This article does a decent job at casting the decentralized database argument as disruptive innovation without resorting to hype. But I wonder about the other f

This article does a decent job at casting the decentralized database argument as disruptive innovation without resorting to hype. But I wonder about the other fundamental drawbacks of blockchain for developing new applications: - Immutability: building any software requires iteration, and therefore mutability. How do you build a startup if you can't ALTER TABLE on your database (or it requires a new database)? - Irreversibility: if there are any bugs, or mistakes in the code, you can't fix it later (and in fact, someone may be able to steal a lot of your tokens) - Dependency on real-world data sources: even if smart contracts are ironclad, they can dependent on potentially erroneous or manipulable data sources - Off-chain centralization: many of these blockchain apps such as the play-to-earn games have a lot of centralized control that is not on the blockchain and under the discretion of the developer. Probably because of the reasons above At a meta level, the hardest problems are people problems and not solvable by technology. Trying to solve people problems with ironclad code seems like a fool's errand. If there was a killer app that wasn't handicapped by the issues above, that would make me a believer.