Alexis King

@AlexKnauth I disagree, because the current behavior violates the lens laws: `(lens-transform foo-a-lens x identity)` can potentially return a value that is not `equal?` to `x`.

As far as I know, no, not directly (otherwise I would be using it!). Any struct type can be used as a supertype with [`make-struct-type`](http://docs.racket-lang.org/reference/creatingmorestructs.html#%28def._%28%28quote._~23~25kernel%29._make-struct-type%29%29), so if the struct type...

@LeifAndersen I’m not sure that it’s an issue with renaming, specifically, but more an issue with how the information structs provide is insufficient for a lot of practical use-cases (both...

> Create your struct type, and never ever ever give anyone access to your struct outside of a closure. Yeah, this is effectively what I mentioned a little bit earlier...

After more investigation, I’ve found this isn’t actually related to the fixed-function versus programmable pipeline, but rather when the WGL context is created. If I add `(with-gl-context void)` to the...

The machine I’m testing this on has an NVIDIA GPU. The issue appears to manifest when the WGL context is created before the frame the canvas belongs to is made...

> I don't quite see how this operation is a _closure_ operation. It isn’t (which is why I said I don’t really know what it should be named). > Now...

> Since you'd presumably annotate your grammar with Paul's amazing binding annotations, you should demand that in addition to substitute define-language also generates functions such as free-vars. That would certainly...

Could you elaborate on that? I understand the intent behind executable building (though I thought Racket had more elegant ways to deal with that), but why are `set!` transformers necessary...

@rakeshkky @coco98 Could I have some more context on this? Is there a feature request somewhere for this change that has informed the protocol design? Have we talked to users...