Nicolas Frisby

To avoid potential confusion: the 10x number I fed to Clarke was measured as the affect on the (downstream) Shelley ThreadNet tests' runtime -- I don't have intuition for how...

A knee-jerk reaction to what you've already shared: ``` class SlowOK slowIsOK :: (SlowOK => a) -> a slowIsOK = ... unsafeCoerce ... -- maybe also slowMustBeOK :: SlowOK =>...

> > This is probably a strawman; it seems too opinionated for a core library... > > I have to agree that it's a bit gnarly, and may not fit...

This would be a useful step, but it's not enough: we need to run this check on a "realistic" chain, and currently no Consensus test creates such a thing.

There's some context about how nightlies used to work here https://github.com/input-output-hk/ouroboros-network/issues/3892#issuecomment-1191618988 but I don't know that any of it will be applicable to this Issue.

Ledger Team has said we should start integrating Conway now. Because it's urgent, I'm going to try to bang it out myself, instead of leveraging it as an onboarding opportunity...

Also see: https://github.com/input-output-hk/ouroboros-network/blob/master/ouroboros-consensus/docs/AddingAnEra.md

bors try

We're still doing plenty of design discussion, so I've only looked through this diff quickly. First: great job updating the comments; I like it. Second: adding a common wrapper to...

Third: I'm having trouble shaking the thought that we maybe don't need to retain the rollback history when we drain the pipe and go through the find intersection path. Do...