If someone releases a new Bitcoin client (and miners) that run on only 1% of the available processing power, and gets more than 50% of the network to adopt this new client, then total electricity costs drop would drop hugely
No, because you can't cryptographically enforce that all compatiable clients use only 1% of their available processing power.
If such a client took over the bitcoin network, somebody would use a patched version of that client that pretended to use only 1% of the computing power, but actually used all of it. There's no way the other clients could know, so the whole spiral would start anew.
I was worried about this so I worked out the math. It wasn't as bad as I expected.
Let's say it's 20 years from now, the total size of the bitcoin economy is $1 trillion, and there are 10 million bitcoins in existence. (We could find the number of bitcoins exactly, but 10 million makes the math easy and it's within a factor of 2 of the ultimate limit. There are about 6 million coins right now.)
$1 trillion / 10 million coins = $100K per coin, so we can expect people to spend almost that much money to generate a coin.
The rate of coin production is currently 50 per 10 minutes, dropping in half every four years, giving us 9 coins per hour in 2031.
That's 78K coins per year, at $100K per coin, or $7.8 billion in computer time and energy spent per year.
Let's say half of that is energy, call it $4 billion, including the energy to make the dedicated mining computers. At ten cents per kilowatt-hour, we're talking 40 billion kilowatt-hours per year. Divide by the number of hours in the year and we get 4.5 gigawatts.
In other words, a $1 trillion bitcoin economy can be run on the output of roughly five typical nuclear power plants (assuming it's 20 years from now).
If we get to $1 trillion in only 12 years, energy usage will be four times higher, since coins will be worth about the same but four times as many will be generated per hour.
If people use custom ASICs to generate coins, they'll be costly but more energy-efficient, weighting expenditures more heavily on hardware than energy.
People can grant transaction fees to miners to make their transaction go through faster. If transaction fees become prevalent, resource usage will be higher, since there will be additional reward for completing blocks. But $7.8 billion is already almost one percent of the bitcoin economy. If the average bitcoin changes hands once per year with a one percent fee, or ten times per year with a 0.1 percent fee, we add $10 billion to mining rewards and approximately double our resource usage.
I don't know much about economics, but this seems to illustrate why bitcoin must ultimately fail. Since producing new bitcoins by design becomes exponentially (!!!) more difficult as time passes, the true cost in resources (electricity, computing time, etc.) to produce a new coin can only increase. That resource use can be justified, but only if the value of the bitcoin is also increasing.
Here's how I see it playing out: eventually it will be too costly for most people to produce new bitcoins. Since nothing forces one to use bitcoins, bitcoin holders will cash them out, and the exchange rate will plummet. Deflationary panic? http://en.wikipedia.org/wiki/Deflation#Money_supply_side_def...
It will always cost about the same to produce a bitcoin as the bitcoin is worth. If it costs more, people will stop doing it and the difficulty will drop. If it costs less, people will profit by doing it and the difficulty will rise as more people join in.
It's already difficult for individuals to produce bitcoins, but people have solved that by pooling their resources, getting regular rewards of fractional bitcoins.
That's helpful, thanks. Still, it's only profitable to mine bitcoins at any computing cost if they can be exchanged for other items of value. If the value of bitcoins drop and people leave the bitcoin economy, making the last N bitcoins easier to compute won't make them more valuable.
The way I understand bitcoins might be flawed but would certainly appreciate clarification ...
My problem with bit coins is that early people who start mining have a greater advantage than later ones. While that's not the classic definition of a Ponzi scheme (in the classic definition, payouts at each round are funded by new participants), it is a feature bitcoin shares with them. If someone started mining bitcoin when it first started, they have a significant stash created by now. As people start to assign value to the virtual currency, these people are getting a significant proportion of value for simply being early adopters.
"these people are getting a significant proportion of value for simply being early adopters"
I'm not one of them but have no gripes with that. What you describe is also commonly known as the entrepreneurial taking-unknown-risks for an unknown-future-reward.
It's the same with government currencies -- the government/banking complex gets newly created money first, and gets to spend it when the new money supply still has the purchasing power of the old money supply, ie. before inflation. This is their bonus for successfully defending their monopoly on money creation. Not that I'm in favour of such a privilegue. Which is the appeal to a growing number of users: yes, early adopters may be rewarded in proportion to how early they adopted, BUT no-one can unilaterally create new Bitcoins at the push of a button or by signing some law and even the early adopters could not initially create more bitcoins or faster than prescribed by the open source peer to peer protocol.
It's not the same as taking early risks; this is a very constrained and artificial environment in which bitcoins are generated. One can be pretty certain that if one generates coins early, one can then make money selling them to others. This is much more similar to a pyramid marketing scenario, where the earlier franchises are inherently worth more, and eventually the value generated for the terminal generation of participants is close to nil.
I'd argue you're not exactly correct about the release path of currency in a fractional reserve system, although I don't disagree that elements are currently implemented in quite unfair ways. I'd also suggest, though, that government fiat currency works very much better than anything commodity based, or indeed than a partially fiat system that functions exactly like a commodity based system, as bitcoin does.
One can be pretty certain that if one generates coins early, one can then make money selling them to others.
I don't know about that. Back in 2009 or 2010, Bitcoins were worth practically nothing and almost everyone predicted that they'd never be worth anything. And if no one mined, maybe Bitcoin would never have taken off (to the extent that it has) and it would still be worthless.
I'm suggesting that if you believe in the model, and even the most obvious ramifications of such an economy, the value of participating in the system is to be as far upstream as possible. I don't believe in the model, so, I'm not suggesting it has real value now or going forward.
You're also actually assuming something that runs counter to the stated bitcoin model, as described in the FAQ (which seems to be down at the moment). That is that there's an investment in making bitcoins, which they would say is not so. The intent in burning cycles and power is simply to ensure security, while the value comes from faith in the currency. Hence, there is no investment in creation: the investment would instead be in accepting bitcoins in transactions, which iterates the pyramidal nature of the system.
The situation is not much better with inflating fiat currencies. Those who are closer to the money spigot have an advantage over those who are further way.
It's no where near as pronounced with fiat currencies as it is with bitcoin. Only a few currencies have experienced this level of hyperinflation that bitcoin has (e.g. weimar republic, zimbabwe), and those currencies have failed.
If inflation is defined as money supply growth, then bitcoin is not hyperinflating. The rate of bitcoin inflation is decreasing over time until it reaches zero.
If inflation is defined as the rise in prices of goods, then bitcoin is deflating.
Isn't bitcoin experiencing some sort of hyperdeflation, not hyperinflation? The strange thing here is not that too many coins are being created, it's that the demand for bitcoins is increasing faster than the supply. The coins are becoming hypothetically more valuable, not less.
Under hyperinflation, each monetary unit is worth less and less. Your complaint about bitcoin is that each unit is worth more and more. Your analogy makes no sense.
If what you're suggesting is that coins can only sit in one address for a limited amount of time, then that would not do anything because people would just create new addresses to send the coins to.
The standard Silvio Gesell expiring-money scheme doesn't have anything to do with where the money is at any given time. The way it works is that for the government to accept a note, it must be up-to-date with stamps, one stamp for every month since it was issued. The stamps (traditionally, bought from the government) cost 10% of the value of the note.
Nobody wants to be sitting on a pile of this sort of cash at the end of the month, because they'll lose 10% of the pile of cash. The intended effect is to dramatically increase the velocity of the money, so that a small stock of it can sustain a very large flow.
I don't know enough about this stuff to sensibly evaluate the claims its proponents make for it.
Wealth consists of things like cured diseases, tasty and nourishing meals, being dry when it rains, and seeing the people you love as often as you want; and the resources and skills we use to create those other kinds of wealth, such as antibiotics, bread mold, spices, beans, houses, corrugated steel, iron ore, blast furnaces, airplanes, kerosene, and CAD software.
Gesell's scheme does not destroy any of those things; it only destroys money. Money is not wealth. Money is just a way to facilitate cooperative ventures with people we don't know and trust. With money, I can cooperate with 200 other people to pay for an intercontinental flight, and with the hundreds of thousands of other people who drilled and refined the kerosene, aluminum, and other resources needed to keep us in the air.
But the money game has a lot of drawbacks. So it's sensible to experiment with it and see how we can improve it. Gesell's scheme was one such experiment. It was reported to be very effective at creating wealth by its proponents when it was tried.
I don't understand the rationale for penalizing saving. The car I couldn't afford on this month's paychecks alone is wealth. Food and shelter after I'm too old to work will be wealth.
You can hardly expect anyone but a proponent to advocate for an idea...
Also, asking for a control group in an artsy fartsy field like economics is too much. Where is the control group that guides the decisions of the central banks? There is none! OMG GREAT DEPRESSION!
In order to spend bitcoins, you need someone willing to exchange them for goods or services. Why would anyone do that if the bitcoins they received were going to expire?
For instance, let's say I wait until the literally last minute to spend my expiring bitcoins. I want to exchange my almost-expiring 10 BTC for $100. Would you accept this offer, knowing that those 10 BTC will vanish shortly after you accept them?
Alternatively, maintaining a hard limit on the number of BTC saves resources, compared to a system that allowed for a fixed growth. Once the 'mining' stage ends, no new computational resources are invested. The system accumulates a kind of computational capital (a mathematical problem difficult enough to be intractable) which needs no further upkeep, other than a network presence to provide verification.
It would only be extended with new transactions, not blocks. Computational resources will be devoted to verification rather than brute-forcing hashes (a much less computationally intensive matter).
No, blocks are still created in exactly the same way, the only difference is that no new bitcoins are created as a side-effect. It is necessary for new transactions to be locked into the block chain with a bruteforced hash, as it's this expenditure of processing power that makes the transactions permanent and prevents double-spending.
I just imagined that the bitcoin world post-mining (after or near the 21M BTC is reached) would just consist of the transactional interfaces and not need the specialized GPU miners.
Unfortunately, that's not how it works. The "miners" are really "transaction settlers". It's the longest block chain that represents the globally consistent view of which transactions have settled.
OK, that was an aspect I was unfamiliar with. So after the mining phase ends, transactions become zero-sum BTC games, where the biggest 'miners' accrue their proportional share of the total fees of transactions validated.
I don't think the bitcoin system is in any way designed to benefit the miners. Its intent is to create a digital currency that has a guarantee of retaining its rarity in the future. Making mining difficult and inefficient is kind of the point.
The hypothetical solution is flawed: it would take not 50% of miners to adopt a 1% implementation; it would take 100%, since if only 1 defects, that one wins all the coins and all the cooperative 'prisoners' lose.
There are two mechanisms by which defectors could win coins.
First, they could simply mine more coins. For them to win almost all the (new) coins this way, they have to do almost all the mining. Say, 80% of it, although it's really a matter of what you mean by "almost all".
Second, they could collude and subvert transaction audit trails so that the Bitcoin system transfers ownership of all existing coins to them. For them to win all the (new and old) coins this way, they only have to do more than 50% of the mining.
So, if collusion is feasible, the second approach is easier; so let's forget about the first one for the moment.
The point at which the defectors, each using 2 orders of magnitude more CPU time, are doing more than 50% of the mining, is when they control more than about 0.9901% of the network's total raw computing power, because at that point, the cooperators control only 99.0099% of the network's total raw computing power, and the cooperators are using only 0.990099% of the network's total raw computing power to prevent counterfeiting.
If a single defector controls only, say, 0.5% of the network's computing power, they will get only about a third of the new bitcoins mined, and will not be able to subvert the network.
Only needing to have 0.99% of the network collude in defection to break the system is a pretty bad problem, but it's not nearly as bad as any single person being able to break the system.
Don't the miners provide network services as well? Like how real mines don't just take from the ground, they also co-exist with civilian infrastructure built to keep the miners alive and functioning, and in bitcoin's case, the mining rigs dually enable transactions?
The same argument can be applied to gold: resources are being utterly and completely wasted on securing Fort Knox, let's shutdown it and store all gold on a lawn without guarding it.
Mining is securing the bitcoin network in the similar sense, as Fort Knox secures gold.
Comparing it to gold doesn't make sense; we've been off the gold standard for a long time. What it should be compared to is minting coins, printing money, and banking. (As far as banking goes, I'm not sure bitcoin really solves any of those security issues.)
Why can't we tie the bitcoin mining process to some sort of useful calculation? Maybe some variant of folding @home or something? We just need some problem with a hard to calculate, but easy to verify 'right answer' right?
More than that. It needs to be a problem that takes a predictable time to solve, it needs to be a problem that we're unlikely to discover faster algorithms for, and it needs to be something we can tie to a set of data.
Cryptographic hashes are designed to have these exact properties. If we wanted to replace this with a "useful" calculation, you'd essentially need to create a new hashing algorithm that also had some practical use beyond hashing a set of data.
Given that creating good cryptographic hashing algorithms is hard enough, I suspect that creating one that is tied to a "useful" result is not feasible.
No, mining solves a specific problem: it generates an agreed-upon ordered list of transactions, so if anyone double-spends, everyone agrees that the first recipient owns the coin. The new coins are just an incentive to solve the problem.
To solve this problem, the input has to be bitcoin transactions, rather than biologically-relevant amino acid sequences.
Even if such calculation could be possible, bitcoin network probably would be less secure, because attacker can attack it and be compensated for attack.
This is just a "prisoner's dilemma". The alternative is that we trust everyone not to use their full ability to generate blocks. Obviously this would completely defeat the point of bitcoins.
Yeah, that's why the only thing I could think of that would 'fix' this, is to release new miners that use less processing power, and a new client that only allows input from those miners. Even so, such a thing would be equally as difficult to introduce as, for example, a rule that would hand out 200BTC instead of 50BTC.
Most of the time graphics cards aren't used for anything remotely as socially beneficial as contributing to the bitcoin network. Rather than complain about the waste, find a way to do something more useful with them and set up an equivalent reward system for people to respond to.
What you're proposing is a cartel of users who agree to only use 1% of their computing power. The Prisoner's Dilemma still applies and thus the incentive to cheat still remains because using more than 1% can give you a competitive advantage. By assuming every one will compete as much as they can the prisoner's dilemma disappears because there is a limit of how many resources a user can expand. The main issue is whether bit coin is a valid expenditure of those resources
So, when we waste our processing power on credits generated on limited supply by the Bitcoin network, it's a waste, but when we waste our time on acquiring credits created in limited quantities by the (Central) Banks it's okay.
I guess the issue here is that calculated bitcoins don't contribute anything. The grid should be used for something useful such as folding proteins or even SETI for that matter.
I would only add that we're using a LOT of energy to create a fake (read: virtual) resource. The best part about virtual resources is that they don't use many real resources, but this one is the opposite.
Granted, it's a 1-time cost, but as the article notes, it could be created with a much lower 1-time cost.
I'm guessing that's the cost of printing a single dollar bill. There are also hundred-dollar bills, which presumably cost the same.
It'd be pretty interesting to get the total cost of printing, transport, and storage, and compare to the cost of maintaining a similar amount in bitcoins (which I calculated here: http://news.ycombinator.com/item?id=2602667).
I didn't count the $100 because I don't believe they're printed/circulated/destroyed/reprinted as heavily as the $1. Who knows about the 5, 10, 20 and 50.
I also don't think this $0.042 covers the labor cost of distributing, collecting, securely shredding and tracking the movement of all these bills through the system every 18 months.
I was thinking more along the lines of what the total cost might be. There's about a trillion dollars worth of cash circulating, but since it's not all ones we can't just multiply a trillion by $0.042.
In any case, it sounds like the bitcoin cost might be in a similar neighborhood to paper costs.
Unless you live in Ted Kennedy's or Trent Lott's district, where the cotton farms and paper mills are.
For decades these senators continuously sabotaged measures to eliminate the paper Dollar and get the US entirely on dollar coins. The US is probably the last major economy to still be using paper for such small denominations.
Depending on where you draw the line between "major" and "minor" economies, the US can be the only major economy to do anything at all, since its economy is bigger than that of any other country.
However, in my recent and local experience, small-denomination paper is not so unheard of:
Country Smallest paper bill in USD PPP GDP/year
Argentina $2 0.50 $640B
Brazil 2 reais 1.26 $2200B
Uruguay $5 (now rare; usu.$10) 0.27 $41B
That's not quite true. Bitcoins cost virtually nothing to create, but you are only allowed to create them if you succeed at securing a block of transactions. All of the processing power used in the Bitcoin network is designed to verify the transaction chain and prevent double-spending. The bitcoins generated as a result are just a reward to compensate people for their energy and hardware costs.
I recently looked into generating bitcoins, and that's not really true. Very little power goes into generating a successful bitcoin, true... But a LOT of power is wasted by failing to generate them. That's why it takes days, weeks or months (depending on processing power) to generate a successul bitcoin, on average.
You're confusing solving a block with creating bitcoins. The distinction is important (in my opinion), because it determines the reason all that processing power is needed.
A frequent mistake is to assume that all that processing power is used to make bitcoins hard to create. This is incorrect; the processing power is used to verify the block chain. The bitcoins created as part of each block are just a reward to make it profitable to participate in the network.
You're trying to win a lottery. With everyone racing to create a "solution" that fits the difficulty criteria there's no guarantee that you'll even win a block if you throw enough horsepower at it. A newcomer with a 900Mhz Pentium could find the correct solution on his first cycle.
There's a lot to say. I personally think the Bitcoin debate is interesting because it's making a lot of people question the concept of money - especially in the current state of the world economy.
It's also a new and interesting intersection of technology, economy, and personal trust.
All those things are true. But I still am not seeing any points raised in the discussion here that haven't already been made in any of the countless other discussions that have already played out here on HN in the past few weeks.
It's not that Bitcoin isn't new and interesting-- it is-- but that there is nothing new and interesting to say about it today that wasn't said (here) yesterday.
To be blunt, you're doing more harm than good by complaining about the discussion. For every person that cries "dupe", there are ten more who are grateful for that dupe. Dupes are a good thing.
If the conversation points are correct, then there's nothing wrong with making them.
As something like Bitcoin becomes more popular, people will naturally have questions and concerns. Should they search for the answer before asking? Maybe, but it doesn't really matter if they don't. People are free to downvote them for asking. But they don't, and the resulting conversation becomes interesting for those who haven't seen its variants, such as me.
You've been a member of HN almost as long as I have. Haven't you noticed the destructive effects meta-discussion has on articles? If so, why are you perpetuating it?
You've been a member of HN almost as long as I have. Haven't you noticed the destructive effects meta-discussion has on articles? If so, why are you perpetuating it?
Because I've also noticed the pernicious effect of waves of discussions like this that never die. Articles about Bitcoin (and other similar topics) come to dominate the front page for a while, and crowd out other, more interesting content.
I don't usually involve myself in meta-discussion, but this seemed like such an egregious example that I felt I ought to speak up.
If you have, welcome! But understand that not every day is chock-full of new and amazing technological innovations and announcements. Sometimes you're going to see multiple articles from different sources on the same thing. If you need NewFlashyShinyThingOMG constantly...maybe Gizmodo would be more your speed.
No, you guys. You don't get it. Bitcoins are going to take over the world. We're ALL going to be part of an economy of Bitcoins, just like we all universally adopted Swatch Internet Time[1]!
The future economy may not (I'm guessing will not) be bitcoin. It will probably have a few very different ideas built into the core of it. But the foundation is being laid right now as we discuss the problems of bitcoin.
Except for the guarantee of uniqueness. Which is one of the most powerful invariants against corruption and government-sponsored inflation, goals which are worth these effort.