The Haskell Interlude

Peter is a professor at the University of Freiburg, and he was doing functional programming right when Haskell got started. So naturally we asked him about the early days of Haskell, and how from the start Peter pushed the envelope on what you could do with the type system and specifically with the type classes, from early web programming to program generation to session types. Come with us on a trip down memory lane!

In this episode, we focus on a particular part of Haskell: teaching it. To help us, we are joined by Jamie Willis who is a Teaching Fellow at Imperial College London. The episode explores the benefits of live coding, and why Haskell is the best language for teaching programming.

Franz Thoma is Principal Consultant at TNG Technology Consulting, and an organizer of MuniHac. Franz sees functional programming and Haskell as a tool for thinking about software, even if the project is not written in Haskell. We had a far-reaching conversation about the differences between functional and object-oriented programming and their languages, software architecture, and Haskell adoption in industry.

Welcome to the Haskell Interlude. Today, Matti and Mike talk to
Jeffrey Young. Jeff has had a long history of working with Haskell and
on ghc itself. We talk about what makes Haskell so compelling, the
good and bad of highly optimized code and the beauty of
well-modularized code, how to get into compiler development, and how
to benefit from Domain-Driven Design.

Jeff is currently on the job market - if you want to get in touch,
email him at mailto:[email protected].

We are joined by Kathrin Stark, a professor at Heriot-Watt University in Edinburgh. Kathrin works on program verification with proof assistants, so her focus is not exactly on Haskell, but on topics dear to Haskellers' hearts such as interactive theorem provers, writing correct programs, and the activities needed to produce them. We discuss many aspects of proofs and specifications, and the languages involved in the process, as well as verifying and producing provably correct neural networks.