Content pfp
Content
@
0 reply
0 recast
0 reaction

avi pfp
avi
@avichalp.eth
this is a nice articulation of why prover networks should have a metered execution like evm
2 replies
0 recast
5 reactions

Cassie Heart pfp
Cassie Heart
@cassie
The same can also happen for tripartite MPC markets (buyer - garbler, seller - executor, validator - verifier): a verifier of execution cannot disambiguate a slow executor, a poor connection between garbler and executor, or a faulty executor, without having an explicit cap (either in circuit size or time, or both) We have a continuation model, where there's an explicit upper bound. It avoids the "trusted arbitrator" issue
1 reply
0 recast
2 reactions

avi pfp
avi
@avichalp.eth
how does continuations prevent never ending loops?
1 reply
0 recast
0 reaction

Cassie Heart pfp
Cassie Heart
@cassie
They don't! But they do bound the maximum allotted work per interval
0 reply
0 recast
1 reaction