Last week the Chicago Board Options Exchange became the first established exchange to offer Bitcoin futures. This past weekend, the world's largest futures exchange—the Chicago Mercantile Exchange—started offering them.

It happened quickly, and it's appropriate that this would happen in Chicago. The city is an economic power in large part because it pioneered the modern concept of futures, smoothing out the chaotic markets of Midwestern agriculture and allowing grain to be sold as a commodity as opposed to individual lots from individual farms. In doing so it took something real and made it virtual—rather than trading grain in hand in a place, the promise of grain to come could be bought and sold.

Now Chicago is home to a market in something entirely virtual, created on computers and existing only as a series of agreements between people to buy or sell them. Bitcoin is unconnected to any physical thing, or any government, or any established financial institution. And as its price skyrockets, people are trying to figure out what it is.

Even at its most basic level, the concept is profoundly strange: what this useful 2013 Quartz explainer calls "competitive bookkeeping." For all its futuristic wizardry, Bitcoin is vaguely based on our oldest concepts of what a currency is: People put in work to liberate a thing that has a value in the world and reap the reward that the world has put on it. Gold miners pan or dig for gold; Bitcoin miners use their computers to solve mathematical puzzles. More interestingly, Bitcoin has followed the path of real-world commodity mining; in the beginning it was very easy to do, because the mathematical puzzles were simple enough to do on a laptop. Early Bitcoin miners were like Gold Rush miners, rushing off with pans and pickaxes to a land of plenty. And like those prospectors, early Bitcoin miners have gotten extremely wealthy.

But Bitcoin is designed so that the math problems have gotten progressively harder, demanding larger and more expensive hardware to solve them. This is where Bitcoin genuinely crosses into the real world. As Eric Holthaus writes in Wired, the Bitcoin-mining industry is consuming more energy than 150 countries annually, and each day the capacity it adds requires about the annual energy consumption of Haiti. Just one transaction "requires the same amount of energy used to power nine homes in the U.S. for one day."

It's not unlike physical commodities in that sense. Once the easy pickings are gone, extracting the rest requires ever greater feats of engineering. With physical commodities, the extraction varies based on real-world conditions; the limits of Bitcoin are literally designed by people, including a permanent cap ensuring that only 21 million Bitcoins can ever be created.

This creates very strange problems for Bitcoin. Go back to the phrase "competitive bookkeeping" for a minute. Bitcoin transactions are protected by powerful cryptography, which guarantees that a transaction happened. If I give you a dollar bill, I've given you a physical thing that I can't conjure up again out of thin air. If I PayPal you a dollar, PayPal tells my bank to tell your bank that I've given you a dollar. Bitcoin is completely decentralized, so that agreement is done with mathematical problems.

"Every time somebody wants to send Bitcoins to somebody else, the transfer has to be validated by miners: They check the ledger to make sure the sender isn’t transferring money she doesn’t have," according to the Quartz explainer I cited before. "If the transfer checks out, miners add it to the ledger. Finally, to protect that ledger from getting hacked, miners seal it behind layers and layers of computational work—too much for a would-be fraudster to possibly complete."

Our bank pays guards and tellers and programmers to make sure the money is safe in storage and transit. Bitcoin does something similar: miners get Bitcoins for validating transactions. This is why the bookkeeping is competitive. Instead of putting all my money in Citi or some other bank and letting them handling all my transactions for me, those transactions can be validated by any Bitcoin miner.

Those transactions are validated in blocks, which is Bitcoin is based on a system called the blockchain. Each block is limited to one megabyte in size, a limit established by pseudonymous Bitcoin creator Satoshi Nakamoto. Adding a new block requires doing the hard math that keeps the system secure.

So you see where this is going. Every block is about 2,000 transactions. Validating each of the 2,000 transactions requires a lot of computation, which takes time and costs the miners money. So as more Bitcoins are mined and more people start using them, the demand to process transactions increases, so transactions get more expensive, slower, or both. Right now Bitcoin can process about three transactions a second, less than a thousandth of what Visa alone can do. You can buy a fast transaction or wait for a cheaper one. "It works a bit like Uber's surge pricing, except the user sets the fee based on how long she's prepared to wait for the transaction to go through," writes Lenoid Bershidsky in Bloomberg.

There's resistance to increasing the size of the block limit for a couple of reasons. One is that Satoshi created the limit, so it has something of the specter of holy writ; more abstractly, currencies are based on rules that everyone agrees to, so there is a natural conservatism even when the rules seem to be antiquated. Another is that larger blocks require even more computing, which would theoretically make Bitcoin less democratic. A fight over the limit was enough to cause a currency split back in August. And what was meant to be a nationless currency could run up against the limits of nation-states: Chinese Bitcoin miners fear that their bandwidth couldn't deal with large block sizes.

Another potential solution is an entirely different network, the Lightning Network, which overlays the Bitcoin network and… well, it's complicated. It would add complications to the appealing simple philosophy of Bitcoin, which was critical to its rise: a distributed, (relatively) democratic, completely independent and virtual form of currency. What happens when that philosophy runs up against technological limitations? What gets changed?

This is why Daniel Cooper argues in Engadget that Bitcoin is "failing as a currency," and makes a convincing argument that, at least for now, it's "locked into its destiny as a replacement for gold, not dollars and euros." Like gold, it's mined; it's hard to conduct small transactions with; it's expensive to move around. If it only makes sense for very large transactions, it could end up being a place to park money rather than spend it.

Right now the price of Bitcoin makes it difficult for regular people to get into the market. It's hard to say what they're missing because no one knows what to expect. But even if you're not in it, it's a fascinating story: the birth of a new currency and the lessons that we're learning (or re-learning) from scratch, in the context of a new era.