Crypto is the first sector where rampant speculation began before mass-market consumer applications were on the horizon, and that's owing to the initial prototype having its own self-enclosed and permissionless financial system and token that could be speculated on by the general public. So the gulf between when the tokens accrued significant speculative value, and when the sector produces broadly useful applications, is exceptionally large.
That means a lot of malinvestment in nonsense projects, but also significant capital to accelerate the building out of the useful parts of the technology that will get us to broadly useful applications. I provided a list of what I believe are the necessary building blocks for mass-adoption that are being worked on right now:
https://news.ycombinator.com/item?id=31797395
Optimistically, I'd say we're two years away from production quality Rollups, which will be able to host mass-consumer applications. More cautiously/conservatively, I'd say we're four to five years away, and with emerging technology, I think it's better to go with the cautious/conservative estimates.