Tony Morris

@propensive we have to play fast and loose with implicits. No guarantee of uniqueness, can always pass different one: http://blog.ezyang.com/2014/07/type-classes-confluence-coherence-global-uniqueness/ …

@puffnfresh "[...] can always pass different one" ... no, you can prevent that, that's the reason why its safer than Haskell. @propensive

@oxnrtr how do you prevent it? i.e., disallow https://gist.github.com/tpolecat/38e323101d050c91a88f … / @puffnfresh @propensive

@tpolecat @puffnfresh @propensive Why would I want to prevent that? That's exactly the behavior I would implement, if it wouldn't exist.

@oxnrtr @tpolecat @puffnfresh @propensive Because you can achieve the same goal with less hack. It's a win that costs you nothing.

@dibblego @oxnrtr @tpolecat @puffnfresh @propensive I am curious what you folks think of http://homepages.inf.ed.ac.uk/wadler/papers/implicits/implicits.pdf ….