A proposal: FixedToken, cloud computing model

During spikes and selloffs, the stock market offers tools like money market funds, treasury notes, and stable value funds--which are also seen as an attractive market alternative to the more risk averse. These funds are measured--not by how much they grow--but by how little/rarely they shrink. The crypto exchanges don't yet offer this tool, but there's certainly a need for it. For example, when someone is about to retire (or spend a lot out of their investment portfolio), they start selling stocks and putting that money into something more stable, so they don't need to risk selling even more stocks (in a bear market or volatile conditions) to make a large purchase.

Nobel laureate Milton Friedman and other economists note how frequently fiat currency inflation is defeated almost immediately by, "simply shutting off the printing presses," or, as during the Civil War, having the North destroy the money press of the confederacy. In order to maintain a much less volatile altcoin, its Founders are limited to managing a small 2-3% pre-mine (operating like the Bitcoin Foundation) which is used only for marketing, branding, and paying for the development of eCommerce for their altcoin. As the only real incentive for their work, get a larger post-mine or parallel mine of say, 10%, so that they have the strongest incentive possible to minimize volatility and to maximize market capitalization. This is similar to a company whose founders merely own shares, putting the Founders in the same boat with other shareholders .

I propose building a "FixedToken" altcoin constructed on a peer-based network based entirely on a wallet program written in Java that takes the entire computational workload and runs it as background threads on user systems. Rather than follow the practice of automatic production of new coins at fixed/declining rates, members of the FixedToken community perform CPU/GPU mining only when:
I. there is a growth in demand for the FixedToken, and only until the price increase is brought back down; or,
II. they are taking on cloud computing services. Since CPU's are very capable in handling double precision floating point operations (used in supercomputing tasks), miners would download small pieces of complex problems (protein folding, molecular simulations, analyzing SETI@home signals, auto crash simulations, raytracing for film) that are normally handled on expensive supercomputers and clusters. Leasing time on supercomputers costs universities and companies thousands of dollars per hour.

However, customers would pay the mining pool to perform their proof-of-work on our laptops and desktops by writing their program as thousands of threads in Java, OpenCL, or PyOpenCL, returning results to the end customer. Due to the advent of GPGPU's and and 8-32 core CPU's on the desktop with SIMD double-precision hardware, this may become a viable enterprise.

If the exchange prices for FixedToken start dropping, the open source software "shuts off the printing presses" immediately in order to prevent a condition where there are many more sellers than buyers, even using the cloud computing revenue to buy up and lock up excess Tokens from the exchanges. If the exchange prices increase too rapidly, anyone with a wallet could simply run the printing press program as a thread performing proof of stake (by entering a CATCHA in the wallet program to prove that they're human), running in the background on their laptop, Android Phone, desktop, or server with the minimum consumption of electricity, computation, and human time.

In other words, the intent of this proposed backup stable value fund, FixedToken, is to minimize all bad (electricity, risk, volatility, fraud, uncertainty, wasted human effort), while maximizing the value (stability, savings, market cap) that it gives back to those who use it. Having a much larger number of users with a small, growing stake in a less volatile currency might be more effective in this approach, because it brings more cloud computing power online for end customers and minimizes risk to buyers and holders of Tokens. The end customer could buy a FixedToken for $1 today (currently representing roughly 700 TFLOP/$), then use it in the future as computing power grows with Moore's Law, without a loss of initial investment. In 18 months, as computing power doubles, that Token would buy roughly 1400 TFLOP, depending on how technology progresses.

In the stock market today, money market funds, treasuries, time deposits, blue chip stocks, index funds, and stable value mutual funds are able to use smaller rewards to maintain a large, less volatile market capitalization over the short run. Precious metals are used to minimize risk over the long run. In the same way, an altcoin might be put together (using methods developed for Quarkcoin, Peercoin, and Namecoin) that splits the combined block chain reward resulting from increased value amongst many CPU-miners. However, that small number of coins should be less volatile, and might be accepted by eCommerce sites, investors, merchants, and individuals with earnings that are more averse to the the thorny price spikes that keep many pragmatists away, or that greatly limit their investment in any crypto-coin.