Callum Waters

> The new defaults should be tested in practice, to double-check on the performace impact, especially on high-activity channels. I could write a tool to grep chain validator sets and...

Some of these pages like `indexing-transactions` I don't think need touching. Sergio, Samantha and myself have had some prior discussion on working on the new ABCI docs so perhaps we...

I was thinking about this a little more. I know previously we had thought about writing a guide to building a public forum, but I think taking dydx's use case...

Well if you wanted to demonstrate a non-trivial case (i.e. where the txs are returned back to Tendermint), then an orderbook is a good example

My reading on this so far is that the epoch block has a lot of computation (distributing the lp rewards) and results in the emission of a lot of events....

> Should these be using log.NewLazySprintf to avoid hex formatting if info-level logging is disabled? Yeah I guess that makes sense

Hey @erikgrinaker, can you please elaborate on the issue - specifically what is the current verification process that is used for fast sync and what part of the lite client...

Thanks @milosevic. Just for further clarification, I would like to extract what you said into each of the features that need to be implemented, followed by any questions I might...

> If the trusted header is outside the trusted period, we cannot synchronize safely. Why is this the likely case? For fast sync, where we start our trusted state at...

Block sync will still need to be able to occur outside the evidence age parameter. I think there should be a decoupling in the detection logic (i.e. compare headers with...