ok so "decentralized AWS" doesn't actually make sense, but it's a bit hard to see why so here's the nuance: the spirit of the question is "I want to achieve 99.999% uptime". And in particular, negate the risk exposed from yesterday's outage where an entire AWS region can go offline. This could be addressed by deploying fallbacks across either 1) multiple AWS regions or 2) multiple cloud providers. from an engineering perspective, I really struggle to think of any non-financial use case where it makes sense to spend the significant eng effort to deploy and maintain logic on decentralized infra. In conclusion- its hard to see, but in practice it's larp bc you can just deploy fallbacks across more locations.