49 pointsby ofalkaed14 days ago8 comments
  • ofalkaed14 days ago
    The code and details about various versions are available on the homepage which weirdly does not really explain the language.

    http://www.nsl.com/k/xy/xy.htm

    • jasonwatkinspdx14 days ago
      So this is the blog of Stevan Apter. As you might guess from having a 3 letter vanity domain, he and this website have been around a long long time. It's mostly code golf style stuff posted to the J Software and Concatenative Languages lists. So yeah it'll be terse and obscure outside that community.

      The domain name stands for No Stinkin' Loops, a reference to how APL style languages lift iteration to combinator expressions.

      • ofalkaed14 days ago
        I did not know any of that, did not even consider if it was a vanity domain but did briefly wonder if nsl was the domain for some small country. I think the homepage does a good job of covering everything we need to know and in someways does a better job, but this posting provides a better overview for a HN submission. I included the homepage since some would probably skim the post or only read as much as they need to get the idea and miss the note at the bottom, someone would have probably posted asking where the code was.

        Thanks for the background info.

        • jasonwatkinspdx12 days ago
          Yeah, I stumbled onto J et all after seeing an interesting entry into the ICFP programming contest. So I was on the concatenative languages email list for a bit and knew nsl.com from there.

          I've always found this snippet interesting: https://nsl.com/papers/kisntlisp.htm

          Comparing the lisp to the non linenoise k syntax you can see there really is something compelling to this sort of lifting of iterator at a time fiddling to a higher level of abstraction where you just think about the shape of what you want.

    • nitrix14 days ago
      I’m the Alex mentioned in the acknowledgments, feel free to ask your questions.
      • ofalkaed13 days ago
        In what way did you carry the ideas forward? Did you create a language which builds on it or did you just help Stevan develop the ideas of XY? Curious in anything you can contribute.
    • mncharity14 days ago
      Including discussion of XY 2.0 and its flatness (all partitions of a sequence of program tokens are semantically equivalent).
    • gnabgib14 days ago
      (2004)[0] And yet you submitted it twice in 5 months? weirdly indeed.

      https://news.ycombinator.com/item?id=44880307

      [0]: https://news.ycombinator.com/item?id=9216548

      • ofalkaed13 days ago
        It is the same topic but not the same submission, if topics can't be repeated HN will die in short order. Are you suggesting that I am trying to revive my 20 year dead language by posting it twice within 5 months? my posting history should prove that I am not capable of creating such a language.

        Not criticizing you, you perform a great service to HN, I am just confused by your post and that might just be a failing of mine. Either way, thanks for all you do.

  • tobr14 days ago
    Submitted headline is missing an ’en’.
    • ofalkaed14 days ago
      Something about that looked wrong but I just could not quite see it. To late to edit.
    • mncharity14 days ago
      "The Concatenative Language XY" discussion (typo in title).

      For search.

  • volemo14 days ago
    This helped me finally grok continuations!
  • jaberjaber2314 days ago
    interesting mix of k and joy. the queue manipulation primitives like -> and => have no equivalent in joy, lets you do things like call/cc in a few lines
  • mncharity14 days ago
    Fwiw, bing/chat identifies the implementation K version as a K2, in a style which might be an easy port to K3.
  • wvlia514 days ago
    I'm making a new language inspired in XY
  • zabzonk14 days ago
    So, basically a less-readable Forth?
  • wosined14 days ago
    Interesting. But looks like assembler and more complicated.