Perhaps the most important abstraction a Haskell programmer must understand to effectively write modern Haskell code, beyond the level of the monad, is the monad transformer, a way to compose monads together in a limited fashion. One frustrating downside to monad transformers is a proliferation of
lifts, which explicitly indicate which monad in a transformer “stack” a particular computation should run in. Fortunately, the venerable mtl provides typeclasses that make this lifting mostly automatic, using typeclass machinery to insert
lift where appropriate.
Less fortunately, the mtl approach does not actually eliminate
lift entirely, it simply moves it from use sites to instances. This requires a small zoo of extraordinarily boilerplate-y instances, most of which simply implement each typeclass method using
lift. While we cannot eliminate the instances entirely without somewhat dangerous techniques like overlapping instances, we can automatically derive them using features of modern GHC, eliminating the truly unnecessary boilerplate.
Since I published my blog post introducing Rascal, I’ve gotten some amazing feedback, more than I had ever anticipated! One of the things that was pointed out, though, is that Rascal is a language that already exists. Given that the name “Rascal” came from a mixture of “Racket” and “Haskell”, I always had an alternative named planned, and that’s “Hackett”. So, to avoid confusion as much as possible, Rascal is now known as Hackett.
With that out of the way, I also want to answer some of the other questions I received, both to hopefully clear up some confusion and to have something I can point to if I get the same questions in the future.
Note: since the writing of this blog post, Rascal has been renamed to Hackett. You can read about why in the followup blog post.
“Hey! You got your Haskell in my Racket!”
“No, you got your Racket in my Haskell!”
Welcome to the Rascal programming language.
Object-oriented programming languages make unit testing easy by providing obvious boundaries between units of code in the form of classes and interfaces. These boundaries make it easy to stub out parts of a system to test functionality in isolation, which makes it possible to write fast, deterministic test suites that are robust in the face of change. When writing Haskell, it can be unclear how to accomplish the same goals: even inside pure code, it can become difficult to test a particular code path without also testing all its collaborators.
Fortunately, by taking advantage of Haskell’s expressive type system, it’s possible to not only achieve parity with object-oriented testing techniques, but also to provide stronger static guarantees as well. Furthermore, it’s all possible without resorting to extra-linguistic hacks that static object-oriented languages sometimes use for mocking, such as dynamic bytecode generation.
Currently, npm is the package manager for the frontend world. Sure, there are alternatives, but for the time being, npm seems to have won. Even tools like Bower are being pushed to the wayside in favor of the One True Package Manager, but what’s most interesting to me is npm’s relatively novel approach to dependency management. Unfortunately, in my experience, it is actually not particularly well understood, so consider this an attempt to clarify how exactly it works and how it affects you as a user or package developer.
I started programming in elementary school.
When I was young, I was fascinated by the idea of automation. I loathed doing the same repetitive task over and over again, and I always yearned for a way to solve the general problem. When I learned about programming, I was immediately hooked: it was so easy to turn repetitive tasks into automated pipelines that would free me from ever having to do the same dull, frustrating exercise ever again.
Of course, one of the first things I found out once I’d started was that nothing is ever quite so simple. Before long, my solutions to eliminate repetition grew repetitive, and it became clear I spent a lot of time typing out the same things, over and over again, creating the very problem I had initially set out to destroy. It was through this that I grew interested in functions, classes, and other repetition-reducing aids, and soon enough, I discovered the wonderful world of abstraction.
Haskell was a language I had played with in the past but never really used for anything terribly practical, but lately I think I can confidently say that it really is an incredible programming language. At the same time, it has some significant drawbacks, too, though probably not the ones people expect. I certainly wasn’t prepared for some of the areas where Haskell would blow me away, nor was I capable of realizing which parts would leave me hopelessly frustrated until I actually sat down and started writing lots and lots of code.
Racket ships with
racket/generic, a system for defining generic methods, functions that work differently depending on what sort of value they are supplied. I have made heavy use of this feature in my collections library, and it has worked well for my needs, but that system does have a bit of a limitation: it only supports single dispatch. Method implementations may only be chosen based on a single argument, so multiple dispatch is impossible.
Macros are one of Racket’s flagship features, and its macro system really is state of the art. Of course, it can sometimes be difficult to demonstrate why macros are so highly esteemed, in part because it can be hard to find self-contained examples of using macros in practice. Of course, one thing that macros are perfect for is filling a “hole” in the language by introducing a feature a language lacks, and one of those features in Typed Racket is ADTs.
Elm is a wonderful language for building web apps, and I love so much of its approach to language design. Elm does so many things right straight out of the box, and that’s a real breath of fresh air in the intersection of functional programming and web development. Still, it gets one thing wrong, and unfortunately, that one thing is incredibly important. Elm took the “functions” out of “functional record types”.