author: Vitalik Buterin
published: 2011-07-16 20:00:05 UTC
In the first part of this series, I talked about the use of Bitcoin in physical stores, and the conclusion was that with the introduction of smartphone applications that allow Bitcoin users to scan a QR code and immediately send to an address it can become incredibly easy to spend Bitcoin physically - the aforementioned QR code scanning is essentially a one step process. Bitcoin spending online has a parallel to this - Bitcoin payment processors like MyBitcoin exist for merchants to set up which generate addresses for users expressing their desire to buy a certain product to send to - one can see this process in action with BitcoinCashout's custom system which works in the same way. However, the problem is that these merchant services are not widely adopted, in the online world partly due to the Bitcoin community's "everybody's a merchant" decentralization and despecialization. In reality, payment with Bitcoin tends to work like this:
This is all done informally, so every step has a few hours of email communication lag. The use of payment processors combines steps 1,2 and 3, so it instead becomes the much simpler 2-step process: (1) B requests and sends coins, (2) A provides. There is a workarounds that is used now to shrink the number of steps from 4 to 2 - the transaction being the communication. B would send the payment, say, 0.2 BTC, to A, and then send a message saying claiming the transaction. This, of course, falls apart when A has more than one customer, and it gets confusing whose transaction belonged to whom. A solution was to require B to send a "marker amount" of bitcoins to A along with the actual payment - eg. 0.20031415, and then send a message to A immediately claiming that unique payment amount - essentially allowing a limited amount of communication with the transaction. This brings to light a minor flaw in the implementation of Bitcoin - earlier versions of the Bitcoin client allowed addresses to send messages to each other, but one could not send a message with a transaction. Because of this weakness, the feature was never used and is now removed. The marker amount strategy works, but it is highly inconvenient, so naturally merchants now tend to use payment processors.
One of the most significant innovations for Bitcoin businesses is services like BitPay. BitPay functions as a fully fledged merchant system for Bitcoin, and markets itself specifically as a substitute for credit cards and PayPal. It's most important innovation, however, is instantly converting BTC received from customers into conventional currencies, effectively insulating vendors from the chaos of the Bitcoin exchange rate, making Bitcoin attractive to them despite its current instability. The key insight is that such a service can just as easily be implemented for customers - a Bitcoin payment application which offers a single button to convert dollars in a MtGox account into Bitcoin and immediately send them to the desired address. Essentially, MtGox would function as a bank account using Bitcoin solely as an intermediate currency to work around the conventional payment system and offer Bitcoin's advantages like low transaction fees and anonymity. This may not work in the long term - MtGox and Tradehill have stated that they are willing to cooperate with authorities and this may eventually require them to demand proof of physical identity for accounts, but as I will explain later the point of this is as a short and medium-term strategy for Bitcoin adoption. With both the customer and the merchant almost entirely protected from the Bitcoin exchange rate (not fully protected due to the 10-minute window during the transaction, although potentially even this can be eliminated by either specialized insurance services or by direct instant credit transfer within or between payment processors and exchanges), this effectively splits up Bitcoin users into three categories:
Without automatic currency conversion, it's necessary for customers and merchants to be speculators at the same time, which discourages adoption. However, with automatic conversion, the three roles can be neatly separated, and Bitcoin becomes a pure conventional currency transfer system. If such a usage style becomes widely adopted, the value of Bitcoin will initially go down slightly, as people using Bitcoin as a currency will no longer need to actually hold any of it, but Bitcoin will not need to have a high value, or even a stable value, to be useful - Bitcoin must only have a sufficiently high value for it to be possible to buy the desired payment amount on the exchanges; keeping in mind that in such a system exchanges will have a lot more volume than they do now, for payments under $1000, even 1 USD per BTC is enough, and as for stability all the fluctuations are absorbed by speculators, so customers and merchants do not need to care about how stable or unstable the rate is. Bitcoin's clear advantages of low transaction fees and the option of anonymity it will be able to pick up adoption in the same way that credit cards, Paypal, Facebook and any other service that is only useful for interacting with other users of the same service gained adoption.
It is important to stress that this is not a viable long-term solution - governments may reach MtGox and Tradehill and regulate them in such a way that they become useless for this purpose, it's centralized, giving MtGox and Tradehill too much power, it reduces Bitcoin's value making it weak for large-scale transactions, it destabilizes Bitcoin's value since it would end up resting solely on speculators (not as big a deal as it seems: many speculators are in it for the long term) and it removes some of Bitcoin's long-term advantages such as deflationism. However, when there is a critical mass of Bitcoin users and the currency's value stabilizes, or the US dollar starts hyperinflating, or this solution becomes no longer viable, switching from it to using Bitcoin as a fully fledged currency will be effortless and not everyone has to do it all at once - a customer or merchant does not even need to know if his business partner is a speculator or an instant converter. Thus, the fact that automatic currency conversion is not where we want Bitcoin to go in the long term is not relevant since it does not drag us far away from what the correct long-term strategy is - it rather acts as a stepping stone on the way there.
In all of this, the key is technology - technology built on top of the Bitcoin system to simplify payment, and insulate payment from the exchange rate, that will set up Bitcoin as a competitor to credit cards and PayPal. Even in a limited role as an intermediary, Bitcoin is still incredibly powerful - it allows exchange accounts to be supplied anonymously, buying Bitcoin with a credit card or Paypal, sending it through one or two Bitcoin laundries for anonymity and sending it to the account. Bitcoin's advantage of deflationism can only be realized through stability and high volume, but for that to happen something must attract users to provide the high volume, and that can only be a strong payment system, which we are well on the way to realizing.
With immediate hopes of Bitcoin's value exploding past $100 shattered, the community's focus has now moved to improving Bitcoin's viabi...
One of the main arguments in favor of fiat currencies as opposed to currencies based on a scarce resource, whether gold, silver or bitcoins, is that time must be wasted mining something that has no...