I am constantly surprised by this point of view. Blockchain is a solution to a number of problems that were previously unsolvable, yet so much of the discourse is around the idea that decentralised trust doesn't really solve any useful problems. True many of the applications are hype, but we need to recognise these for what they are and move on to more interesting things.
To me the best example is the travel industry. Decades ago a small number of companies managed to get manual back office systems moved to centralised booking and reservation systems. This had a huge impact on the ability to sell inventory and open up the world of online travel agencies.
These centralised platforms are still around and they are now a huge problem. The software is bloated, complicated and slow, integrating with any GDS is painful and only ever allowed on their terms. It is amazing to me the amount of time and effort and startups that have gone into just trying to find the cheapest flight available.
This is fertile ground for decentralised trust. Having a local copy of the live flight inventory or a large aggregate set of hotel inventory would revolutionise the travel industry. Sure there are also applications around settlement and financing, but to me the most exciting thing is the ability to build customer applications on large sets of provably correct, trusted data. And to do this freely in the manner you chose with the technology you chose.