PayPal says ... 12/12.  Their estimates have been known to be off in the past.  We got a surge of popularity for PayPal and ACH is painfully slow.

full nodes need a full copy of the blockchain back to the genesis block.  The satoshi (reference) client is a full node (zero trust).

As bitcoin continues to grow more users likely will use "lite-nodes" or eWallets. 

To accept bitcoins you only need a single address however
1) using a unique address per user and/or order makes checking/tracking payments much easier.
2) you don't need to use a third party service however some merchants may find it easier to use a third party service.

Bitstamp = 3
FastCash4Bitcoins = 3
BTC-e = 3
CampBX = 4
Bitfloor = 6
Intersango = 6
MtGox = 6
Virwox = 6

bitcoin.de = ?
BTC24 = ?
Virtex = ?
BtcChina = ?
Bitcurex = ?

In theory fractional reserve banking is possible however there are some unique constraints.

1) There is no central bank.  Central banks also engage in FRB except they create the monetary base out of thin air.  With Bitcoin the monetary base is fixed at ~21M BTC.

2) There is unlikely to be anything like FDIC.  FDIC distorts the market in that the cost of insuring funds is paid for by taxpayers not depositors.  While it is possible that in the future Bitcoin banks will exist and will even be insured it almost certainly will be private insurance.  This makes using a Bitcoin bank more expensive than simply holding the coins yourself.  Some people will still use banks but a smaller % than in most fiat economies.

3) The reserve ratio will likely be higher.   Private insurance companies will be profit driven (not public policy driven) and smaller reserves means more risk.

4) As long as BTC is appreciating in value relative to other currencies demand for loans will be modest.  Say you need to buy a car.  You could get a $10,000 loan in USD (which are dropping in value making repayment easier) or you could get a 100 BTC (assume $100:1 exchange rate) loan (which are rising in value making repayment harder).  Some people will choose to borrow in BTC but it will limit demand.

5) Less need for a bank.   Without a bank (or bank like entity) it is very difficulty and costly to engage in commerce in USD.  Imagine no PayPal, no credit cards, no prepaid payroll cards, no checking account, no direct deposit.  Even routine things like cashing your paycheck to buy a game on steam becomes next to impossible.  The utility banks bring drives deposits into the bank.

The effective money supply is Monetary Base * Money Mulitiplier.   

In most FRB systems the central bank can influence both the monetary base AND the Money Multiplier.
In Bitcoin the monetary base is fixed (once all coins are minted) and the Money Multiplier is unlikely to be very large.

Imagine a hypothetical scenario where 30% of all Bitcoins are in Fractional Reserve Banks.  The insurance company requires a 50% reserve.   That means the money multiplier is 1 + 0.3*0.5 = 1.15.   The money multiplier with no Fractional Reserve Banking at all would be 1. 

Traditionally the money multiplier in the US has been roughly 2.5 to 3.  This has crashed recently due to the recession (well due to Fed's policies) however a money multiplier to 2, 2.5 or 3 seems highly improbable in Bitcoin economy given the increased utility of the currency and the reduced utility of a bank.

Yeah OTC doesn't do much good if you don't validate ratings.  No that doesn't just mean look up a rating either.  You need to strongly authenticate (via ident command) that the person you are speaking with has that rating.  Anyone can "say" they are anyone else (usually someone with a high rating).  Unless you ident the user and validate the rating, OTC provides no protection.

Good idea. Added to the todo.

However 6 confirms isn't magically safe.  It isn't like 5 confirms = massive risk and then 6 confirms = impossible.

Lets assume the attacker has hashpower that equals 20% of the network. 

To reverse an unconfirmed transaction will be 100% successful if using a finney attack or 20% of the time by brute force.  Obviously too much of a risk for high value transactions.
To reverse 1 confirmation will be successful (0.2^2 ) 4% of the time.  An attacker could reverse roughly 1 in 25 deposits.  That likely is insufficient.
To reverse 2 confirmations will happen (0.2^3 ) 0.8% of the time.  An attacker could reverse roughly 1 in 125 deposits.  Pretty small attack vector but still plausible.
To reverse 3 confirmations will happen (0.2^4 ) 0.16% of the time.  An attacker could reverse roughly 1 in 625 deposits.  The attack is non-viable and very obvious*
To reverse 4 confirmations will happen (0.2^5 ) 0.032% of the time.  An attacker could reverse roughly 1 in 3125 deposits.  The attack is completely non-viable.

* With a 0.16% success rate the attacker would only reverse on average one in 625 deposits.  Given there are only 144 blocks per day the attacker would need to deposit a MASSIVE amount of funds every hour (24+ times per day) for an average of 4-5 days before being successful.   The signature would be very obvious.   The attacker will on average lose 625 blocks to orphans for every successful attack.  The lost blocks would be worth roughly $203,000.  So to yield a 30% bonus on that would require a $300,000 double spend.  Think it might be obvious someone with a level 3 verified account depositing and withdrawing $300K in BTC every hour for days and days?

MtGox 6 confirm policy is simply an anachronism.  Why 6?  Why not 60 to be super duper sure.  Satoshi never intended the #6 to have divine like powers.

In MtGox defense one would think that BANK ACCOUNTS would have a checksum but they don't.  Who knows how many countless hours are lost resolving banking issues due to the fact that it is completely impossible to validate if an account number is valid (at least in the US).

All flood insurance providers (which are essentially just payment processors for federal flood insurance program) accept payment through escrow.  They can bill your mortgage company and the mortgage company will simply increase your escrow amount.  Similar to home insurance you don't need to use the mortgage company providers and you don't need to pay out of pocket.  That is why escrow accounts on mortgages exist.

Yeah I think I might know a little something about how Bitcoin works.  Thanks for asking.


For me personally?  No.  However "I" an not the entire planet worth of potential users.  Did you read the post?  Many people will SIMPLY NOT WANT TO BE RESPONSIBLE FOR THEIR MONEY!  The risk of uninsured loss, theft, or robbery is simply too much of a risk. 

I mean lets pretend Warren Buffet had his entire network in Bitcoins.  $40 billion in completely irreversible, untracable funds.  Think that might present a security risk to him?  Think maybe he would want to NOT have direct access to his own funds? 

Many people simply aren't tech savy enough to run their own wallet (securely both against loss and theft).  Do you think your grandmother is likely to install the blockchain, pick out a 12 digit passphrase, keep routine offline backups, ensure the system is free from malware, etc?

Here is one example:
http://www.npr.org/blogs/money/2012/08/03/158047349/how-americas-biggest-bank-makes-money

Note the recession and near 0% rate banks can borrow at has reduced the share that fees make up.  In this article Chase "only" generated 35% of revenue from fees.

At least in the US banks make more money from fees then they do from interest.   Bitcoin is about freedom and freedom means freedom of choice.  If Bitcoin became very popular (say PayPal sized in volume) most users would WANT to put their money in a "bank".  They don't want the responsibility that comes with the freedom Bitcoin offers.  Sure Bitcoin gives you the freedom to bypass the banks but it also gives you the "freedom" to lose your entire life savings as easily as you lost those vacation photos from last year.

No reason a bank couldn't be very very very profitable making fees on overdraft "protection", off blockchain transactions, and currency exchanges.

Yeah feeling much better.  Site should be complete "soon".

Exactly.  The decline in purcashing power of the dollar is pushing the longer multi-year trend however anyone saying silver is "stable" just uneducated.   The daily and weekly moves aren't suddenly caused by a major change in purcashing power of the dollar.

I mean if it was then one would expect McDonald's combo meal to be $4 this week, $5 last week, and $2.25 at the start of the year.  Sometimes during the day the price of a meal will jump $0.50 while you are standing in line.

The United States Note was the last national currency of the US.  I was incorrect in saying it was never unbacked fiat.  Due to the cost of the civil war (and the removal of the Congressional seats of the south where "hard money" was more popular) the US federal government did print without backing of precious metal from 1861-1874.  Part of the reason was that the civil war was very unpopular in the North (depsite the whitewashing attempts over the years).  If the civil war had to be financed by raising taxes it likely would have never been won.  Ironically it was the use of "taxation" via deception through inflation which is what I indicated a "print to spend" system does.  Much harder to explain the loss of purchasing power due to inflation to the masses then a direct tax.  The change was very unpopular and was repealed.

http://en.wikipedia.org/wiki/Resumption_Act


Even with the war time unbacked spending there are two things which separate it from an unbacked fiat currency.  The first is that after 1874 all US notes could still be redeemed for gold or silver even the ones issued between 1861 and 1874.   Due to the law of 1875 the government in the long run didn't issue unbacked currency it instead incurred a debt.  Iit needed to boost its silver/gold reserves of the coming decades so the reserves matched the number of notes minted.  Today US notes are no longer redeemable for gold or silver however they remain a debt of the US government, as the government needs to exchange them for Reserve Notes (which is must issue debt to acquire).  The treasury estimates there are roughly $230 million US notes (and slowly falling due to loss and redemption) in circulation and they contribute to the total indebtedness of the federal government.



If you look close you will notice something interesting about this note.  Today most people would call this a "twenty dollar bill" right?  Well not exactly.  Notice it says "will pay twenty dollars to bearer".  This note wasn't twenty dollars.  This note was an IOU.  A promise to pay the redeemer twenty dollars.  So what was a dollar?  371 grains of silver.  So you could redeem this note (which isn't dollars) for twenty dollars worth of silver.  Notice that dollar was a unit of measure.   Like saying a gallon of water, or an ounce of gold you could say twenty dollars of silver.  

Over time the language changed but a promise of redemption existed up to the very last US note.  After 1900 dollars were no longer redeemable for silver (only gold) so when the treasury would run short of gold and had an excess of silver it would issue silver certs. 

You do know the exact same thing was true of GPUs.  Improved efficiency only matters during the transition.   An ASIC may get 200x the MH per J but when difficulty goes up 200x the electrical cost per BTC will be roughly the same.

The transition period isn't going to be that long.  When these units ships they are going to ship in mass (dozens, hundreds of TH/s per week).  Why?  Once that dam breaks the retail value of new sales is going to plummet.  By shipping faster an ASIC producer can lock is millions more in profits.  That combined with difficulty adjusting every 2016 blocks and the type of magnitude increase expected there will be no window of "off the charts rewards".

Not every bitcoiner but certainly those who want to go to expos, conferences, and trade shows.  The sad thing is on top of being a horribly bad comment (almost black flag bad), it is completely untrue.  

Hi my name is X and I am advocating "Y".  Let me start by ensuring my target audience tunes out by throwing out a completely irresponsible and false statement.  My target audience won't know it is false and since I am the expert are likely to believe me.  Anyone potentially interested is now not interested and this is how I advocate for "Y".

The only good news is Bitcoin is so powerful that when thrown under the bus by it's own "supporters" the potential still shines through the horrible presentation.


Actually it was BTC in their eyes.   Nice article though.   Good to see more "mainstream" articles about bitcoin.    I also love the writing style of the author.

The US government never issued debt free fiat.   The US govt did issue gold back currency but that wouldn't be fiat.  During the period of "free gold" you could bring 1 oz of gold to US Bank and receive 1 oz worth of dollars or bring dollars worth 1oz of gold and receive 1 oz of gold.  The gold would then be shipped to a US repository and formed the "backing" for the notes issued.   Alternatively you could bring 1 oz of gold to the US mint and they would mint it into a new US coin (for a small seignorage fee). In the early history of the US government it didn't have much of a credit rating.  The gold backing was the only reason anyone (foreign govt or individual) would accept the notes. 

Countries which simply print as much money as they need end up in hyperinflation.  

I mean look at the US debt expansion (excess spending is covered by debt financing rather than pure printing).  In 1941 the US national debt was ~$50B.  Really think about that.  The combined surpluses and shortfalls of the US federal government for the first 165 years was $50 billion.  By 2000 that had reached $5,600B.  Still it took six decades.  Another way to look at it is the federal govt was spending beyond its means by "less than" $1,000B per decade.  The federal debt is now $15,500B.  An expansion of almost $10,000B in little over a decade.  The government overspent in the last year 20x as much as it did in the first 165 years COMBINED.    Without the "debt ceiling" the projected budgets (which likely are overly optimistic) through 2016 put the debt expansion to $20T (more than 120% of GDP).  The government is spending over a $1T a year more than it has.  A trillion dollars a year.

You honestly think it would be better if governments simply printed what they wanted when they wanted? A trillion, ten trillion, hundred trillion, a quadrillion dollars?   Even with the token constraint of the need for the government to balance receipts and revenue (and fund the difference with debt) governments (not just the US but all modern governments) have borrowed utterly insane amounts.  

They have shown an utterly reckless inability to "leave within their means" even with the ability to unilaterally confiscate any and all wealth they see fit.  Government aren't responsible.  Given them the ability to print unlimited money and they certainly will try to print an unlimited amount.  

The end result of every government which hasn't balanced receipts with revenue has been Hyperinflation.

Official proof that "scam accusations" have lost all value.

If you want to debate come armed with facts (or at least fake stats which could possibly be mistaken for correct).

Hint: interest doesn't make up 95% of anything.
Wealth has increased for the 99% over the last 40 years.