Jonathan Immanuel Brachthäuser

> I wonder if this is possible in Effekt without using the special syntax for bidirectional effects. What do you mean with "special syntax for bidirectional effects"? ``` [error] ./issue.effekt:4:27:...

Thanks for your questions, pointers, and fruitful discussion. I'll try to address the questions as best as I can. ### 1. Encoding In the following example failingGet1 will type check,...

## Regarding 3) I agree that it is not always desirable to duplicate `p`. Especially, since it will *always* be the same value. We chose this particular syntax / implementation,...

Thank you for all your valuable comments and references. ## 3 Wow, that's a tricky one. We can only hack this together using mutable state, I guess: ```scala interface Nd...

``` EXPR1 || EXPR2 = infixOr { () => EXPR1 } { () => EXPR2 } EXPR1 && EXPR2 = infixAnd { () => EXPR1 } { () => EXPR2...

The second is probably better. In the future, then, we need to optimize the inliner to make sure this is actually inlined.

This is a blocker for #449

The machine IR looks fine to me ```scala def main_4200 = { def loop_4204 = { let utf8_string_literal_6221 = "\21"; let x_6220 = println_310(utf8_string_literal_6221); subst [__4284 !-> x_6220]; subst [ev6212_6212...

Would it be an option to take subtyping into account, when checking this? Or would this be the same as option 2?

Hey this one line has a very long history. We discussed it here at some point: https://github.com/effekt-lang/effekt/issues/16 - Introduced `env`: https://github.com/effekt-lang/effekt/commit/69655ae917b531f96516170dc24a2d51e23a4bcb#diff-5634c415cd8c8504fdb973a3ed092300b43c4b8fc1e184f7249eb29a55511f91R85 - Removed `env`: https://github.com/effekt-lang/effekt/commit/31779448a6254573af6ec711d724f259e5cfadfb - Reintroduced `env`: https://github.com/effekt-lang/effekt/commit/234314ffca1784803d81a766dcc167b5f697d2ad -...