It's really unfortunate that rent-seekers (may have) ruined blockchain as a technology. The fundamental problem IMO is that the most useful applications have no speculative potential, and structuring blockchain applications in a way that creates speculative potential seems to inherently reduce their usefulness.
Decentralized currency on the blockchain? Great! But a currency's much more useful if its value is stable, so if you want to maximize its usefulness (assuming you don't want to tie its value to an existing currency, as existing "stablecoins" do) you probably need a mechanism to increase or decrease the money supply in response to changes in demand. Poof, there goes the speculative potential.
Or consider another potential use case that I've thought about a lot, raising funding for co-ops: it's hard for a startup to raise capital without issuing equity, and decentralizing that process could create realistic alternatives to the stock corporation ownership structure for big nonfinancial organizations. Or similarly, as an example that might be more familiar or important to folks on HN, it would make a lot of sense as a decentralized funding mechanism for FOSS development. (Yes, there have been attempts to do this, which I won't link here.) But no one's content to just build a protocol that lets people put (say) ETH in a pot to fund business investment or software features and then let others claim that ETH in exchange for implementing the stuff. No, you have to create your own DecentralizedFundingToken with a structure that's contrived to create speculative potential to reward "early adopters" and then do an ICO.
I currently own no cryptocurrencies. I owned <$10k in Bitcoin between 2011 and ~2014 (no, not at mtgox).