This is a great intro to bitcoin. It made me feel more optimistic about the prospects of the currency.
It highlights many of the risks and benefits but misses out on what I think is the Achilles' heel of bitcoin, the ability for sovereigns to shut down the exchanges.
Without the ability to convert bitcoin to fiat easily, the currency loses its usefulness as a proxy for fiat currency. Does anyone have a convincing argument for how the exchanges are not vulnerable?
The situation is like the chicken-and-egg dilemma.
Definition 1:
'Good producer' is a producer that accepts bitcoins.
Definition 2:
If all the good producers together produce all that you need, then the number of producers is 'enough'.
Definition 3:
'Ideal state' is when enough good producers exist.
Theorem 1:
A producer will accept bitcoins if either of the following is true:
i. Ideal state has been attained.
ii. There exists exchanges that trade you physical currency against bitcoins.
Observation:
Once we are in the ideal state, we no longer need to have bitcoins converted to physical currency. So we don't need the exchanges to exist forever. We need exchanges only till enough producers have started accepting bitcoins.
Except the only ideal state is when you can buy anything in the world, not just "enough."
The great thing about good currencies is they can be exchanged for almost any currency and therefor can be used to purchase almost any good or service that is for sale.
Good point. I agree that major exchanges are vulnerable to regulation and shut down. I think localbitcoins.com (cash in hand or private transactions for bitcoins all over the world) is probably part of the answer (but not a very liquid or effective answer yet).
Even if BitCoin can't be exchanged for some fiat currencies, there is always the prospect of exchange through some middle step: either less restrictive foreign currencies, or commodities such as gold.
At this point, a full-scale governmental assault on BitCoin could slow it down or scare some people away, but it's highly doubtful that the growth could be stopped altogether. The bigger danger is that they'll embrace it wholesale and start taxing BTC exorbitantly instead. :P
Aren't the exchanges only necessary initially, for some definition of "initially"? Surely the goal is that, eventually, enough people are mining (or even receiving bitcoins as payment for labor) and trading among themselves for goods/services without any other currency in the picture.
There will always be other currencies, and currencies that are not exchangeable are less desirable and therefore less valuable.
Do you care if you are paid in Chinese Yuan or USD? The currencies are interchangeable, but Yuan is very hard to convert it USD and you'll lose a lot when you do.
Regardless of how big bitcoin gets, everything will not be transacted in bitcoin. People try to accumulate capital to buy big things, like cars and houses. If you can't buy a beach house in Costa Rica - and any beach house, not just from a few sellers - then the currency is less valuable and people will prefer fiat currencies.
Bitcoin must be easily exchangeable to be widely adopted.
Does anyone know of an explanation of the bitcoin protocol in plain English? I'm a non-math guy who's trying to write a legal treatise on the use of bitcoin, and I'd like to give some background on how it works to my readers. Can anyone help?
Thanks. I'm not a legal student, I'm an attorney doing some research on the issue. I was just looking for some help understanding the more difficult math concepts. Sorry that I don't live up to your standards.
"Bitcoin is obviously not the solution to these issues, nor will it be a key enabler to crime. It is just a currency with some interesting features on top: peer-to-peer, security, public transactions and potential anonymity."
Not a key enabler, but I think criminals would definitely appreciate a hardly traceable and potentially anonymous currency, don't you think?
It's not an opinion unfortunately, there is already evidence floating around that they are using it or that they are very interested about it.
Because of this I don't think bitcoin should replace a national currency anytime soon, can you convince me otherwise?
I can't. I think criminals will find their way with Bitcoin or without it. It's easy to fall for the easy argument: "the standard banking/monetary system is even worse, so bitcoin is ok". As a small example HSBC was considered too big to jail (they laundered last year about twice the size of the Bitcoin Economy, helping criminals, drug lords, etc), and only got a slap on the hand (a $1.8bn fine). The powerful can launder their money, do tax evasion, use tax havens ... now it's available to all technicaly skilled ones (good or bad?). Drugs sold on silk road were evaluated (http://arxiv.org/abs/1207.7139) last year to about $14m/year ... a good 20% - 30% of the bitcoin Economy, so yes it's looking pretty tainted so far. I would however highlight one innovative point: If you slip (leave any trace, don't use TOR, use a physical address, leave your IP), the community can track you down! And I think it will happen. Because ledgers are public, you can make algorithmic criminal enforcement :)(for those criminals or corrupt politicians who are not careful enough)
I also think criminals will avoid it like the plague. Bitcoin could conceivably tell law enforcement everywhere you received money from and everywhere you spent it. I realize it is one long number transacting with another long number, but law enforcement have computers too, and they have time on their side. Using bitcoins could be like leaving DNA on the scene of the crime - probably not a good idea if you want to get away with whatever you are doing.
If I understand the idea correctly, using a different key pair for every transaction should help mask user histories. Not sure if this is easy in practice.
I just got to wondering, how would charging for a monthly service work in bitcoin? A user can't just give you an account number and say bill me $X every month, right?
A wallet service which you've entrusted your coins to can obviously do recurring payments (bonus: can't hurt you any worse than your total deposit).
Your local wallet software could do this (in a more user-friendly fashion than programming a crontab to invoke bitcoind).
A credit card would work, regardless of whether you settle your account with the credit card company in bitcoin or dollars.
The Bitcoin protocol supports future-dated transactions, but which can be canceled by a user at any point before the future date; so for a 'subscription', a user sends a few transactions worth $X future-dated to each month for the next, say, 5 months. The service holds onto them, and tries to commit them when they mature; if the transaction works, the user gets access for the next month. (See https://bitcointalk.org/index.php?topic=22772.msg287925#msg2...https://en.bitcoin.it/wiki/Contractshttps://en.bitcoin.it/wiki/Script )
Technical details aside, I don't want to be tied into paying for a service in a currency that fluctuates 30-50% a month. I think that's by far the biggest problem bitcoin currently has.
I suspect in the long run most of the bitcoin economy, if there continues to be such a thing, will work through payment processing services.
These will resemble the banks of today, in that you'll have an account with the service and they'll manage moving your money around for you. Some of this may end up just decreasing one account and increasing another within the same provider.
A lot of people assume these payment services will be just like banks, but that's not necessarily the case. There's certainly room for innovation, and so far there's nothing whatsoever that would require them to be subject to any particular nation's regulations. We'll see how it shakes out.
The simplest solution is to send a monthly invoice like a mailed utility bill.
It might be neat to create or extend a bitcoin client that accepts invoices/requests that you can approve like many bank's bill pay services, but without the middle man. You would need to figure out how to prevent getting spammed with requests.
It probably isn't worth the effort though, because I'd bet that one day someone will just make create a credit card based on bitcoin just as they have for most other currencies (I don't think one exists yet).
Anyone can build anything really, so long as it conforms to the bitcoin protocol as implemented in the other clients.
So yes, you could probably write some sort of crypto-system that allowed continuous authority for a signed-address to be given a certain amount every so often, on request.
You wouldn't want to right now though, with the volatility in the value we're seeing right now you'd be a fool to agree to pay the same amount for something next month as you do this month.
A satoshi (not satochi) is 0.00000001 BTC, not 0.000000001 BTC.
Not sure where the $6,000/BTC comes from for $80b in annual transactions per 10m BTC, but consider velocity of money. A BTC can be spent multiple times in one year. Also consider that monetary base (raw currency) is not the same as a credit account, but you can spend either.
So in practice, you might only need 0.1 BTC per 1 BTC in annual transactions.
> A satoshi (not satochi) is 0.00000001 BTC, not 0.000000001 BTC.
I have no idea, visually, what the difference is between those two numbers, and I so suspect no one else reading does either, without manually counting the zeros one-by-one.[0]
Here's an idea: we sometimes use commas to the left of the decimal point to make the numbers easier to read, perhaps using apostrophes to the right would also make things easier to read? E.g.:
A satoshi (not satochi) is 0.000'000'01 BTC, not 0.000'000'001 BTC.
I can easily tell the difference now. (The idea behind using apostrophes and not commas is that you won't inadvertently confuse the right-hand side of the decimal point with the left.)
Anyway, carry on. :)
[0] There's also scientific notation, but that only works in cases like this with lots of zeros between the decimal point and the number in question. With a number like Pi, scientific notation is useless.
Good points about velocity and monetary base. "satochi" spelling and decimals corrected. Thanks. The $6000 is just an order of magnitude >>>(80/12)*1000. Too many unknowns for anything beyond guesstimates: What percentage of the currency is hoarded by then (1/2? 1/3? 3/4?). How many bitcoins are lost forever? As you point out velocity, or even limits on transaction speed from block-size ...
Even for a wild guess I think you're off by at least an order of magnitude if you don't take into account velocity (typical dollar is spent four times per year; edit: and bitcoin, as a digital currency, is likely to have higher velocity) and the different types of money (monetary base is maybe one fifth of total money supply).
Note also that your math involves 12m BTC but your page says 10m.
I may very well be off by a factor of 10. Agreed. But what is your estimate? and what exactly would M2,M3 and M4 be?
Let's assume 12m has been mined, 1m has been lost forever, 8m are hoarded, and a velocity of 10. You then have 30m bitcoins in transactions but 3m bitcoins circulating. In this case my estimate is off by a factor of 2 - 2.5 (but it could be 10, sure). But What if it doesn't take over half of paypal's transactions, but rather 4 times all of paypal's transactions? That's why I think any estimate is pretty hopeless. A good order of magnitude is ... if it reaches paypal scale it will be in the thousands.
Since there is no fractional reserve banking (yet, anyway) in the bitcoin economy, I don't see how the monetary base can be 1/5th of the money supply (but I might be missing something).
Is the part about it costing $25,000 Million to "trick the network" true any more, given the advent of ASICs into the mining arena? User 67117 on BTC Guild is currently processing 6,348.98 GH/s, or roughly 10% of the entire mining network. While not cheap, it appears it could be relatively affordable to hit the 50% mark.
The difficulty auto-adjusts to keep supply constant, so market forces will drive the cost of mining a bitcoin to just below the value of a bitcoin. Hence, ASICs should have no major long-term effect on the difficulty of taking over the network.
Arguably they'll even make it a bit harder since you'd need specialized hardware, and couldn't just buy up a bunch of GPU EC2 instances on a whim and take over the bitcoin network.
As of today, you can't buy the ASICs because they don't exist, so I guess it's true (plus minus a lot of millions as it is a rough estimate). But it will be a lot cheaper very soon for a brief period of time before miners up the arms race [-- see reply below for clarification --]
This is incorrect. Avalon has shipped a number of Asic units over the past 2-3 weeks, as confirmed by several trusted members of the Bitcoin community, including by one of the core developers of the Bitcon client.
My understanding (and it could be wrong) is that ASICMiner has made their own asics for their own use. They are not selling them.
Their business model is that they have a sub-pool in a larger mining pool, and they financed their operation by selling shares (denominated in Bitcoin). The shares pay out weekly dividends, based on the awards the mining pool provides to them.
There's a couple (very, very tiny) bitcoin denominated stock exchanges, and on some of them you can buy ASICMiner "passthroughs". These represent shares in ASICMiner that are controlled in large blocks by an individual. That individual passes through the dividends to the passthrough shareholders, minus a small management fee.
It's an elaborate system, but I guess money finds a way, even when it's bitcoin.
That is correct. ASICMINER went through an informal IPO, selling at the time the equivalent of $200k or so of shares. With these funds plus the project's creators own investments, they designed a custom 130nm ASIC. http://bitcoinmagazine.com/asicminer-starts-hashing/
Sorry, I didn't express myself properly. AsicMiner does have ASICs, and Avalon also has shown them to work. All I'm saying is that a government or individual can't buy today enough ASICs off-the-shelf to conduct the 50% attack.
The article talks about crowd processing eliminating falsified balances, but it doesn't go into some exchanges getting hacked. Hasn't that been a security concern? Haven't people lost money that way? I know my credit card numbers can get stolen, is that the right comparison? Do exchanges just make you whole?
It highlights many of the risks and benefits but misses out on what I think is the Achilles' heel of bitcoin, the ability for sovereigns to shut down the exchanges.
Without the ability to convert bitcoin to fiat easily, the currency loses its usefulness as a proxy for fiat currency. Does anyone have a convincing argument for how the exchanges are not vulnerable?