I read some criticisms of Bitcoin (including those from a previous HN post about Bitcoin) and one of the biggest criticisms from an economics perspective (if I understand correctly) is that the Bitcoin network is at the mercy of positive or negative feedback loops (hyper inflation or deflation).

All current currencies are controlled by central bodies that can dampen the effects of these feedback loops.

Does anyone with an economics background have an idea of how the Bitcoin network could be adapted to counteract these feedback loops? For example, perhaps the network should make the transfer of small amounts very liquid whilst the transfer of larger amounts becomes progressively more "viscous".

How is it at risk of a feedback loop in a way that other currencies aren't?

It seems bitcoins defend specifically against hyper inflation by limiting the number of coins that will ever be produced.

Decreasing supply of a good with a constant or increasing demand, ensures its increasing value. The govt is able to control the supply of money.

Real-world currencies have backing. The ability of dollars to hyperinflate or hyperdeflate is bounded by the fact they can be used to pay debts to the United States Government that are denominated in dollars. Their value recursively emanates out from there, but that's the base case. That's also why the fall of the US government would be pretty much immediately followed by the fall of dollars, as has happened to other government-backed currencies. Stripped of its backing, currencies completely collapse in practice. The recursive base case is gone and if the currency briefly holds on it's a Wile E Coyote off the cliff moment, and in the modern connected world I can't imagine it would be more than a few minutes.

At some point, BitCoin faces some dip in currency confidence. Not because it's "BitCoin", but because all currency face periodic confidence crises. All the major world currencies have faced them in the past three years. But they didn't simply collapse because they have backing. I don't know what stops the first BitCoin currency crisis from completely collapsing the currency, because when BitCoin holders ask themselves, "Hey, what is this really good for?", I don't have an answer. I know what a dollar is good for: Not being put in jail by the United States Government due to unpaid debt. Ultimately, without trying to be too philosophical, dollars are backed with men with guns (and the privilege of them not being pointed at you).

Contrary to apparently popular belief, currencies aren't merely arbitrarily-agreed-upon numbers that we all trade with. Every currency I know has some actual backing. Even the fancy electronic ones we've seen spontaneously develop online, they all have some form of local economy-appropriate backing. It isn't always "gold" (literal or electronic), but so far the "men with guns" approach has proved successful in the real world. BitCoin is an apparently-good design for the arbitrary number approach, but until it solves the backing problem I won't be putting one dime into it. As much as I don't particularly enjoy having the US government as my currency backing I do not see how going to a backing-free currency is the solution.

The currency can still hyperinflate even if it is physically impossible to produce more by virtue of people raising all BitCoin prices as their confidence in the currency collapses. But instead of hyperinflation producing lots more BitCoins, then collapsing the economy, hyperinflation will simply directly collapse the economy as it takes ever increasing amounts of the BitCoins in the world to buy a service, until eventually even every BitCoin in the world isn't adequate. Hyperinflation is a symptom of lack of confidence in a currency, not a cause. (Of course observing hyperinflation can further decrease confidence, but the hyperinflation started in the first place because of lack of confidence.)

BitCoin advocates tend to get very angry when I point this out. I think it's because they have no answer to this. The site used to have a FAQ that addressed this, but it just sort of mumbled words and now it's gone because it was actually better just to ignore the problem. My challenge would be to anyone who claims that currencies are somehow backing-free is show me the successful currency of any kind that really is just an arbitrary number with no backing that has reached any sort of significant size, shall we say, a million dollars or so worth of an economy? (All the electronic currencies like Microsoft Points are well above that size, for instance.) And I'll show the backing. It isn't always a physical item, especially in the electronic world, but it's always something that serves as a locally-appropriate backing.

("So why's it worth so much now?" It's in a bubble generated by all this publicity. What happens the first time this bubble even threatens to pop? That's when you'll really find out who is right, me or them. Oh, and I'd predict an even larger burst of publicity and public braggadocio if it looks like that's going to happen; that will be the only way to forestall the inevitable another few days.)

BitCoin advocates tend to get very angry when I point this out. I think it's because they have no answer to this

Who got angry over this? I happily concede that bitcoin have no intrinsic value.

When the price of bitcoin rose so fast, it is only natural that a correction followed.

At some point, BitCoin faces some dip in currency confidence. Not because it's "BitCoin", but because all currency face periodic confidence crises.

Let see if your hypothesis bare this out.

The part you should have highlighted at the end is where I predict collapse, not where I predict it'll have a confidence issue at some point. There's no way BitCoin will avoid some sort of issue at some point and just monotonically increase in value forever and ever, amen. If nothing else, after its inevitable and mathematical wild success (cough) BitCoin will one day attract the attention of the IRS who will decide they need to tax it. If you manage to get that far without a confidence crisis, that's an automatic one. You'll lose a chunk of people for whom the primary draw was that the IRS didn't care about it. Confidence crisis isn't the question, collapse is.

(And that's the gentle one. There's also "The US government has decided that BitCoins are primarily a money laundering operation" and it starts throwing people in jail for using them. I consider this a less likely outcome by far, but still on the table.)

When I say confidence crisis, I do mean that as a distinct thing from a collapse; as I said, all currencies face confidence crises.

The part you should have highlighted at the end is where I predict collapse,

Fair enough.

Since your argument is so good, I linked to it on the bitcoin forum. http://www.bitcoin.org/smf/index.php?topic=4832.msg70583#new

I don't know who the bitcoin advocates that hate hearing this is coming from but the bitcoin community had always debate possible attack vectors and ways to counter it, if indeed it was a problem. Indeed, the government's response is a constant worry and there had been various opinions as to how big of a problem it is and how to win.

Running away from problems does not help us or anybody. It's best to confront it head on.

Indeed, nothing ever monotonically increases, or keeps growing forever. Didn't we already made that mistake with the housing crisis? There will be drops, eventually, maybe even big ones.

I think the biggest confidence crisis possible for bitcoin would be a security breach discovered in the software.

Other things, like the IRS interefering, will cause smaller panics. There will still be a lot of countries left where it can be used. Bitcoin is by no means a US-only operation.

The surprising thing for me for bitcoin is that it's actually being used to trade things, and is not just seen as an investment vehicle. This makes me think it can succeed.

So your argument is basically that government comes to tax bitcoin, or make it a part of a legal currencies, bitcoin will suddenly drop it's value? You could argue the opposite just as successfully - bitcoin will increase it's value even more because it will become a de facto legal currency (adoption will increase massivelly).

And the "bitcoin = domestic terrorism" attack is quite hilarious (in it's futility) if you compare it to war on drugs. Demand won. You can't make war on something large part of populations want and win.

"So your argument is basically that government comes to tax bitcoin, or make it a part of a legal currencies, bitcoin will suddenly drop it's value? You could argue the opposite just as successfully..."

Only with a heaping helping of wishful thinking. In the real world, when a currency incurs previously-nonexistent liabilities (in the accounting sense) the reaction of people isn't going to be piling in even faster. It may not kill BitCoin but it isn't going to be a moment where the value rises.

"And the "bitcoin = domestic terrorism" attack is quite hilarious (in it's futility)"

For the record, I disagree with the argument that alternative currencies are solely for money laundering or terrorism (note you added the terrorism connection, I just mentioned money laundering). I'm simply saying the government may make it. If you're going to be paranoid and cynical, do it right. If the government perceives BitCoin as a threat to its power, it isn't just going to come out one day and say "We perceive BitCoin as a threat to our power and so we're going to just stomp it out." They're going to have some reason with vague plausibility for enough people to give them cover to do what they want. (In fact I think that there may not even necessarily be any one person who thinks to themselves BitCoin is a threat to the dominance of the US Government and we must come up with some pretext to stomp it out, these things can sort of emerge from the successfully-evolved system itself.)

I actually approve of alternate currencies and expect that they will exist in the future regardless of what governments say. I think the embryonic versions already exist and the technological trend is unstoppable. I just don't think BitCoin is it, as it is today. Someone pointed out to me that someone could take BitCoin and actually provide some sort of backing, and I think that would be a potent combo, though given the "men with guns" option isn't really available that seems to only leave physical assets, which is tricky to pull off at scale. (You would need to actually be ready to provide all the physical assets if there is a run on the currency, no excuses, no clever contractual "no we didn't really mean it", you actually have to have it. Perhaps ironically, if you can, you may never have to, but if you can't, you will certainly have to.) If there's some sort of third option, someone might be able to actually provide the recursive base case and put BitCoin on a firm footing.

Totally agree with you on possibility of governmental attacks on bitcoin, thats why i added the war on drugs analogy, and it doesn't quite matter what political moral high ground excuse will be used to attack bitcoin, be it terrorism or money laundering. That's not the point.

My point is that Bitcoin has backing that is even more powerful than men with guns or physical assets. That backing is subjective value of Bitcoin in people's minds. Think Apple or Luis vutton, it's valuable because people think it is, and if the ideas on which Bitcoin is based are a solid foundation, the value of Bitcoins will only grow.

So for your third option of backing i'm thinking in terms of http://en.wikipedia.org/wiki/Subjective_theory_of_value the "men with guns", or "gold in storage" has the same value as "men who combined great ideas(PGP P2P Crypto Currency) and produced value (Bitcoin)".

Also Zimbabwean dollar is backed with men with guns, and it doesn't guarantee its value.

Id smuggle a mobile phone into my jail cell and trade bitcoins from a nice warm bed with 3 meals a day.

Making bitcoins illegal would make them worth more as drug dealers and pimps would value their qualities more.

tl;dr bittorrent.

Hows shutting down piracy working out for you ?

The IRS taxing bit coin will have about the same effect as piracy being illegal.

It's too easy to use it anonymously and too hard to crack down on it for there to be any real effect.

Can you point out the diff between that and the stock market? Isn't a stock in the stock market not really backed by anything other than people's speculation?

A stock is backed by companies' performance, I think.

Similarly to jerf, I think if you point this out to someone "in the know", and dig deep enough, they won't get mad, but it'll turn out that it's "really complicated" and "you wouldn't understand unless you were heavily involved". It happened at least once, and I wouldn't be surprised if this was common.

Yes, that's essentially correct. And that's why stocks can plunge and lose practically all (or all) their value relatively quickly, as happened to General Motors and many banking stocks in 2008, for example.

Edit: To clarify, even when a company pays dividends which holding the stock entitles you to you are speculating the company will continue to pay, or be able to pay, which is why you continue to hold the stock. This is why stocks are risky -- they have no guaranteed value (or backing) and can crash theoretically at anytime.

A stock should be backed by all the assets of a company, shouldn't it?

Stock in a company is equity. You own a slice of the company. If the company pays dividends, you are entitled to a portion of its future earnings.

Not at all. A stock represents a share of a company's earnings, which has real fundamental value. There are bubbles and random swings, but at the end of the day, a stock is worth something like the present value of future dividends. (Even for a non-dividend-paying stock - the idea being that eventually the company will mature and start paying a dividend.)

One thing I like about your post is that it gives me hope of understanding how, if Federal Reserve Notes are based on "nothing" since the 70s, our currency hasn't collapsed entirely yet? I love Ron Paul but he's got to explain this one to me before I buy into it wholesale.

You say that it's because the US government forces other people to accept it as payment for debt? I don't understand how that enforces any particular value. It enforces that it's accepted, but what's to stop someone from asking for 10 times more than they do?

Or do you mean that for a debt incurred at time A at a given interest rate, it's guaranteed to be a certain ("inflated", but predictably so) amount by time B? What about with Bitcoins? The owner of the debt could lose credibility if s/he decides to change the contract and say "Bitcoins are looking bad these days, you owe me gold now instead". But I guess guns work a lot better than reputation loss against a currency collapse. Hmm.

Or did I misunderstand your point altogether?

I do believe you misunderstood the point. Let me try to clarify from my own understanding.

There are legal tender laws, but these are not the point. They are largely irrelevant.

The point is that the US government itself accepts dollars as payment for debt. This may not seem like a big deal, until you realize that governments are unique in their power to force a debt onto you via taxation. Just like their monopoly on physical power, they also have a monopoly on economic power - for a number of reasons, which are not really important here, but the concepts are indeed very similar.

So the government forces debt on people, which can only be paid using the dollars that the government issues. This creates demand for dollars and therefore value.

Side note: You can learn an important point lesson from this. In a fiat money system, taxation has nothing to do with financing government spending. Government (by which I mean the union of all governmental institutions, i.e. executive + legislative + Fed etc.) can spend whenever and whatever amount it likes. The point of taxation is to create a demand for currency, which entices the private sector to offer goods and services for sale in exchange for that currency, thus enabling government to execute its mandate.

Returning to Bitcoin, there is no entity that can force Bitcoin-denominated debts onto people, and therefore the only people using it are doing so either out of curiosity or out of ideological dislike of government-run monetary systems - or perhaps out of the desire to make money off people who are simultaneously fools and members of the former two categories.

This obsession with 'backing' demonstrates a gross failure to understand the subjective theory of value.

What backs gold??? Nothing other than the subjective valuations of individuals. The same is true of fiat currencies, except here individual subjective valuations concern matters such as 'not getting arrested' or 'not getting shot.'

Bitcoin is a technologic, cryptographically-rooted, informational commodity. It has unique, desirable properties inherent to its design and structure. Individuals subjectively value these properties. Their reasons and value scales differ, but the simple fact that people are already trading Bitcoins proves that it is subjectively valued.

You talk about 'confidence crises.' This is unique historically only to fiat currencies (and paper currencies built fractionally on commodities like gold). The subjective valuation of fiat currencies is derived from political considerations. As such, confidence in them is built on the ever-shifting sands of political perceptions.

When was the last confidence crisis in gold? There hasn't been one. People value gold for its physical properties defined by the laws of nature. Those properties do not change and individuals have continued to subjectively value them.

The same is true of Bitcoin. Its properties are rooted in our present understanding of cryptographic principles. As long as individuals continue to value properties such as anonymity, decentralization, finite supply, low transaction fees, ease of digital manipulation, ability to integrate smart contracts, etc... we have no reason to expect a 'confidence crisis.'

This will be true as long as the cryptographic logic buttressing the system remains sound – just like the laws of physics underlie the desired properties of gold. Breaking this cryptographic logic (akin to cheap transmutation of lead into gold) would require breakthrough advancements of our knowledge of cryptography and discoveries on hitherto unsolved mathematical problems. But even in this case, the open-source nature of Bitcoin allows it to evolve new cryptographic implementations that would avoid such problems.

This is pretty insightful, the part about hyperinflation vs confidence.

I'm not a BitCoin supporter, but I still think you are wrong on this count.

Confidence depends on perceptions. Nothing can command confidence consistently if the underlying economy is bad or broken. In the case of BitCoins, not being able to artificially increase the supply is a virtue. (ps I'm not a supporter because bankers always figure out a way around limited supply of a precious commodity).

> The ability of dollars to hyperinflate or hyperdeflate is bounded by the fact they can be used to pay debts to the United States Government that are denominated in dollars.

This is not really 'backing'. Backing is when your currency can be traded for a specific amount of a specific thing. Your currency actually 'stands for' something else. Saying the US Dollar is 'backed' by the fact that you can pay debts to the US govt is isomporphic to bitcoins being 'backed' by the fact they are anonymous and you can trade them for DNS hosting.

And fwiw.. the bitcoin economy is roughly ~4.5 million USD right now.

I don't know enough about economics to give a qualified answer but let me try. All currencies are vulnerable to feedback loops but governments can do things like change the interest rate to encourage/discourage spending. I don't understand how this would be possible with Bitcoin.

I overlooked the Bitcoin limit - price inflation will certainly be limited through this mechanism. But I wonder if this might not cause prices to drop at a constant rate (since Bitcoins are so divisible) - this would include labor costs (although how wage stickiness would hold up to this remains to be seen).

Read about money creation[1]. It is not a conspiracy that it relies, simplifying, on debt. One wants to bind somehow an amount of money with the output.

Right now, BitCoins ignore the output. Even if there is no limit, it links the amount of money with either the computational resources or number of miners. There is no way to fix it. The limit is an ironic bounty.

[1]. http://en.wikipedia.org/wiki/Money_creation

I think the answer is that multiple competing currencies give stability. Hayek wrote a book about it. So Bitcoin by itself might not be stable, but Bitcoin in an economy that also has other private currencies might do just fine.

When I read about BitCoin a while back I thought it might be a cool startup idea to build an "online banking" interface around it, along with mobile apps, etc. To sort of, bring it to the masses (non-technical folks). One analogy might be Github which took another burgeoning open source project and made it more accessible through an excellent web interface.

However, I got discouraged after reading some things about the failed e-gold project: http://en.wikipedia.org/wiki/E-gold#2008_court_trial

Namely that it looks like you'd need to be licensed as a money transmitter to avoid serious legal trouble if the thing took off. And I assume being licensed as a money transmitter means reporting identifying info on who is sending/receiving money. Anonymity seems to be big amongst the bit coin community but may not be as important to the masses. Not sure.

There may be some way to avoid it, sort of like how torrent sites don't keep any copyrighted info on their servers, or actually going through the process of getting licensed, but I assume there would still be tons of legal scrutiny just like the torrent sites receive. The idea of running a startup where you are being sued by everybody is not especially appealing.

Has anyone else pondered this?

Yes, I wanted to do it, just to demonstrate that BitCoin is just as susceptible to problems of money supply as any other currency.

The basic idea is to allow people to deposit BitCoins, but then issue a BitCoin derivative subject to fractional banking rules and charge interest. This should in theory give rise to the same problems current banks have, where the debt is never fully repayable (unless you create more money to pay the interest), and requires a banking system to continually generate debt.

watching a cartoon doesn't mean you understand banking.

FUCK "money as debt". talking people down from their newfound expertise is tiresome.

Hardly. Bitcoin is somewhat of an improvement on gold, since you don't have to dig it out of the ground, only to bury it again and put guards around it.

However, the monetary system is already hacked. I'm just hoping other hackers have some insight so that we can have a more effective, secure currency.

How is it hacked?

By hacked, I suggest that there are loopholes which allow some body/some group to have undue influence on the creation, destruction and distribution of money. In responsible hands, this level of undue influence doesn't lead to negative outcomes, but in irresponsible hands it is disaster.

For instance, where we see hyperinflation in Zimbabwe, it is really about the government gaining "root" access on their Reserve Bank, and "creating" money, while diluting what others already have.

On the other hand, when business confidence is low, people will store their "money" away for a rainy day, but that can lead to people sitting on the sidelines rather than engaging in production. i.e. people are poorer for it.

The other thing about money is it is reputationless. Although governments have a notion of tainted money, or money which has to be "laundered" to become legitimate - when white collar crimes are committed, let us say Enron-style, the CEOs still have money that they can spend and lead a good life. Contrast this with some aboriginal cultures which were moneyless, reputation is very important, and reputation is currency in its own right.

Perhaps the side effect of Bitcoin tracking transactions is that money no longer becomes reputationless. Society can choose to shun money that has a tainted reputation.

I don't really know of any solutions. I'm just bringing this up for discussion.

Your portrayal of hyperinflation in Zimbabwe is questionable.

To understand Zimbabwe, you first have to understand that it suffered from a destruction of supply capacity due to a stupid land reform that had nothing to do with money. The motivation for reform was part good (white farmers who had profited from colonialism were disowned) and part bad (land was given to political cronies who had neither the knowledge nor inclination for farming). Once the supply capacity went to hell, price increases were a simple matter of supply and demand.

Add to that that Zimbabwe had to start importing food, affecting its current account and causing their currency to drop w.r.t. foreign currencies.

Add to that that Zimbabwe had significant US$-denominated debts at the IMF, and the required sum flow of US$ for debt payment plus imports out of the country simply overwhelmed the amount of US$ they could possibly hope to have flow into the country from exports - their exports basically collapsed, because the productive capacity of the economy was largely destroyed.

At that point, instead of doing the only reasonable thing and defaulting on their debt, the Zimbabwe government decided to start the printing presses and print Zimbabwe currency, largely for the purpose of directly exchanging it against US$. Needless to say, supply and demand caused the Zimbabwe dollar to drop in value, which - via increased import prices - caused domestic inflation.

The important lesson here is that initially, it was the hyperinflation that caused the printing of money, not vice versa as people usually assume. Of course, the fact that the government kept adding to demand helped the hyperinflation to continue. But given the desolate state of the economy, simply not adding to the demand wouldn't have fixed anything either.

I agree that sanctions exacerbated production problems.

However, it doesn't detract from the point that money is hackable. I don't know if it is solvable though, because governments have coercive power over what constitutes legal tender.

Interesting. Must be a scheme based on infinite supply of the said derivatives and, thus, the ability to create a lender of last resort with a printing press, right?

Won't work, as far as I can see (one of the reasons being the absence of legal tender law for your derivatives). If your derivatives are redeemable in bitcoins, bank run is still possible (and easy). If it's not, then you can't really affect the money supply of bitcoins.

Or is it some other kind of scheme?

Bitcoins still need to be securedly stored, and this is beyond the capability of 95% of the world. In the end you'll have a consortium of "guards" who will offer storage services. If humans in the future are like humans of the past, then fractional banking is assured. If there is an economic boom, then I can assure you that these new "banks" will offer letters of credit, which is essentially a derivative.

That's not enough for FRB to work. However big your consortium, bank run is still a very real possibility (the history of Bank of England has some insightful examples, Nixon closing the gold window is another recent one [and it wasn't even redeemable dollar! it was a restricted bullion standard.]).

FRB can exist long-term only if bank runs are impossible. The only way to make bank run impossible is to have lenders of last resort which are able to redeem all the notes in circulation.

With gold-backed notes, you can't issue more notes than you have gold (all the attempts to do so historically ended with bank runs pretty quickly). With purely fiat currencies, it's not a problem (you can print as much as you need to cover whatever derivatives are afloat).

Bitcoin (or vanilla gold-backed money) does not prevent FRB per se. It just makes sure that FRB will be very limited and will usually go down with a bang fairly soon.

> Bitcoin (or vanilla gold-backed money) does not > prevent FRB per se. It just makes sure that FRB > will be very limited and will usually go down with > a bang fairly soon.

What you say makes sense. Since Bitcoin has the same properties/faults as gold insofar as FRB is concerned, why is bitcoin interesting?

Gold is susceptible to global adversary (government) attacks. You can't have instant payments over internet with physical gold -- you need some kind of virtual banknotes to change hands instead, and to be redeemable in physical gold sometime later. That would be perfectly fine in a world with small, reasonable governments, but in the real world we live in, it's not. Two problems arise:

1. Sending physical gold for payment clearance, especially though state borders, is problematic (just plain illegal in some countries, may be taxable, etc). Doing it often and in small portions is also economically unfeasible [you need economy of scale to lower security costs, logistics, etc]. The usual market solution for that is creating big clearinghouses, with branches everywhere, etc. But:

2. Running a big clearinghouse for gold makes you effectively a bank. Running an unregistered unregulated bank is a crime in most countries. So you have to register, get all the licenses, keep all government requirements fulfilled, you can be subjected to all kinds of inspections, etc. On top of that, in some countries gold payments violate legal tender laws or something else, and since you are already on gov radar, it's a non-starter. After E-gold crackdown, we know these are not purely theoretical concerns.

Bitcoin is free from these limitations. Instantly redeemable, no vulnerable physical storage, distributed, no need for high profile points of failure, etc. Bitcoin still lacks one of the standard requirements for commodity money, though ("must have some non-monetary use/value"). How critical is it? We'll see :)

OK, I take it that from government attacks you are referring to gold seizure in E-Gold and during the Roosevelt era.

I don't know if this can be avoided. Governments could coerce the release of private keys so that the BitCoins could be reassigned to another party. (by the way, what happens if the private key is lost?)

During the Roosevelt era, it is simply outlawed, and gold has to be exchanged for fiat currency. I'm not sure if BitCoin can circumvent that.

To seize noticeable amount of bitcoins, you have to work through individual bitcoins holders. Bear in mind that bitcoin transactions are not channeled through any central points under gov control, nor are they automatically personalized -- so the gov must first somehow find out where to look. The whole process will take months, and the borders are effectively transparent for coins all the time. If you know the seizures are underway, you may as well arrange for your coins to stay out of your access and U.S. jurisdiction for a while.

To seize noticeable amount of gold in gold-backed paper money system, you have to issue one order to all registered banks. It takes a day or two.

But that's besides the point. Nothing will circumvent the confiscation scheme in which the attacker has power to do anything they want to you, and you can't do anything to protect yourself. It's simply not the problem to be solved by monetary system.

This already exists: http://bitcoinusa.com Easy-to-use web service, from a legitimate company that complies with all laws.

e-gold primarily failed in the marketplace, like all the other gold currencies. This was really of a piece with PGP and the other cypherpunk notions: they didn't take off because people were not yet invested enough in their digital lives to care. It was largely experience with this that brought me to the conclusion that the end of privacy is inevitable.

I wrote the software behind two of e-gold's "major" competitors, and contracted for e-gold as well very briefly, so I'm not exactly unbiased, by the way. I don't know much about BitCoin, though.

BTW, this video was motivated by the first and largest bitcoin bounty which were supported entirely by user donation, in the form of bitcoin. See the thread at: http://www.bitcoin.org/smf/index.php?topic=697.0

I am an escrow agent for this animation bounty that is currently worth 8472.05 BTC(from memory) and now that's worth like six thousand USD. Originally, it was worth around two thousand dollars.

What's the quickest and easiest way to buy a couple hundred BTC?

For small amount of bitcoin, you can use CoinPal at http://coinpal.ndrix.com/

Even though I did not personally used it, many people recommend the service.

There's also #Bitcoin-OTC where trading is done over IRC. Here's the Order Book, viewing USD offers: http://bitcoin-otc.com/vieworderbook.php?sortby=price&so...

1. Open an account on any Bitcoin exchange (mtgox.com, bitcoinusa.com, etc)

2. Fund your account by wire or ACH transfer (supported by most exchanges)

3. Place a 'buy' order in 2 mouse clicks to buy BTC at the current market rate. Done.

Or if you know someone in the real world who has coins, just make him a cash offer to send you directly the BTC (using the Bitcoin software, without going through a third party such as an exchange).

I like the idea of Bitcoin, but I've been running the client for about two weeks on my quad-core machine and still have not generated any Bitcoin. It's a little disappointing!

Yes, I know I could buy some, but until there are enough venues that accept Bitcoin, I'm only motivated enough to get free Bitcoin. The only one I've seen that I'm interested in so far is a site that lets you bet Bitcoin on pro Starcraft 2 matches, although the idea of buying Call of Duty: Black Ops for Bitcoin is pretty good too.

There's no such thing as a free lunch.

When you're using quad-core machien to generate bitcoin, you're probably wasting electricity. That doesn't stop people from pooling their resources together and mine, though.

Lol, CPU mining is not sensible, use GPU. And why do you think that it would be easy to get free money? Bitcoin value is pretty high, so there are people running GPU farms & mining it. Competition is high.

I think currently mining is profitable, you can pay your hardware investments back maybe in 2-6 months.

you can pay your hardware investments back maybe in 2-6 months

Can you clarify the logistics of this? Do I borrow BTC, find a hardware vendor that accepts it as payment, then repay with what I've mined? Or do I find a lender that will give me USD and accept BTC in return? Or do I spend USD on hardware and put the resulting BTC towards my regular costs of living? I doubt my landlord/utility providers would accept rent/bill payments in BTC. I've seen the "trade" page on the wiki, and it only lists a few dozen merchants, generally dealing in web site services and luxury goods.

You can sell BTC for USD.

https://mtgox.com/ - liberty reserve http://coincard.ndrix.com/ - paypal

A two year old video card can crunch numbers faster while using less power than a 2.8ghz Core i7 Quad.

Two things to check out that you can find on the bitcoin forums: pooled mining and GPU miners

With both of those, you can earn 1-2 BTC/day on an older video card.

Ah, ok this makes a lot more sense. Getting started with DiabloMiner and http://mining.bitcoin.cz now.

I don't use Bitcoin anymore (now that it's popular, I'm a cryptocurrency hipster), so I can send you the few BTC I have remaining if you want to play around with them. Just leave me an address. :)

Hey, much appreciated! Here's my special HN receiving address:

1HzJQSHcRTbMmuT3Hb2myDEybDwnD4LgGp

You have 5 BTC incoming once I catch up on crunching the block chain (I'm a few thousand blocks out of date) and there's a few confirmations. Have fun!

hey, I'd like to play around with them if you have some to spare! 18CsFA1WJh4zCMTPWoNcwmzfod9vrdZkNS

I only have 0.07 left, but you can have them (minus the 0.01 transaction fee). :)

http://www.alloscomp.com/bitcoin/calculator.php

There was a pretty big spike in difficulty after BTC reached $1.

If your speed doesn't allow you to get a block at least every few days, then you should use bitcoin pool,

Bitcoin mining pool with currently open registration: http://deepbit.net/

I don't understand why mining is so computationaly intensive. Well, I understand that it's used as a way to limit the rate that new coins are introduced, but are all these mining computers doing anything useful, or are they just wasting electricity?

Right now the bitcoin network is calculating at about 500 GHash/second. Only 1 hash is needed to process an entire batch of transactions (a block). Well, technically it is 2 SHA hashes that are needed per block. So 499,999,999,998 of the hashes calculated in the next second will be wasted.

It gets worse. That is 500 GHash per second. Over the approximately 10 minutes between blocks, only two of those 300,000,000,000,000 hashes that were calculated will actually be used even.

Of course, there is a reason for this. Hashing to obtain a relatively unique result is called a "proof of work" and is the method that Bitcoin uses to ensure that the amount of currency issued occurs only at the levels specified in the source code. http://en.bitcoin.it/wiki/Proof_of_work So far, that method has been the only one found so far that works as the transaction processing system for a decentralized payment network.

They're doing something useful.

1. They secure the network from attackers and check for double spending attempt.

2. They process transactions.

If you can implement a bitcoin clone without the computational load, it will be a revolution. But this is the best we've found so far: it secures the network, records the transactions and does the initial distribution of coins.

Bitcoin needs a problem (any problem) that is computationally difficult to solve in order to be able to offer certain security guarantees. This arbitrary problem was chosen by its designer:

  For a given target T, find X such as SHA256(SHA256(X))  < T

It is correct that in and of itself this computation is useless. But exactly like http://en.wikipedia.org/wiki/Hashcash it is vital for it to be difficult to solve. In fact the whole network adjusts T dynamically to keep up with the fantastic exponential hash rate growth of the network these last 13 months: http://blog.zorinaq.com/?e=49

Could you use rainbow tables to speed this up?

No because parts of X are effectively variable, change all the time, and depend on things not controlled by any single entity (transactions, block chain, etc).

Technically X is a block header https://en.bitcoin.it/wiki/Protocol_specification#Block_Head... Miners change the 'nonce' part of X to attempt to satisfy the above equation.

Put simply: bitcoin mining is a race with hypothetical hackers. A hacker trying to create fake bitcoins would have to create hashes faster than the real bitcoin network.

I tried to get some free coins through the bitcoin faucet to a newly generated address (18YTP3wEbaiNX2QaQ4k1Gi89JgXntxAiff), but so far haven't received anything. Might be a problem on my end, or the transactions might just be slow, or they didn't send anything to me at all, but so far bitcoin transactions don't give the same instant gratification paypal or credit card payments give me, where you press the 'buy' button and get access to a service immediately. Perhaps this will get faster when the network becomes bigger, or if I use one of those digital wallet sites instead of running the client myself?

If a transaction fee is included (which the bitcoin faucet doesn't include), the process is sped up.

http://en.wikipedia.org/wiki/Bitcoin#Transaction_fees

Your client has to do some stuff before being able to connect and receive the notification. Keep it open for a few hours and you'll see the faucet coins.

If the newly generated address is on your local machine, and it's a new client.. you will have to verify the block chain before you can conduct transactions. Probably you are blocking it up.. I got my .05 bitcoins right away.

It's hardly fair to compare Bitcoin to credit cards in terms of "instant gratification", unless you turn back the clock to when you said, Oh, there are such things as credit cards? Cool, how do I get one? I'm confident that the difficulty of acquiring your first bitcoin will compare favorably to the difficulty of acquiring your first credit card.

You can't compare Bitcoin transactions to Paypal or credit card transactions. More analogous would be a Bitcoin transaction vs. dollars sent by mail.

Paypal and credit cards are just transactional layers on top a currency, they only represent the underlying currency. The actual exchange of the currency can take an arbitrary length of time (if it even happens at all).

Nothing precludes Bitcoin from having the same services built on top of it. When the time comes for Bitcoin to have truly instant transactions, the market will create this necessary layer.

There is ongoing attack on the network with a flood of low sum transactions ( http://bitcoincharts.com/bitcoin/ ). Because of this, transactions without fees are delayed.

What is the point if only 21 million can be generated? Are they infinitely divisible or something? Thats less than one bitcoin per person in California alone.

They are divisible to the 8th place. Even then, that still could be fixed.

Interesting, I should open a bitcoin bank.

Would gold and silver be valuable if there was an infinite supply of them? You can't print/generate money indefinitely (unless you think quantitative easing is a legitimate economic stimulus strategy and you happen to be the Federal Reserve).

If the desire for BTC goes up, so will its value. There's no problem here.

Money is a mean of exchange that happens to be a store of value. A thing that punishes the exchange is not an effective money.

In an economy with a finite amount of money, even potentially, the exchange is not rational given that the money's value grows over time.

That's why the focus is on generating BitCoins.

After all the BTC are generated, only market conditions and demand will affect its value. It's not guaranteed to deflate perpetually.

You can't print/generate money indefinitely

Why not? Our current currencies (which are a lot more useful than BTC) are not backed by any finite commondity like gold.

Exactly, they're fiat currencies: http://en.wikipedia.org/wiki/Fiat_money

Interesting, but until some real use case or adoption, these Bitcoins sound like Internet stocks to me.

edit: I didn't really look into it, but I'm going to now.

There's lot of real adoption. The bitcoin forum alone get 500 posts a day and a whole lot of traffic. Maybe you should check us out at http://bitcoin.org

I myself run a little magazine about bitcoin that is hosted on a VPS that I pay in bitcoin. I even own a domain name that I paid in bitcoin.

The economy is still small though, but with adoption, it could become real big.

No it can't the total money supply is limited to 21 million bitcoins, this will create drastic problems unless they are infinitely divisible.

They are divisible to the 8th place. That's 21 quadrillion unit of bitcoins.

And that could be increased indefinitely if the need arises.

So the economy could become dependent on everyone keeping their software up to date..... As a web developer this scares me.

I myself run a little magazine about bitcoin that is hosted on a VPS that I pay in bitcoin. I even own a domain name that I paid in bitcoin.

What kind of exchange rate on bitcoin to USD did you get on these purchases?

Last time it was 0.90 USD something. The exchange rate changes with the market.

Does getting Call of Duty: Black Ops(Steam Account) for $26 worth of bitcoins count as enough of a "real use case" for you? http://www.bitcoin.org/smf/index.php?topic=4807.msg70315#msg...

Ya, scratch that.

Which brings up a good point. With Bitcoin, there is no concept of chargebacks. Once bitcoins are sent, they are gone.

When you wish to make a purchase with a trading partner where trust has not yet been established, the http://ClearCoin.appspot.com escrow can be used to protect the buyer.

For this transaction escrow would have been free to use, as the amount is under 100 BTC.

Recommended reading on the kind of social changes a popular cryptocurrency like Bitcoin could usher in: https://encrypted.google.com/search?q=a+lodging+of+wayfaring...

How about a Hacker News clone, powered with bitcoins?

With witcoin, you earn money every time someone votes or replies to your post.

http://witcoin.com

So, wait. You can print your own money subject to a cap? ie. creating bitcoin only involves spending CPU cycles (or electricity)?

Money supply is limited to 21 million bitcoin ever. Plus, as more people bring more mining computers online, it become harder over time to mine.

While I understand that this is the current plan, what is the guarantee that this plan will not be changed? Is this a technical limit, or a social one?

The "plan" is implemented through the bitcoin protocol and software.

Any change would need 50% + 1 of the bitcoin network to follow the new rules.

An effort to change the software to "mint" a greater amount will be countered by those who already hold bitcoins, and can add mining capacity to resist the change.

Another implementation could be launched with a different limit, but that would be a completely separate, new currency.

Yes, but it's very very hard to do it now. Also it's limited and the more you print the less you will be able to print.

Not quite sure what the appeal is here?

Low transaction fees are nice but why not use something like Dwolla which is dollar denominated?

The "mining" concept also confuses me. Is the currency backed by CPU cycles? If so, what are the CPU cycles being used for?

Dwolla is great, ... but for smaller amounts their $0.25 per-transaction fee is a limiting factor. Especially if what you want to buy is only $0.99, for instance.

Also, when you receive Dwolla as payment and want to spend that money using your bank ATM/debit card, after you withdraw funds it then takes a couple days for those funds to reach your bank (via ACH).

For now, Dwolla is available for U.S. accounts only as well. Bitcoin has no such limitations -- it is just software and a network protocol.

The "mining" concept also confuses me. Is the currency backed by CPU cycles? If so, what are the CPU cycles being used for?

Bitcoin is not backed by anything. CPU cycles are used to ensure the security of the system.

Low transaction fees are nice but why not use something like Dwolla which is dollar denominated?

That's like comparing eggs to chickens. Bitcoin is more than just a transaction clearing house, but also a currency that is completely separated from the control of governments and traditional banking institution.

The advantage of bitcoin is that anybody on the planet can use it. I can send you bitcoin just as easily across the street or halfway around the world. Bitcoins cannot be frozen as easily as your bank accounts can be frozen. There's also the fact that bitcoin cannot fail due to one single company.

Not quite sure what the appeal is here?

It appeals a lot to technogeeks who happens to also be libertarians, people who worries about economic policies of governments, and so on.

But bitcoin in my opinion can facilitate greater efficiency in the world economy by making it easier to transfer capitals across border. That alone is valuable.

_It appeals a lot to technogeeks who happens to also be libertarians, people who worries about economic policies of governments, and so on._

I fit that description and agree that private currencies such as Bitcoin would be awesome, but aren't there laws in place to secure the government's monopoly on money? ( http://www.lewrockwell.com/paul/paul619.html )

Does Bitcoin circumvent these laws by being completely virtual?

It is impossible to have monopoly on money.

Simply put, money is whatever market participants agree on.

Sure. But the government can still insist that all transactions paid in BitCoin are subject to a sales tax, which is payable only in legal tender (not BitCoin).

Money is whatever the market participants agree on, and laws are whatever the government agrees on.

It requires cheap trackability to tax.

Of course, BitCoins are easily trackable, thus doomed. It was possible, though, to design them without this property, ensuring that costs of tracking offset the potential tax gains.

EDIT: It would be fun to consider the arbitrage opportunities rendered possible by such sales tax.

My understanding is that it's easy to track the transfer of bitcoins to accounts, but very difficult (impossible) to figure out who is behind an account unless they choose to reveal their identity. Since accounts can be generated on a whim, and with no identification, the upshot is that taxation would be unenforceable.

They do not think so[1]. It is easier to fail than with the old-fashioned money, given that all the transactions are publicly logged.

http://www.bitcoin.org/wiki/doku.php?id=anonymity

Mining is the method how transactions on the Bitcoin payment network get processed and stored. The only reason you would mine is if you wish to invest the significant amount of time, equipment and electricity necessary to do so. Compensation for the mining activity comes from "rewards" in the form of new bitcoin currency that is issued.

The more cryptographic hashing calculations that your hardware can perform, the more likely you'll earn the currency "generation" reward. The reward is issued roughly once every ten minutes. All the miners are competing to become the recipient of that next reward.

Because there is no central authority, the currency is protected by the network. The more mining activity, the less likely an adversary could mess with the payment network's transactions. Now that Bitcoin is among the largest distributed computing networks that exist, corralling the necessary computing power to act as an adversary is becoming something that is more and more difficult.

Okay that makes sense, thanks for the clarification.

Is the currency backed by CPU cycles? Essentially yes.

If so, what are the CPU cycles being used for? Essentially nothing. Just doing time consuimg calculations. Makes it hard to fake.

why not use something like Dwolla which is dollar denominated

One of the big advantage of bitcoin as I see it is that it is not pinned to a single country, it is a truly international currency. Pegging it to the dollar would reduce its attractiveness for many people.

While the video does look great, I'm afraid I still do not know how real-world value is attached to BTCs--it seems as if a 'step' is messing in the 'how it works' explanation.

I'm assuming that was the major desired end-result of this video; if it wasn't, it was the question I was hoping most to get answered.

Supply and demand. On the supply side, it's limited to 21 million bitcoin forever.

Demand is supported by people who accepts bitcoin. Currently, the price is around 0.80 USD per bitcoin. It used to be only 5 cents per bitcoin.

And the ones who own the initial supply are the ones cranking out the solutions to the problems?

Yes. Those who solved a problem get to send 50 BTC + transaction fees to himself.

Yes. Note that solving the problems is useful to the whole network: it records transactions and secures the network.

What's being solved are not just abstract mathematical problems, but equations that have implications on the whole system?

The "equation" is a simple cryptographic hash performed on a block of data, as described earlier in this thread: http://news.ycombinator.com/item?id=2358396

The equation itself is trivial -- it can be calculated hundreds of thousands of times a second by even a really slow CPU.

An artificially set difficulty is implemented and the result is that only one of those attempts across the entire network will succeed in about ten minutes, which is the targeted time for processing a batch of transactions (a block).

And it was well over 1 USD at one point (the first time the subject really blew up on HN).

Am I right in that Bitcoin isn't backed by computational complexity; but, instead the continued ability for machines on the network to communicate with one-another?

Bitcoin is backed by cryptography. The bitcoins in your wallet cannot be spent by anyone else.

Now if there were no network access, then there's a problem. I might accept your transaction but without access to the network I couldn't know that you didn't already spend those bitcoins elsewhere.

surely the answer to that is that I shouldn't accept someone's bitcoins without being connected to the network to be able to check if they were already spent. this kind of "offline" transaction isn't allowed with regards to a credit card, why would it be allowed in our system?

Yes, for bitcoin there are no offline transactions. A transaction is recorded by telling everyone on the network it happened. Then everyone knows which addres spent and recieved bitcoins.

IANAeconomist, but a currency is back by trade. You and I accept a dollar or a gold coin because other people accept them at (usually) parity for value-- call it a network effect.

But, a dollar or a gold coin has limited value if you can't trade it. The same for a bitcoin; except, it can only be traded via coordination with many other actors in the economy. (The longest block chain is the "truth.") While a dollar or a gold coin can be traded physically, a bitcoin is never traded until the economy as a whole agrees-- the record of the transaction appears in the chain.

Therefore, the act of trading bitcoins, and by extension bitcoin itself, is backed by the presumption that Internet access will be pervasive and unfiltered.

Reply on news.ycombinator.com