Tweetovi

Blokirali ste korisnika/cu @seagreen__

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @seagreen__

  1. Prikvačeni tweet
    6. pro 2019.

    Into Haskell, Unison, and getting people's data back onto their own computers (Perkeep, Sandstorm, etc). Writing a book about making a simple compiler. Open invitation: if you're interested in any of those or are trying to learn functional programming feel free to message me.

    Prikaži ovu nit
    Poništi
  2. proslijedio/la je Tweet
    31. sij

    Sorry community, I feel like I have failed you. My team will be discussing which tools they want to migrate to instead over the next few weeks, minus my highly opinionated position on it all 😢

    Prikaži ovu nit
    Poništi
  3. 30. sij

    The only three valid "protocols" for dealing with an MVar are take/put, put/take, and some threads taking and some putting, right?

    Poništi
  4. proslijedio/la je Tweet
    29. sij
    Odgovor korisniku/ci

    Yep, we used it in a code base I worked on a few years ago. In my experience when I need channels for inter-thread communication I never need bi-directional, I usually have some producers and some consumers, but never both in single thread. This lib helps in those cases.

    Poništi
  5. 29. sij

    More generally, I love the idea of main being a one-liner that summarizes the design of the program.

    Poništi
  6. 22. sij

    I'd love to be proven wrong. Also, re this line: > The language that GHC supports is incredibly complex. GHC 8.6.3 man page lists 115 language pragmas. The man page is missing pragmas. I don't think there's a complete list of pragmas on the web.

    Prikaži ovu nit
    Poništi
  7. 22. sij

    Fantastic post, I agree with everything in here, except the hopefulness. I think the community already has principles to guide the language. These principles are: cool features are great, feature flags are great, defaults don't matter much.

    Prikaži ovu nit
    Poništi
  8. 14. sij

    3 years ago I made a prototype of it: I can't have been the first though, does anyone know of any academic work or other projects in that area?

    Prikaži ovu nit
    Poništi
  9. 14. sij

    Anyone have ideas you toss around in your head for years and could use other people's thoughts on? My current one is a schema system for JSON based on algebraic types. It seems like a really good idea, but doesn't get much attention.

    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    6. sij
    Odgovor korisnicima

    In comparison with Elm and React, Concur makes explicit the previously behind the scenes event loop. And gets rid of the mutation (setState). This allows using different event loop abstractions for different parts of the UI. This is powerful. Recovering TEA is a simple one liner.

    Poništi
  11. proslijedio/la je Tweet
    6. sij
    Odgovor korisniku/ci

    I wholeheartedly agree - I think the Concur model is honestly one of the most impactful developments in UIs since Elm/React. Shoutout to who came up with it.

    Poništi
  12. 5. sij

    I was a diehard Model/View/Update fan until I tried this. I've written two MVU games, one in Elm () and one in Haskell (). concur-replica seems like an INSANE improvement and I'm looking forward to trying something bigger with it.

    Prikaži ovu nit
    Poništi
  13. 5. sij

    /cc who paired during TS work on an upstream dep, who I've mentioned this to already, and who wrote concur-replica

    Prikaži ovu nit
    Poništi
  14. 5. sij

    Shipped a little concur-replica example app: Slightly tongue-in-cheek, but I'm pretty happy with the problem statement:

    Prikaži ovu nit
    Poništi
  15. proslijedio/la je Tweet
    24. pro 2019.

    Anyone working on any neat side projects over the holidays?

    Poništi
  16. proslijedio/la je Tweet
    28. lis 2019.
    Odgovor korisniku/ci

    Thank you :) It's interesting how the essence of various build systems shows up in the languages they speak. While we expected to see the usual Applicative/Monad examples, it was a complete surprise for us to find that pretty much every choice of `c` gives an existing system!

    Poništi
  17. proslijedio/la je Tweet
    28. lis 2019.

    I love charts like this; an excerpt from Build Systems a'la Carte at . A great denotational design breakdown of build-systems! This slide shows how constraining effect systems results in different build-system models! Fantastic talk !

    Poništi
  18. 19. pro 2019.

    🎁First holiday project shipped🎁 Introducing Acme PCF Specification: "RealWorld[1] but for a small language." [1]

    Poništi
  19. 18. pro 2019.

    if case "foo" of "bar" -> True _ -> False then 5 else 3 This isn't even with BlockArguments. I didn't understand haskell's syntax well.

    Poništi
  20. 18. pro 2019.

    One liner: negate let a = 5 in a

    Prikaži ovu nit
    Poništi
  21. 18. pro 2019.

    Didn't realize how much BlockArguments changed, eg it makes this valid: bar :: Int bar = negate case () of () -> 5 Details:

    Prikaži ovu nit
    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·