Daniel Spiewak

@djspiewak

It's been done before. The mathematicians complain.

Boulder, CO
Vrijeme pridruživanja: lipanj 2007.

Tweetovi

Blokirali ste korisnika/cu @djspiewak

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

  1. proslijedio/la je Tweet
    3. velj

    I continue to be asked about Techempower Benchmarks. A diatribe: I don't maintain them for my projects or recommend looking at them. Some projects optimize for a high score, and some demonstrate idiomatic usage. Either may be insightful, but the mix is not.

    Prikaži ovu nit
    Poništi
  2. 31. sij

    Cats Effect 2.1.0 has been released with a ton of new goodies! Many, many, many thanks to each and every contributor.

    Poništi
  3. 28. sij

    WTF this project contains *semantically significant source files* which exploit upper/lower-case name collision?! In other words, it literally cannot be built on OSes with case sensitive filesystems.

    Poništi
  4. 27. sij

    I identify so much with this.

    Poništi
  5. 27. sij

    Spoiler: it probably won't.

    Poništi
  6. proslijedio/la je Tweet
    26. sij

    Would it be fair to say that `traverse` discards information? The input is a known structure (spine) with unknown leaves. The output has both unknown spine and leaves.

    Prikaži ovu nit
    Poništi
  7. 26. sij

    "Oh I know! I'll just crack open my editor and pick up where I left off on debugging this stateful race condition…" *some time passes* "I hate me of a week ago. IS IT SO HARD TO LEAVE A DAMN COMMENT EXPLAINING THIS STUFF?!?!"

    Poništi
  8. proslijedio/la je Tweet
    19. sij

    I’m pleased to present the flame-bait sequel to my previous blog post: “No, dynamic type systems are not inherently more open.” (Please send me your hatemail at your earliest convenience.)

    Poništi
  9. proslijedio/la je Tweet
    14. sij

    Created a handle for The Scala Logs ( ) for those interested in a podcast.

    Poništi
  10. proslijedio/la je Tweet
    13. sij

    on Catalina: If you are having issues with "cannot be opened because the developer cannot be verified." ->

    Poništi
  11. 12. sij

    IMO, these ApplicativeError laws have been missing: Affects both Haskell and Scala, though afaict, Haskell's various implementations of the class are already (almost coincidentally) correct.

    Poništi
  12. proslijedio/la je Tweet
    10. sij

    "the ship is sinking around us, we're about to die, and we can't even reliably get the chairs to stick in the right place since the deck may (or may not) be tilted off to the side" Never been much for action movies, but who needs them when you have 's code reviews?

    Poništi
  13. 9. sij

    "a profunctor over A and B and HFunctor over F, well, it's not actually profunctor but Arrow, Strong, Costrong and other funny profunctor children" Ooooookay, SlamData engineering Discord. Calm down.

    Poništi
  14. 7. sij

    Random nugget of insight from found while perusing unrelated threads: Source compatibility is broken in Scala when you *add a public method to a type*. Even ignoring diamonds, this alone is murder on source dependency schemes in this language.

    Poništi
  15. 6. sij

    Update: it's actually FreeT that's violating a slightly different law, not Kleisli.

    Prikaži ovu nit
    Poništi
  16. 6. sij

    A minimized reproducer! .Kleisli currently violates the monad laws:

    Prikaži ovu nit
    Poništi
  17. 5. sij
    Prikaži ovu nit
    Poništi
  18. 5. sij

    Follow up: it appears that Kleisli or FreeT's map function is… oddly broken?

    Prikaži ovu nit
    Poništi
  19. 4. sij
    Poništi
  20. proslijedio/la je Tweet
    4. sij

    Wrote an answer to the question “What are algebraic effects?” from a Haskell perspective in a reddit comment, for anyone interested:

    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:

    ·