Jeff Burdges

I'd think these wind up being bespoke for block production methods we provide: At least sassafras, babe, and aura should all be merkle proofs into specific parachain state, and a...

At a theory level, there is only really one caveat here: Alternative host implementations must make similar changes in their ed25519.

I noticed threads mentioned in https://github.com/WebAssembly/design/blob/master/Nondeterminism.md so I guess any threads we use should come from the host, so several runtimes' entry points? I suppose the runtime could make some...

https://github.com/paritytech/substrate/issues/1459#issuecomment-566045237

I'm not overly worried about resource limits because if we've some block dispute caused by resource limits then we've a straightforward bugs in resource allocation. We should declare the required...

Yes, I suppose wasmi/wasmtime version upgrades might pose the most real world threat, interesting thanks! :) I'd think the relay chain could dictate the required wasmtime versions, so backing checkers...

We cannot consider timeouts to be slashable, but afaik we have not carefully thought through the related issues: Imagine one validators declares that some candidate timed out. Is there any...

I've moved timeouts discussions to https://github.com/paritytech/polkadot/issues/1628 so maybe this issue could refocus on determinism.

We know more efficient storage migration schemes, but nobody clarified why we'd ever justify the implementation effort. I'll describe one fairly straightforward approach: We first require storage checkpoint protocol where...

Yes, we can run a while with just disabling, and then bring in the slashing very slowly. We can also just plan to refund the first say five slashes after...