Tony Morris

.@purefn this whole issue is an instance of getting equations between weaker and stronger constructions backwards.

@bitemyapp hmm interesting. I see what you mean. I'm curious what @dibblego thinks.

@purefn @bitemyapp http://hackage.haskell.org/package/validation …

.@dibblego @purefn yeh. My point is to re-add Monad for AccValidation (monoidal Validation)

@bitemyapp @purefn It cannot be done such that ap=(<*>)

.@dibblego @purefn when AMP rolls in, it won't be 'liftM2 id' anymore. Also, that equality is not a monad law anyway.

@bitemyapp @purefn It is a reasonable expectation. I also don't see the penalty otherwise.

.@dibblego @purefn it's a reasonable expectation and there's an argument to be made for orthogonality of short-circuiting vs. monoid