You’re seeing everything just fine.
Imagine you’ve spent your whole life investing millions of dollars in entrepreneurs. Building models and testing hypotheses with no real agency: you can’t control what the CEO does or the product the engineering team builds or the way they market the product. Your job is to provide returns for your LPs. That’s it.
Now imagine, instead of spending thousands of hours interacting with and managing people, their problems, watching painfully as they try to figure out the market they specialize in, I tell you that you can invest in something else. That thing, we’ll call it a token. This token does work. It can be used to compute something. Or helps make something easier. I’m not clear on the specifics, but I can tell you one thing: if you buy it, and if you tell other people about it, merely the act of owning it increases its value.
I have just given you something you don’t get a lot of in your career, true, direct self agency. You are unequivocally sure that you are creating value because it’s you who is selling the value! You bought, the price rose, you know if other people get excited the price will rise again. It’s exciting, you don’t have to deal with people mucking about directly, and you’re doing your job: generating returns.
The house of cards here is the assumption that these tokens can be used for something, anything beyond a store of value. Because if that’s true, it’s not really speculation. You’re an early adopter of a paradigm-changing protocol.
The default assumption here is that engineers are like ants: we build for the sake of building. Give us a canvas and we will build dApps (or what-have-you) just for the sake of building them.
The truth is, engineers are not ants. We build things to make our lives easier or more fulfilling. I started programming, for example, to build video games. Many people build things to automate their homes. When you build for an employer, the end goal is always to do something for the customer.
The draw of Ethereum, seems to me, to be that eventually somebody will have to build something insanely valuable, right? Statistically, it’s gotta happen. I think that’s a ridiculous premise: if centralized tools are always faster with more mature ecosystems, engineers will always go there to solve their daily problems. The question to me really seems to be, “do we need truly immutable ledgers and transaction histories? Do we need 100.0% reliability in transaction history at the cost of orders of magnitude more energy expenditure, or is 99.9999999% okay?” As a cute anecdote, Life has been remarkably successful with a 10^-7 error rate in DNA polymerase. I have a feeling most industries will be fine without, and it’s actually probably cheaper to have a reasonable facsimile (or statistically insignificant approximation) of immutability (human intervention and accountants included) than a guarantee. Especially if, in order to be guaranteed, it needs to be public.
So, what you’re seeing is speculation that engineers could build something one day using an application development paradigm that’s not actually clearly or obviously more powerful. It’s worse speculation than betting on startups with teams of people who reliably execute, masked by the illusion of self agency as an owner and promoter of a token.
tl;dr: Somebody remind me of this post in a decade and we’ll see where it all ends up. The technology will grow and evolve. I’m not sold on any of this, and investors who aren’t building anything but themselves are holding tokens telling people to go experiment and build things isn’t an attractive signal. (It seems like VC flipped on its head and the incentives are poor for engineers - so, for now, let these investors play hot potato with each other for a while. I’ve seen a few of Vitalik’s tweets and it seems even he’s concerned about these incentives.)