Manish Goregaokar

Nah, just whitelist `[service]`. Or we'll call it `[third-party]` or `[external]`

> Would you consider closing the issue until there is more evidence something needs to be done? As an unsafe reviewer the standard I tend to apply is that it...

@Byron I think that's a start but still not easily reviewed. I'm also not convinced on that Send and Sync bound, the safety comments on the Send and Sync bounds...

Yeah the main thing to be careful about is that `&T` is `Send` only if `T` is both `Sync` and `Send`, which is not reflected in those bounds. a RefCell...

No, your type is logically an `&T`, it's not about whether `T` is an `&Type`, it's about whether it is `RefCell`. If your `T` is allowed to be `RefCell` your...

> Now the question is how this "array with members that are (trust me) owned by a single thread at a time (but accessed by reference)" can be expressed so...

Okay, the invariant is trickier because `future_child_offsets` is also weird.

It's fine I figured it out, I've got a partial PR. It wasn't that it's confusing, it's that the invariant has multiple caveats which is tricky for invariants of this...

https://github.com/Byron/gitoxide/pull/1237

Native origins in the spec are intended to just be transforms, but you can define them to track something so the transform changes per-frame. This is essentially shortform for "every...