Incorrect test assertion for Chapter4, Task 8 about learn4haskell

[vrom911]

This is a correct assertion and we expect the solution to respect this test case.

[CharlesSchimmel]

Oh no how embarrassing for me. If that's the case, I don't think the prompt for the task makes that clear. The prompt doesn't indicate to return m False if the first term is False instead of lifting the whole function into m, which would differ it from something from liftM2 (&&)

learn4haskell/src/Chapter4.hs

Line 616 in 123981d

=💣= Task 8*: Before the Final Boss

Edit: I was told that "Oh no how embarrassing for me" could be interpreted sarcastically, I just meant that it's embarrassing to think I knew better than the learning material, and then also be wrong

[vrom911]

This is the lifted version of && which is also lazy, so no surprises in how it works. You don't expect False && undefined fail, you expect it to return False, so this is a sane assumption. We already specified the constraint of Monad in the function as well, which is also a hint. We don't want to provide problem solutions in the task description. We can always explain everything in the review if people would have questions about the task.

[CharlesSchimmel]

I'm worried that this is confusing as it differs from something like (&&) <$> (Just False) <*> Nothing as well, and communicates the idea that all lifted lazy functions will act that way. I don't know enough to articulate it, but it seems that even though (&&) is lazy on its own, lifted versions still evaluate the context of the second term. Anyway I won't belabor this more, thanks for taking the time to explain, and these exercises were great.

[vrom911]

I am not sure what you are trying to prove in here, but your solution (&&) <$> (Just False) <*> Nothing doesn't even use Monad. That means that it just could not be lazy, as laziness is only achievable in here due to Monad. Applicative apriori won't give you that.
Moreover, this is only an exercise. It is not the library that you have to use. If you think that all people are doing a poor job implementing this function not your way, try to open this issue in all the libraries that provide such a function: directory, extra, protolude and many many more!

Here is not the place to push your visions on laziness and strictness as this is the learning material on Monads that we provide.

[CharlesSchimmel]

I'm sorry if I came off as adversarial, I am just trying to reconcile what I thought I understood with the explanation you gave. I'm am
barely past total beginner and thought I found something that was overlooked, and trying to help and learn more. I'm just trying to understand why liftM2 (&&) would not be equivalent.