Yotam Barnoy

I would prefer to see this introduced to the ecosystem and experimented with before the compiler adopts it. I don't find the examples in the changed files more readable. Instead,...

It just seems premature to me to introduce constructs into the compiler codebase when they haven't become popular in the ecosystem. It makes the codebase harder to understand, because we...

This seems like a very complex implementation for something that should never happen. Lazy shouldn't ever be used for synchronization. It seems like the only reason we're forced to use...

Are there any methods that would allow us to ditch the atomic after we force the value? After all, we only want to limit access to the unforced value --...

I have to admit that I didn't even know about the existence of the `Undefined` exception for recursive forces. This is an extreme edge case though, and I have to...

I would love to have a bank of such proposals so we can look at all the options holistically rather than one-by-one, and then choose the best choices. Would the...

On my personal project, I just came up with a need that may be helped by something like this proposal. For performance reasons, you often need to mix mutable and...

How many people actually make use of the add+shadowing functionality?

`Mutex.with_lock` definitely reads better to me than `Mutex.with_locked`, even if the latter is technically more accurate. Just to confuse things, I'll throw in a suggestion from the C++ world: `Mutex.guard`...

> Not a native english speaker but a bit suprised by this. Did you read it with the argument ? I.e: `Mutex.with_locked` implicitly requires you to work to parse it...