This article is surprisingly good. I like the mention of Bitcoin being the "world's first" decentralized digital currency. This is a clear way to convey to the reader that Bitcoin very is different from everything else. I use the same term in my Bitcoin introduction: http://blog.zorinaq.com/?e=66
The Economist author only made one error: Bitcoin is not the only decentralized digital currency. Some forks exist (eg. Litecoin), although they have very little traction currently; see the Alternate Cryptocurrency board: https://bitcointalk.org/index.php?board=67.0
Totally off-topic: the "vertical-align: top" on line 30 of your style.css is making any non-bold, non-italic body text get vertically misaligned with neighboring bold or italic text in Chrome 22 on Windows 7.
Correct me if I'm wrong, but aren't the others just merely clones of the original Bitcoin, using the same source code? I would still call Bitcoin the only currency, as it's the base they all share.
It seems like there's a bit of terminology overloading going on. You have Bitcoin the protocol, which all these things are using, but you also have Bitcoin the currency, which is separate from the other currencies using the same protocol but with different networks (or block chains or whatever you call it).
Maybe the eBay bit: Silk Road uses zero auction features (if you don't like a price, you can either not buy, or buy from a different vendor), you can't sign up to be a seller for free, and the general experience is more comparable to using Amazon's third-party sellers.
Are there any digital wallets out there that are:
1) trustworthy
2) allow easy funding (credit/debit card/paypal)
Until that's the case I just don't see this taking off. I'm fairly motivated to buy a few just for kicks, but if I can't be bothered I can't see the general population doing so either.
I see MtGox takes wire transfers, which is almost good enough, but not quite. Didn't they also suffer a breach not so long ago?
I used MtGox a month ago to see what all the bitcoin fuss was about and their UK bank was inexplicably closed during that time. They've been contacting me with updates via email but the money I wired to them still hasn't been credited to my MtGox account which is really frustrating.
I run StrongCoin e-wallet https://www.strongcoin.com, which has been operating for over a year now very successfully. I also run https://www.bitcoinary.com a peer to peer site for exchanging coins in numerous ways including Paypal.
https://www.bitinstant.com is trustworthy and by far has the most funding options, which are the most convenient. Eg. in the US you can deposit cash in a CVS, 7-11, Walmart, and have it credited to your wallet almost instantly.
I've used bitcoinary with success to get bitcoins for Paypal USD (with some paying attention to who is selling them: browsing their history + feedback)
Blockchain.info's wallet is very good. They store your private key encrypted and can only be decrypted client side with your password. They'll also email your encrypted wallet backups when your wallet changes.
Estimated around the end of this year, the amount of Bitcoins mined per day will be halved according to the schedule. If I understand correctly, this means that the cost to mine one BTC will double, roughly, if the mining capacity stays the same.
On top of that, some companies are, or should be coming out with ASIC-based miners, which are orders of magnitude faster in mining bitcoins than GPU's, at lower energy usage, increasing the total mining capacity.
Does the cost of mining one BTC influence the price of one BTC? Will these events have an influence on the price (if they were not already 'priced in')?
Does the cost of mining one BTC influence the price of one BTC?
The opposite. The exchange rate determines how many miners are willing to compete for each mined BTC. You can see that difficulty tends to increase a few weeks after the price increases (keep in mind that difficulty has a 1-2 week delay built in). http://bitcoinx.com/charts/chart_large_lin.png
The cost of mining doesn't influence bitcoin price because creating 50 new bitcoins isn't going to inflate the currency by any perceptible amount. The price does influence those who might want to get into mining. Eventually no new bitcoins will be generated from mining. At that point miners will make money solely from transaction fees when creating a new blocks.
50 new bitcoins are created every 10 minutes. That is 7200 coins a day, 2,628,000 a year. This is massive inflation and definitely has an effect on the currency.
No one really knows what the effect of halving of the mining rate will be. The currency is young, relatively small, and not yet very efficient (from an economic standpoint). However, everyone know the event is coming up. It MAY already be priced in. We won't know until the end of the year.
There is a small risk that this event may cause some disruptions in the underlying bitcoin infrastructure, however. The idea goes: Many miners are running full tilt going into the event. They know their income will half, so they want to get as much out of mining as possible before it happens. This drives up the difficultly level.
After the event, many of these miners may drop off as their operations suddenly become unprofitable (they are already running at the edge of profitability by revenue per kilowatt hour metrics).
When the miners drop off, it takes some time for the difficulty level to adjust (up to two weeks). Before the difficulty level adjusts, the time between transaction blocks will begin to grow, which slows down the confirmation of transactions. In a worse case scenario where a large volume of processing power leaves the network simultaneously, it could take several hours to get transactions confirmed for a few weeks after the event.
Will this actually matter, or even be noticed? Again, noone knows for sure.
People will be replacing FPGAs with much faster ASICs around the same time that GPUs drop out due to the reward halving. Many people are predicting that difficulty will increase 10x, not decrease.
I can answer the last question easily. If the supply of bitcoins rises faster than the real value of the debt they represent, their nominal value against the dollar will fall.
This argument does not hold regarding the rate against the dollar. The rate of supply of dollars is massively outstripping the real value of the debt they represent.
I'd qualify as =might= be factored into the price of BTC. Whether it is or not depends in (large?) part on the financial acumen of people trading BTC and if they're bothering to model what the effects will be.
If you assume efficient markets, yes. Nothing I've seen about the bitcoin market so far would suggest that it's efficient. Low volume, ideological motivations, unstable technologies, few really safe places to store value, etc.
All money represents debt that your future payee "owes" you. It's basically a promise of future payment by "society" to you. There was a great Planet Money podcast about the origins of money, which was essentially promissory notes of farmers to provide X units of food in the future, but I'm having a hell of a time finding it now.
> Does the cost of mining one BTC influence the price of one BTC?
Difficulty adjusts automatically to match the total mining hashrate - there's always about 10 new blocks mined per hour. So if more people are mining the difficulty goes up.
What is changing is the number of BTC you get for mining a block. Currently each block is worth 50 BTC. Soon that will drop to 25 BTC. If many people find mining to be uneconomical they will stop their mining rigs and the difficulty will drop.
What will change is the influx of new bitcoin into the market, instead of 500BTC per hour we'll get 250BTC per hour (on average). Will that make the price go up? Maybe.
On that note, has anybody heard of any data or estimates on the proportion of Bitcoin mining that is done with botnets? I keep hearing that a large proportion of the bigger botnets in the world are being used to mine Bitcoins (when they aren't being leased out for DDoS attacks).
That changes the economic dynamic since the cost of mining is the cost of assembling the botnet (ie. cheap)
While I'm rather ambivalent on Bitcoin, I feel the article should show transaction volume as well, in order for us to make an informed assessment on its future prospects.
Displaying a prominent graph of only BTC/USD as proof of recovery, without an indication of liquidity and total market size, seems insufficient.
But a better indicator that Bitcoin activity is
increasing is the number of Bitcoin transactions on
the network itself.
Is it better? If I shifted some bitcoins around between several of my own wallets, that would create transactions - but would that be a meaningful indication of anything?
I wonder how we determine similar 'activity' as far as normal currencies are concerned? Either it's related to money moving between accounts, or it's related to volumes on an exchange, or something else entirely, but surely those values are just as easily 'manipulated' as any of BTC.
Admittedly the volumes themselves dictate that me moving money from one bank account to another will have less of an impact on the final number, but then again, I'm always moving money between my different bank accounts, but never between my Bitcoin addresses. I obviously can't say for certain, but I don't think there's a meaningful number of users moving BTC between their wallets for no reason.
If someone day-trades (constantly buys and sells) on a Bitcoin exchange, it would also increase volume without meaning that Bitcoin activity is increasing. In fact, there are plenty of day-traders on Bitcoin exchanges.
So, yes, I argue that Bitcoin transactions are a better indicator of its activity. Sure, someone could be moving coins between his wallets, but every other indicator also confirms that Bitcoin activity is increasing (nr. of merchants, Bitpay users, etc).
Thinking further on the risks of bitcoin, is it possible that the algorithm itself could have a yet to be discovered weakness, allowing for a trivial solution?
(similar to what happens when some forms of crypto are broken)
Crypto weaknesses are always possible. But generally speaking, widely-deployed algorithm like SHA256, ECDSA, etc (which Bitcoin relies on) don't become suddenly broken one morning. Research progresses slowly (eg. weaknesses found in round-reduced versions of these algorithms), giving ample time to the community to plan for an upgrade when it becomes clear that one of the algorithms will become broken soon.
Bitcoin is also very conservatively designed. For example an address is computed with a RIPEMD160 hash of a SHA256 hash, so one would have to break both algorithms with preimage attacks to be able to generate address collisions allowing an attacker to steal coins from an arbitrary address. Because of this conservative design, it gives it a "safety cushion" to have time to anticipate cryptographic changes if they are needed.
I was thinking the same (but my understanding of all of this is still very weak).
Isn't it a major assumption that these "extremely difficult mathematical problems" remain extremely difficult until all Bitcoins have been mined, 120-130 years in the future?
This statement from the last paragraph of the very interesting Economist magazine article submitted here is thought-provoking: "Bitcoins tend not to be very secure, says Richard Booth, a consultant at RSA, a cyber-security firm. As some users have found to their cost, hackers can sometimes steal Bitcoins from users’ online vaults. In the latest raid, on September 5th, hackers stole $250,000 in Bitcoins from Bitfloor, a large American exchange, causing it to shut down its operation." What does that say about the overall reliability of Bitcoins as a new form of currency?
A while ago I wrote that perhaps the greatest contribution the Bitcoin experiment will make to humankind is to teach you and me and our neighbors more about the realities of economics. And later I added that the Bitcoin experiment will also contribute to greater understanding of attack surfaces and online crime. Many of the ideas about how to mine Bitcoins, store Bitcoins, and trade with Bitcoins as a medium of exchange illustrate both the strengths and weaknesses of any other medium of exchange in a world full of human beings. Seeing the discussion of Bitcoins here on Hacker News reminds me of early online discussions in the 1990s of online payment systems such as PayPal, and the arguments beforehand that PayPal wouldn't have to invest a lot of time and effort (as it eventually did) building defenses against theft and fraud. If a weakness in a system is attached to a lot of money, the way to bet is to bet that someone will go looking for that weakness, even if you haven't thought of it.
AFTER EDIT: Here is a question for all the security-knowledgeable persons who participate here on Hacker News, a question once asked of the inventor of Pretty Good Privacy (PGP). How expensive do you think it would be for the United States National Security Agency (or a comparable organization from another national government) to crack a Bitcoin store, given that we know that some Bitcoin caches have already been cracked?
I'm a bitcoin skeptic, but I think that the problems you mentioned in the first paragraph have more to do with the poor security of those individual vaults, not bitcoin itself.
But, it does bring up an interesting point: as far as how I know bitcoin works, once you steal the bitcoins, they're stolen. This is different from fiat currency, at least online. If someone breaks into Bank of America's servers and figures out how to transfer a bunch of money from other people's accounts to their own, Bank of American can undo it. As long as the person does not get a cash or cash-equivalent, then it should be reversible. I invite the people who are more familiar with bitcoin to correct me, but as I understand the protocol behind it, once a bitcoin transfer from one owner to another owner is made, it's not (forcibly) reversible.
"In the latest raid, on September 5th, hackers stole $250,000 in Bitcoins from Bitfloor, a large American exchange, causing it to shut down its operation." What does that say about the overall reliability of Bitcoins as a new form of currency?
This can be compared to the Wild West times when gangs used to break into banks and steal lots of money. It is not that the money was weak, but that those storing it did not have proper security.
As bitcoins become more popular, you will have more and more competent individuals start offering storage services for these.
Banks of today have not become competent, only they have fixed those flaws that occurred the most in the past. Given that banks continue to have high visibility breaches and leaks of user information, it is not clear that time is what's needed for Bitcoin to get better security.
> What does that say about the overall reliability of Bitcoins as a new form of currency?
As xtracto said, there isn't something inherently insecure about Bitcoin. In fact, AFAIK, there have been no exploits to date of the actual math behind Bitcoin itself. The issue has been breaches of the various institutions dealing in Bitcoins.
When a bank is broken into and cash is stolen, you wouldn't say that dollars are insecure ― you'd say that the bank is insecure. This is no different, except it's occurring online.
All this tells me is that it hasn't hit governmental radar in any meaningful way, yet.
If Bitcoin was annoying enough to be a real problem to the powers that be, the computing resources necessary to hijack the main chain would be online in a very short period of time.
Bitcoin uses a proof of work system to determine the valid chain of transactions, as well as who gets the initial currency payout. Basically, anyone who controls more than 50% of the CPU power in the Bitcoin network can control the currency.
Some of the various forks have tried to work around this by having trusted nodes and the like, but then you lose the decentralized nature and are vulnerable to the trusted nodes going rogue.
Having greater than 50% of the computing power of the bitcoin network does not give you complete power of the currency. You would be able to prevent transactions from occurring, and double-spend recent transactions. Since the bitcoin network is effectively the most powerful supercomputer in the world, this would be a very expensive attack to sustain, without that large of a payoff.
I definitely wouldn't recommend starting with the official bitcoin paper to learn about this. That would be like telling a math student to just read Newton if they want to learn Calculus :). Check out the bitcoin wiki instead: https://en.bitcoin.it/wiki/Weaknesses#Attacker_has_a_lot_of_...
FWIW, when I was figuring out Bitcoin I avoided the paper at first (due to logic similar to yours) but eventually read it because there was a lot of noise in the wiki, etc (may have changed). I found it surprisingly accessible; more so than the vast majority of academic papers I've come across (I have most of an EE degree but little to no formal CS training).
I've seen some time ago comparison of popularity of passing fads and useful inventions.
Useful invention graph looked like this http://3.bp.blogspot.com/_2IAWpIwOiOc/SuvD_xiOfTI/AAAAAAAAAD... (example was IT) and graph for the passing fad dropped down to almost zero after initial peak (example was Macarena dance, if you remember that you are old ;-) ).
No one here has attended a wedding or bar mitzvah? That silly dance is far from dead, which is not all that important depending on your social set but does give some doubt on the analysis.
It always seemed odd to me that the protocol expects everyone to always have the full blockchain (currently over 2 GB of data). The response is always "when it becomes a problem we'll fix it", but isn't it already a problem for mobile use etc? The other problem I never see mentioned is the slow speed in getting transactions confirmed (for me it takes maybe 30 minutes)
The original bitcoin paper described a "Simplified Payment Verification" which doesn't require the full blockchain. This is used by Bitcoin Wallet for android.
Another alternative is to work with a trusted server. This is used by Bitcoin Spinner for Android.
As I understand you can download the latest Bitcoin transactions (the latest spend of each coin) and discard the rest of the transaction history. The problem being though is that you have to trust some third party to give you the correct information.
Try Electrum http://electrum-desktop.com/ It uses servers with the full blockchain to verification and has another neat feature - deterministic wallets. You keep the seed around and you can recover the wallet.
With the original client, having the appropriate port open/forwarded helps as well, and my understanding is that part of the slow sync is the inefficient database design. It's on the list of #TODOs.
You don't need the block chain to use bitcoin. Honestly, there only needs to be one complete copy in total across the entire world. It's just a quirk of the current client that it insists on such a download in an era where it takes more than a few minutes.
agreed. If people can't understand bitcoin in at least some basic way they will not use it. Faq's on bitcoin are full of questions which nobody has ever asked. Even bitcoin enthusiasts can't explain bitcoin, because they don't understand it either. constantly changing account/addresses? i don't like it, would rather one account and have payments tagged with sender info. i still have not managed to find out if bitcoin supports this. can't make sense of messaging aspect, labeling, verification , don't understand mining. Ordinary folks will not buy into this if it can not be explained in two short sentences, no matter how logical or robust or brilliant bitcoin might be.
I'm curious if this last sentence from the article is true:
> It turns out that a currency can thrive even when
> no one is making laws for it.
Were the laws were so weak or hard to enforce at one time that, effectively, there were no laws? Can we look at currencies from that period of time to give us a sense of how bitcoin will work?
(Another poster pointed out that the spec is making the rules, and not a government, so it may be wrong to compare bitcoin with an un/poorly regulated currency.)
It shows bit-coin value to USD, gold, silver, and crude. It's fairly stable against the commodities but drastically spikes in value against the dollar. Our Fed at work...
So bitcoins are a commodity like unto gold, silver and crude, and behave the same way. Cool, but for practical use as a currency it would be better if they were correlated with USD/EUR/JPY, food, labour and other things that normal people want to buy.
Surprising? What's surprising about it? That there's no governmental issuing authority? (It is not true that there is "no issuing authority whatsoever", the issuing authority is the spec.)
This won't surprise anyone with a grasp of economic history that extends beyond 1945.
No, it is not surprising. If you look at the Bitcoin forums to understand the community, it is very clear that a large core of users are really here for the long term. They truly believe in the concept of a decentralized digital currency.
This core group of users (including me) were completely unswayed by the crash of 2011 H2. We all knew that it was nothing more than a correction of the insane valuation bubble of earlier that year.
This is not an asset, it is a currency. Asset values do well when investors buy-and-hold for the long-term. Currencies die that way - they increase in value when they are traded, i.e. transacted in. A high ratio of buy-and-hold investors to merchant activity in a currency is a sign of speculative build-up. Paradoxically, the attractiveness of a currency to trade in is partially a function of how well it holds value - price level volatility bodes badly for that cause.
As a financial market participant, yes, it is surprising that the Bitcoin market stabilised as quickly as it did since it suggests there was an influx of asset and currency owners willing to trade their value into Bitcoins despite the risk of rapidly losing that value in price level volatility. Having examined some AML flow of funds data I can say this resilience has nothing to do with the zeal of the Bitcoin community, which in terms of wealth is insignificantly tiny, and everything to do with its adoption in trades deemed, shall we say, unsavoury by the world's legal authorities. As a supporter of further sophistication in the Bitcoin markets myself, this sentiment of seeing nothing awry in the Bitcoin/USD volatility is worrying, though the faith expressed in the currency is reassuring (these are, after all, fiat currencies).
And I think this is the actual problem with Bitcoin, the community is small and few people are owners of Bitcoins.
How many people sold all their Bitcoins in the wake of that crash? It is indeed unsurprising to me too that Bitcoin recovered after a crash, because the owners of Bitcoin believe in it and so the market kept being rational post-crash.
The real test of Bitcoin has yet to happen, when Bitcoin will be popular enough that regular folks will put their life savings into it.
That test is unlikely to happen. regular folks do not put their life savings into currencies or commodities. They put their life savings into banks. A saving account in a bank does not contain a heap of cash or gold nuggets. Instead its an trust relationship between the bank and the bank customer, like a contract.
I would guess that in "the future", the name of currencies will become less likely to end up in the mindset of the general public. All they know is that a specific thing (a computer, a car, and so on) will cost X amount of what they have in the bank.
That test is unlikely to happen. regular folks do not put their life savings into currencies or commodities. They put their life savings into banks.
And your savings account is denominated in ...?
(hint1: USD EUR JPY CAD AUD)
(hint2: The 'stuff' people exchange for goods and services)
(hint3: Currencies!)
You're right about commodities, though. Unless you are deliberately taking positions in commodity ETFs or futures contracts (something that a person is unlikely to be doing without knowing it), you are probably not taking on commodity exposure.
Regular folks do not put their life savings into currencies or commodities. They put their life savings into banks.
What planet are you from? On my planet, regular folks put their savings into 401k accounts, which invest in stocks, bonds, and sometimes commodities. If you do choose to keep your savings in a bank, you're investing in a currency (probably the US dollar.)
Instead its an trust relationship between the bank and the bank customer, like a contract.
Nope. Actually it's a trust relationship between the federal government, which insures your account up to 250,000 through FDIC, and you. You also have to trust that the federal government isn't going to inflate away your savings (spoiler alert: they are going to.)
I guess it depend on where on the planet one is (or maybe its on the not-the-usa-planet). I am not sure banks where I live even have anything called 401k plans. They do have stocks and bonds, but its not something commonly talked about. pension savings are done mostly through tax but in rare occasions, addition money is added by the place you work from. In very extreme rare occasions, people sometimes add personal savings to it.
I've been trying to buy bitcoins for a month via dwolla and mtgox, still haven't been able to make the purchase. I don't know how "regular folks" are supposed to get in.
I'm wondering, what if government bans using of Bitcoin (or such a digital currency in general)? Not that it will try to block it just that it will be illegal to use it (i.e. for companies to receive Bitcoins). Will it still be used and/or useful?
You can make bitcoin a legitimate payment mechanism. Despite the advertised anonymity it's fairly easy to track down a particular identity from the web of transactions. If people moved on mass into bitcoin it would make their financial history pretty impossible to hide.
The government banned sharing copyrighted materials and everyone stopped sharing movies, right? (It was similar to the experience they had with banning drugs. Now nobody uses drugs, I hear.)
One thing that I wonder about with respect to Bitcoin is the issue of taxation. Governments can support the dominance of their currency within their borders by compelling their citizens to pay taxes in the national currency.
And on the "supply-side", in the US at least, federal spending is a significant portion of GDP, so you will always have a large amount of dollars floating around being used as a means of credit, exchange, and value.
And finally, having control of interest rates is a powerful tool to support economic growth - I would be fearful of a world where policymakers did not have that ability.
The Economist author only made one error: Bitcoin is not the only decentralized digital currency. Some forks exist (eg. Litecoin), although they have very little traction currently; see the Alternate Cryptocurrency board: https://bitcointalk.org/index.php?board=67.0