Bitcoin Box

Ads

Bitcoin, Alternative Applications and the Profit Motive

author: Vitalik Buterin
published: 2011-07-23 03:28:23 UTC

Previously, I talked about the key factor that makes Bitcoin such a stable system - it is based on a set of rules that is not profitable for anyone to deviate from. Everyone has an incentive to play fair and to reject attempts at cheating. However, there is another concept that has made Bitcoin as successful as it is: the profit motive.

The idea is simple: all distributed computing networks previously have relied on either generosity or default settings to keep them up. Freenet, for example, relies on people being willing to donate storage space for free in order to function, and Tor relies on people running exit nodes. However, of all these projects only Tor has seen any significant success, and even it is fairly obscure. There is simply not enough interest for Freenet to develop and hit the critical mass needed to become used by the general public, so it remains obscure and unable to expand beyond the small group of users that need it to avoid censorship. It has seen use in China but in the English media it is virtually unheard of. Freenet could potentially have massive utility for the public if it would grow mainstream - imagine the possibilities that fully decentralized and distributed web hosting would offer. Bitcoin, however, has attracted massive mainstream attention and the value of all the Bitcoins available has grown into the tens of millions of dollars all within the span of a few months. What is the difference? The profit motive. The Bitcoin network relies on miners to keep itself up, but the miners are not mining out of goodwill; they're doing it because it earns them money. As a result, the Bitcoin network currently weighs in at 148 petaflops, 20 times more powerful than the much more established [email protected] and almost 100 times more powerful than any single supercomputing site on the planet. It is only by integrating profit into the task of Bitcoin mining that we have so many dedicated professional miners building GPU and FPGA systems specifically for the purpose of maintaining the Bitcoin network as efficiently as possible. With the example of Bitcoin, we have shown how to create a stable decentralized network for any purpose: have a profit motive built into the task of maintaining it.

Thus, the idea of a built-in profit motive is in fact far more versatile than it first seems. The one alternative application that already exists is Namecoin, a decentralized domain name system that uses a bitcoin-like currency system to limit the supply of domain names to avoid spammers. The key insight that allows Namecoin and potentially many other applications to come is that we can put anything we want into transactions. Namecoin puts in name/value pairs on special coins to store registered domain names. A DNS is only one application; it would also possible to store profiles of individuals or businesses and comments on them, ie. a reputation system on a decentralized blockchain rather than the centralized web of trust we have now. Creating fully decentralized email is also possible, and very convenient - the nature of the Bitcoin protocol would allow every message to be public key encrypted to the recipient automatically - but there are of course limits: hosting entire documents, songs and videos on the blockchain is obviously ridiculously impractical, and for such applications a completely different approach is necessary as only a few computers need to store someone's files for the files to be accessible. Thus, straying from Bitcoin and creating a special-purpose alternative currency is not always the best option - one of the strengths of Bitcoin as it already is is microtransactions, and this can itself be used to introduce the profit motive into everyday distributed computing.

First, let us consider the idea of a decentralized truly cloud-based Dropbox-like service. Wuala has created something similar: a system where people can, instead of buying space, get it for free by allowing Wuala to use their computer as a server for other users. Combining this with a system where people can get free accounts in this way and rent them out on the open market to earn a profit from unused disk space would complete the cycle, creating a distributed cloud file storage system that might eventually take up nearly every internet-connected hard drive on the planet, all thanks to the advent of a profit motive. If it turns out that there is more space than people need, the space will still be put to use as that portion of the population for whom it is worth it will store their files multiple times or with advanced redundancy algorithms, such as those implemented by the online fully decentralized file storage cloud system Tahoe-LAFS. However, one weakness would still remain: Wuala is centralized. But there is nothing stopping a decentralized market-based system from appearing to serve the same purpose.

A simple algorithm would be to create the system out of independent one-on-one contracts, where the file hoster hosts the file from the file owner in exchange for a fee per hour in digital currency, and OpenTransactions may be part of the solution. OpenTransactions intends to create a software library through which it would be possible to run services in exchange for digital tokens which would be centralized but have the necessary improvements of instant transaction verification and no transaction fees (and also increased anonymity with blind signatures, which I described here); these currencies could be issued and backed by competing banks holding a reserve of bitcoin. Looking at the business uses listed, it becomes clear that with OpenTransactions any internet service can be easily done for profit - running a Tor node, running a server and running a mesh network node (essentially a mini-ISP that sells access to only part of the route that a packet would use to travel to its destination, so each packet would pay for dozens or more of such nodes) can become possible. Instead of a small number of businesses operating such services, the types of service can be standardized so that every computer can be running a competing server or node business with no special configuration, and customers can switch from one provider to another (or from one hundred providers in parallel to another hundred providers) just as effortlessly.

Unlike the Bitcoin mining market, which is dominated by specialized hardware that is thousands of times better at performing the specialized task of computing hashes than standard CPUs, tasks like storing files and sending and receiving things over the internet are what standard computers are designed for, so the issue of electricity cost crowding casual users out of the market will in many cases not even arise, or find a reasonable equilibrium (eg. turning on 10 year old consumer laptops for their hard drives will not be worth it, but turning on 3 year old laptops will be). This is the true power of the profit motive in distributed computing: it could potentially create an optimally efficient market that prevents needless waste and puts every byte of hard drive space, every joule of electricity and every second of wireless antenna usage to the best use possible in a system with no single point of failure. And on top of all that with public key cryptography we get anonymous web hosting, email and file storage as a bonus.

Random Articles

blog comments powered by Disqus