This gets to the heart of the problem for all type systems supporting deduction. Haskell is slightly better, but at a certain level of abstraction a function type can’t be separated from its implementation. https://brevzin.github.io/c++/2018/10/20/concepts-declarations/ …
-
-
I have no idea what this means but no one was commenting so i did not want you to feel lonely
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I often read
@TimSweeneyEpic’s tweets thinking: “I’m a C++ programmer now. Surely if I read enough of these, they’ll start to make sense!” They never do.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Tony Hoare: hate to be an I told you so.
-
Thoughts on OCaml?
- 3 more replies
New conversation -
-
-
I definitely agree with the first part, although I doubt anyone would say you've floundered much overall. I've been using Typescript some in a project the past month. It has a pragmatic-for-web-dev feel of letting you sprinkle in some advanced type features as desired.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
https://twitter.com/nuopnu/status/929421223163056129 … It's going overboard that's unscalable in the first place. The complexity, or seemingly inexpressible implementations might be redundant. While you were fighting the compiler for the whole day, you could've written an imperative checker within an hour.
-
Hooking up that checker(s) to some another meta-consistency checker also isn't a rocket science. Decouple it and build horizontal (not vertical) layers, instead of mixing it all up in a one single bowl of soup.
End of conversation
New conversation -
-
-
Millimeters times millimeters results in millimeters squared, validated at compile time.pic.twitter.com/5bKbAvUqiH
-
If you write a lengthy arithmetic function and use the wrong type somewhere along the way, the compiler will whack you.
End of conversation
New conversation -
-
-
“Functional Logic Overloading” http://deinprogramm.de/sperber/papers/functional-logic-overloading.pdf …
-
Note that the link which points to how the Success kind Singleton is defined is now invalid. This is a shame since the typeclass example is IMO (as an Haskell neophyte) the most illustrative of
@TimSweeneyEpic ‘s concern. How the link with == is done is IMO the core of the issue.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.