John Adler

> will this include how we use ipfs? Eventually, yes. But it will probably start with more conservative goals, e.g. just defining _what_ to sample. Then move on to _how_...

A recent topic that came up is "do we need to sample from both rows and columns for DAS?" (cf. https://github.com/lazyledger/lazyledger-core/pull/270) ## Mechanism 1 (from paper) The mechanism from the...

> The question is how much larger in the worst-case Without deduplication of sibling nodes, Merkle proofs against a single root vs Merkle proofs against `k` roots is basically the...

It might be possible to capture all the benefits of an in-protocol history MMR with a history MMR computed entirely client-side (i.e. add a new leaf to a locally-maintained MMR...

Not _just_ "voting power" (to use Tendermint lingo), but potentially also things like rewards and managing slashing/jailing. But in large part, yes it's mostly about changing the validator set each...

Don't think that token can even expire!

I think it's because the token only has read permissions, not write.

My recommendation is to start with native leader selection+staking through the Cosmos SDK staking app rather than through an EVM contract, since it involves much less work I think.

Thought dump: would also be nice if there was a DAG mode, where new "blocks" can be appended on top of any other block instead of just the latest.

> Choose and request some random leaves - a quarter of all leaves. Note that to _guarantee_ we can reconstruct the block, ~3/4 of shares of the EDS must be...