Category Archives: TechRandom

What is Bitcoin, really?

So, I’m making this post to try and help some people I’ve encountered understand what alt-currency is and why it’s important, useful, and here to stay. The recent issues with Mt. Gox and Flexcoin are just birthing pains, don’t let them become the boogey-men that some groups would like.

First off, what is distributed analysis:
Arecibo_Observatory_Aerial_ViewYears ago, SETI (the Search for Extraterrestrial Intelligence) Institute figured out how to leverage distributed data analysis on the vast quantities of data it had pooled up from radio receivers like Arecibo Observatory. [pictured] This data, fragments of white noise from beyond our planet, *could* contain a message from an advanced alien race. (c.f. Contact by Carl Sagan.) The problem is, there is so much of it that a person sitting down to pour over the information would quickly be overwhelmed, and not be able to complete the task. This is where computers come in. If there is one thing that computers do well, it’s repetitive tasks, like looking for variations in white noise. But one computer would not be enough to analyze all this data, so SETI Institute had a program you could download and run as a ‘screen saver’ to analyze this data for them. It pulled a packet of information from SETI, analyzed it using the CPU (I believe) on your computer while it wasn’t being used by you, and then sent the results back. I’m not sure it was the first, but it was the first ‘distributed data miner’ that I’m aware of.

So, what’s Bitcoin (or any alt-currency):
Very basically, it’s an incentive. I have no idea how successful the SETI@home program was, but I imagine not very much so. It relied on people’s interests with no compensation, imagine NASA being entirely donation funded and you’ll get a picture of what I’m talking about here. Besides basic interest, there was no reason for people to participate, and the fewer people you have in a distributed system, the less likely it is to succeed. So the question became, how to get people (or more specifically their computers) involved in the distributed analysis. Enter crypto-currency.

The idea originated, according to bitcoin.org’s FAQ on a cryptography mailing list (cypherpunks) in 1998. It was fleshed out from there, but very little is known about the original creators of the program. Much like the concepts, the development of bitcoin software / algorithms (I’m not even really sure what to call it, but we’ll go with algorithms) has become distributed. The advantage here is that no one ‘owns’ Bitcoin. The disadvantage is that no one knows the full extent of what Bitcoin does. I can conjecture about the origins of money being attached to it, however, and here’s my conjecture:

How to turn data in to money:
I think the first big misunderstanding of crypto-currency is where the money comes from. It’s always been backed by other currencies, not by any intrinsic value of its own.There is no commodity to it, just investment in results. The modes of support are likely the same as the original modes, though the methods are different. You’d be surprised at how much money is out there for some of the calculations that are best performed by distributed data analysis. Finding prime numbers is a good example. Again, computers are really good at this kind of thing, where humans just aren’t, but the power needed to crunch this kind of number is just staggering. So you distribute that challenge among many different processors (one processor checks the number to be divisible by 2,3,5,7,etc and another picks up at 113, 127, 131,etc.) By distributing the calculations we can eliminate candidates faster, and eventually come to the next prime number. [Interestingly, primes are important for the other applications of distributed data analysis, namely cryptography, which I’ll discuss in a bit.] Finding the next prime number would be what we call ‘discovering a block’ which is basically just being the lucky processor in that line to nab the actual new highest prime number. So, in this example, you see how a payoff for that processing power would come from a ‘bounty’ set for the new highest prime’s discovery. This is overly simple, but it drives my next point home, so I need to make it this way.

Understanding a sea of data:
It’s impossible to tell exactly what calculations cyrpto-miners are doing. That’s sort of the point. But if we look at the overall process, we can tell a few things. First off, encoded information (that is, information that has been hashed to obfuscate it’s content) is likely being analyzed and opened. We can say this with a decent amount of certainty because it’s one of the best uses of distributed data analysis. Now, who is driving this decryption, and where it ends up, is well beyond my paygrade, I’ll let you puzzle that one. On the turn of this, new data is likely being encrypted over this network as well. I’d love to think that complex science is being done (or that the SETI information is being poured over) but I doubt that there is an altruistic bent to this anymore. What is clear is this, investors are investing, and data is churning.

When I buy a *coin what do I get:
Basically a promissory note. I mentioned Mt. Gox and Flexcoin at the top, and essentially you are at the mercy of whatever exchange you go with. Your *coin is worth whatever the going rate is, as long as you can get the company to pay for it. This may sound shady, but it’s the way we’ve done business ever since banks entered the picture. The only thing backing these exchanges is the money coming in from investors. If every *coin earner were to rush their exchanges at once, I doubt they’d have the capital on hand to shell out (and some probably don’t have any on hand, but they’ve invested it themselves.) So should you avoid this method of investment… probably not. Looking ahead, even as computers get faster and faster, the need for distributed data analysis is only going to increase. New methods of analyzing telescope data are pushing the boundaries of our understanding of the universe, cameras that work faster than light can tell us incredible amounts about the very basics of electromagnetic transmission, and even the tiniest amount of metadata in large enough quantities can implicate a person that might be very well protected otherwise. I don’t see the need for distributed analysis going away any time soon, and so I don’t see the need for incentivizing it going away soon either.

Think I missed the mark? Have something to add? Comment here: