> Being resilient to unreliable connections in general is a feature that was important initially and has never stopped being relevant.
That is a very abstract way of referring to the feature. Suppose 100 years from now the entire world runs on decentralized blockchains. You could make the same exact abstract point: "Well, leaderless consensus was the original design goal, and it was achieved immediately, so, it was useful from the very beginning".
Of course the core innovation will always be integral to the relevance of that innovation. The point though is that the specific applications envisioned for the internet were very narrow and almost completely unrelated to what we use it for today, when it was invented.
> It seems to me that people knew exactly what they wanted to do from the get-go, it was useful almost immediately, and even though we have exponentially more people, computers, and applications today, the internet itself hasn't fundamentally changed.
Again, only if you conceptualize "knew what they wanted" in a very abstract way that makes the point tautological.
Cryptocurrency as it stands today is already useful. There is no real debate about that. I personally find it useful as an alternative value store. The use cases might be narrow, and not nearly as revolutionary as some of its proponents think (yet), but they do exist. If I am the only person in the world that derives utility from it, then its utility is still non-zero.
But abstracting from crypto specifically, the idea that technologies need to have specific applications in mind before they are pursued is just not how the history of technology has ever worked. Many of the best discoveries were made entirely by accident, in the pursuit of random, interesting things. We make progress by pulling at threads, and pushing boundaries down avenues that lead to highly general, abstract, and powerful primitives. I don't know about you, but I think Byzantine fault tolerant contract execution is a profoundly powerful, abstract, and general primitive, much like the internet itself. We do not know what its specific applications will be, just like we didn't know about Uber, Amazon, or Google in 1969.
Go take a look at the history of the early internet. As recently as the early 90s (more than 20 years after ARPANet came online!) people were saying commerce would never happen on the internet. Nobody would ever feel comfortable buying things that way, it was absurd. Here's Paul Krugman, in the NYT all the way in 1998 saying as much:
https://www.snopes.com/fact-check/paul-krugman-internets-eff...
It's way too easy to see where we are now with the internet as inevitable. But it was extremely far from obvious 30 years ago. Most of the world thought it was a toy. Thought business and commerce and ordinary people would never take it seriously. This was not a fringe view, it was the mainstream consensus.
Now, I don't want to fall into the trap of arguing that because the internet's future was obscure, and crypto's future is obscure, therefore crypto will be as successful as the internet. That's stupid. What I do want to argue is that the premise that technologies need to have clear-cut and specific applications before being pursued is wildly out of sync with how some of the most important technologies in human history have come about.