The ascendence of crypto markets has been humbling to many, including myself. There’s a lot that I got wrong. But, I contend that in just two statements I can convey what us currently salient regarding crypto markets. The first statement I will make seems to be true, and the second is certainly true.

My first statement: the value of crypto assets is still primarily driven by speculation. I am not aware of any good way to quantify this, but it still seems true.

My second statement: much of the value of crypto assets is now certainly real, by any reasonable definition of real.

As for what a “reasonable definition” is, I would argue that most people agree that the dollar has some real value in the present, and therefore a dollar or any synthetic instrument representing a dollar is real. The mere existence of stable coins then suggests that at least some crypto assets have real value. And on the basis of that, the income generated from interest on those assets through DeFi lending platforms is also real.

Now, I suppose I could have made a simpler argument that, unlike my current argument around stable coins and DeFi, would have been equally true five years ago: I could have just argued that the exchange, or fungibility, between crypto assets and fiat currencies implied that those assets represented real, if volatile, value. Honestly, I could have. That is part of what I missed regarding tokens as an asset. But my more significant oversight relates the the nature of tokens themselves, and the complex systems that can emerge with just a few assumptions, paralleling computing itself.

Our computers are magical because all of their capabilities emerge from the fundamental inputs of 0s and 1s; bits, and logic gates. I am only slightly simplifying here; this is essentially true, and incredible. For a long time, it seemed we had just invented counting machines, but then one day we had a mouse and a GUI.

Similarly, every function of a dApp emerges from the fundamental inputs of tokens, smart contracts, and marketplaces for those tokens. So long as all tokens are fungible and tradable, which I have learned through my recent experimentation that they essentially are, one can build highly complex systems with features far more complex than their underlying units.

Nobody sees 0s, 1s, or logic gates when they’re watching a YouTube video, unless they’re watching Stanford or MIT’s open sourced introductory computer science course videos. Similarly, relatively complex applications, such as fully distributed blockchain based cloud computing, are operating with the underlying token architecture completely abstracted away from end users. Loading some ETH into MetaMask and wandering the world of dApps is a like a stroll through the future, hamstrung as it currently may be by exorbitant gas fees (themself a function of the increasing real value of crypto networks, and also self-resolving through the aforementioned principle of fungibility enabling side chain trading with orders of magnitude greater efficiency). In this regard, the token is like the bit. It can abstracted away, but also underpin almost any system. And just as improvements in using bits having enabled seemingly anything in computing, improvements in using tokens are on track to enable seemingly anything on blockchains.

In general computing, Moore’s law allowed for the scaling to the incredible complexity represented by computers today. While Moore’s Law related specifically to transistor density, we also had parallel exponential increases in storage density (also inversely correlated with energy use for read/write, as greater compute is also inversely related to energy use).

In cryptocurrencies, the scaling to incredible complexity will be driven by lower blockchain transaction fees. Due to fungibility of economic tokens, the main networks themselves don’t ever have to become more efficient, so long as they maintain their fidelity. All transactions can occur on side chains. When the first side chains are no longer the most efficient, new side chains can chain off them, creating an infinite chain of chains with a gradient descent towards lowest fees.

We are already seeing this happen through improvements in network architectures, and it’s also worth noting that all of these efficiency improvements also benefit in lockstep from the general increases in computing and storage efficiency that are still ongoing. Apple is still somehow able to make their mobile chips significantly faster while using less power every year. While it may cost $100 to trade $1 on ETH today, in evaluating the future of all that is built on cryptocurrencies, one must instead consider that in the long run it will cost basically nothing to trade anything.

Now is the moment that any keen observed must acknowledge that cryptocurrencies can represent anything that could plausibly be an economic exchange on some atomic level, which especially is our capitalist world, is almost everything. Looking at cryptocurrency networks today as merely instruments for participating in speculative bubbles and subverting laws would be like looking at IBM mainframes in the 1970s and thinking computers will never be useful for anything beyond counting numbers.

In pursuit of magic