Marcel Champagne

@mockersf Thank you so much for investigating! I tried your suggested workaround and it seems to have resolved the issue (unless I've yet to encounter the panic😅). Your hypothesis makes...

Can confirm that the workaround fix is indeed working as I've been running with it for a couple of days without any panics. I also doubled checked and my full...

My biggest problem with the implementation of ComputedStates is as follows. Keep in mind this is a user-facing perspective and I don't have the full context on the actual implementation...

> This is actually something I disagree with for a few reasons: > > First, there are situations where a state's existence doesn't really make sense, for example: "Paused/UnPaused" is...

> We don't have to get the exact API perfect on the first go: I'd much rather move forward and ship this, then adapt based on user feedback and experimentation....

One more thought. If you want optional computed state it seems reasonable for users to: a) Implement their computed states on `Option` OR b) Explicitly define computed state with an...

@lee-orr Ahh, by queried directly you just mean a system which uses Res? Yes, that quite common in my experience and I use that pattern throughout my game as well....

@lee-orr That all makes sense! I guess the trade-off here is do we optimize the APIs (compute trait method, Res, etc.) for optional state or non-optional state? My intuition was...

@MiniaczQ > In this example there are no ZSTs, but GameState computed state makes absolutely no sense during AppState::Menu and it's unreasonable to expect every state to introduce a "none"...

> It's unreasonable to scale that up, it looks fine with one enum, but you will have to do that to every enum. > The same 'none' variant to every...