More Fixpoints!
summary
Library implements lazy sets and lazy booleans by wrapping an imperative propagator network behind a pure API using unsafePerformIO, so recursive equations become productive and cyclic graphs now yield correct reflexive‑transitive closures. Implementation requires monotone operations and a unique (least) fixpoint per equation set, relies on heap sharing, and exposes risks: termination, reentrancy, space leaks and changes to observable optimizations like lambda lifting. Artifacts and source are published with evaluated reproducible examples.
AI-written summary · about
chapters (10)
- 0:00Introduction to purity
- 0:37Defining purity in code
- 1:47Laziness and recursion
- 3:23Knot tying and graphs
- 5:55Challenges with strictness
- 8:46Lazy libraries and IO
- 10:44Broader programming potential
- 13:22Alternatives and trade-offs
- 17:25Implementation and semantics
- 19:40Q&A and discussion
related talks
27mHaskell equations, thirty-eight years later
51mGoing bananas with recursion schemes for fixed point data types
32mFunctional Programming Over Sets with Atoms
19mPure Borrow: Linear Haskell Meets Rust-Style Borrowing
36mConcrete Functional Programming
14mFundamentalist Functional Programming
43mAn Imperishable Wonderland of Infinite Fun
58mKeynote: Why Functional Programming Matters