{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreiftrgy3wtmuuirrhlpr2gkwp4are33jsmj7logcn77sybu4ift37y",
    "uri": "at://did:plc:pi6woz4d47bkuws673w2il2r/app.bsky.feed.post/3mex5tofiyam2"
  },
  "path": "/t/theseus-worry-free-algebraic-and-higher-order-effects/13563#post_10",
  "publishedAt": "2026-02-14T12:06:06.000Z",
  "site": "https://discourse.haskell.org",
  "textContent": "Thank you for citing Heftia. I am its author, and it is encouraging to see more people working in the space of effect libraries for Haskell. I have not been able to spend any time on the library for a while, but I hope to resume work again soon.\n\nI have not had the chance to read Theseus carefully yet, but I think it is a good approach to start from a freer-based implementation to refine the interface and semantics, and then later rewrite the internals using GHC primops while preserving those guarantees. In fact, that is broadly the direction I have been taking as well, and after running some experiments I have become more optimistic than I initially was about how difficult that path is. I hope to be able to share results in the near future.\n\nThe possibility of a sound semantics for combining higher-order effects, especially scoping for resource management like `bracket` or `withFile`, with coroutines (non-scoped resumption) has been a long-standing mystery. I find it extremely interesting, but because it is so difficult I have paused work on it for now and am focusing on the performance work mentioned above, namely the migration from freer to primops. So I am very happy to see you exploring this area.",
  "title": "Theseus: Worry free algebraic and higher order effects"
}