Alexis King

Technically, I think you can implement this yourself dynamically with [`module->exports`](http://docs.racket-lang.org/reference/Module_Names_and_Loading.html#%28def._%28%28quote._~23~25kernel%29._module-~3eexports%29%29). I could understand a desire to have it built-in, though.

Yes, this is semi-planned, I just don't know exactly how I want to do it yet. The solution you mentioned is definitely the simplest one, though.

@0x777 Why is this hard?

Actually looking at the source: this seems to be intended behavior. `rfc6455-send!` checks if the connection is closed, and if it is, it does nothing and returns. I find this...

Matthew responded to this idea negatively in [a mailing list thread from 2015](https://groups.google.com/d/msg/racket-dev/8BFT-RBDp9E/NJZjK0th8vUJ): > That would create lots of trouble, unfortunately. > > Search paths are painful in general. It...

@shekari Could you explain the typing difference between immutable hash tables and readable hash tables? Why wouldn't they be equivalent?

@samth Oh, I see, that makes sense. `ImmutableHashTable` would still get you covariance, though, yes? Wouldn't having explicit mutability still be helpful, though? Contracts support it, and then `hash-set!` and...

Should we fork `HashTable` into `ReadableHashTable`, `ImmutableHashTable`, and `MutableHashTable`? Then just leave `HashTable` as-is? It would make `HashTable` a sort of silly type to use for anything, but I guess...

With `Set`, I think we need to at least support mutable hash sets. I ran into this originally when trying to use mutable hash sets in typed code, and the...

You mean make `Set` and `MutableSet` entirely disparate types, like `Pair` and `MPair`? I think having the type structure of the proposed hash table rework would be nice.