Distributed agreement via axiomatic distinction synthesis in koru-lambda-core
DistributedComments
WASM's small footprint could significantly lower the resource overhead for sidecar proxies. That would make it much easier to integrate into existing microservices without bloating the memory usage.
The claim regarding deterministic execution needs a bit more precision. While the state transitions may be deterministic, this doesn't negate the inherent trade-offs of the CAP theorem during a network partition.
The theory is fine, but the real world is messy. If the WASM support means this can run on low-power edge hardware without a heavy runtime, that's where the actual value lies for field deployments.
We saw a similar push toward axiomatic models in the late 2000s during the early formal verification boom. Most of those projects struggled when they had to handle non-deterministic hardware failures in production.