A few weeks ago, we announced the first public beta of stack, a new build tool for Haskell. Since then we've seen a huge amount of work on stack: code contributions, feature requests, bug reports, and design discussions. In other words, the response we've had from the community has been amazing. Thank you for helping us push forward with improved Haskell tooling.
Today we're announcing a new milestone: we believe that stack is now stable enough to be the standard build tool for most Haskell development. We're taking the beta label off of stack, and recommending people dive in. Please keep in mind that stack is still young software, and there are likely corner cases that haven't been worked out fully. However, the feedback we've received from other users has indicated that stack is ready. The stack team itself (both inside and outside of FP Complete) has been using stack for months now as its primary tool. And at FP Complete, we've already moved our biggest projects over to it.
One question which I've personally held off on addressing until now is a direct comparison between stack and other build tools, cabal being the most obvious comparison. Expect a blog post on the subject in the near future. For now, my recommendation is: try out stack, and see if you like it.
While stack started with the needs of existing Commercial Haskell users, the goal we've always had in mind for it is a tool that lowers the barrier to entry for non-Haskellers. As we've discussed before, FP Complete has done quite a bit of research on this topic, and the data shows that build tool issues have been an obstacle to adoption. We want stack to solve that, and we want your help.
Firstly: keep doing what you've been doing! Using stack, testing it, providing feedback, reporting bugs; all of these make stack better for both our existing community and for newcomers. But the next step is marketing Haskell outside of our community. All of us Haskellers already know how wonderful a language Haskell is. Let's send a clear message to the rest of the world that it's time they try Haskell. stack's aim is to remove those initial barriers. But now we need to tell people what Haskell has to offer.
If you're interested in getting involved in this kind of a push, please join and discuss on the Commercial Haskell mailing list. There's no preexisting formula for how this should be done. But the basic idea I have right now is:
I'm sure others will have better ideas than I on how to accomplish this. If you want to see Haskell take off as not only a successful, powerful language (which it is now), but as a more popular language too, come get involved!
Do you like this blog post and need help with industrial Haskell, Rust or DevOps? Contact us.