miniblog.

← Back to all posts

May 5, 2018 at 11:27

Superb article on type systems, open vs closed world soundness, and type systems intentionally not offering soundness:

Are unsound type systems wrong?

Jan 26, 2020 at 16:29

Back from #POPL! The Saturday tracks were well worth attending. Ronald's discussion of gradual type safety vs soundness was my highlight: https://popl20.sigplan.org/details/wgt-2020-papers/5/Gradual-Typing-as-if-Types-Mattered When talking about soundness, it's important to say exactly what issues you're handling, and the talk explored this.

Gradual Typing as if Types Mattered (WGT 2020 - : First ACM SIGPLAN Workshop on Gradual Typing) - POPL 2020

The Workshop on Gradual Typing (WGT) aims at disseminating the latest results on the integration of compile-time and run-time checking of program invariants, such as the integration of static and dynamic type checking. The workshop serves as an incubator for ideas, open problems, and manuscripts: it is a place where the community can meet, discuss, and give each other constructive feedback. The workshop will encourage participation from researchers in both academia and industry, drawing people from the many active projects on both sides of the aisle. Format: A one-day workshop with 1-2 ...

Oct 21, 2017 at 19:34

Brilliant, practical talk on implementing a JS type checker: https://youtu.be/MuC8I1JBKv0 (including soundness tradeoffs and flow-based analysis)

Aug 23, 2015 at 14:45

Rust nightly is now warning about upcoming changes to the type checker. There are some (obscure) soundness holes:

rfcs/text/0000-projections-lifetimes-and-wf.md at projection-and-lifetimes · nikomatsakis/rfcs

rfcs/text/0000-projections-lifetimes-and-wf.md at projection-and-lifetimes · nikomatsakis/rfcs

RFCs for changes to Rust. Contribute to nikomatsakis/rfcs development by creating an account on GitHub.