yarn

twtxt.net

↳ In-reply-to » It is such a nice feeling that Mu is such a capable little language 😅 And I decided to write code code in Mu by hand 🤚 haha 🤣 and start solving Project Euler problems, like Problem 8 which works out to be a nice elegant solution in Mu:

And just like that Problem 10 is done and correct whoohoo 🥳 It was easy because in Problem 7 I’d already written an iterator to produce infinite primes. So the solution for finding the sum of primes under 2,000,000 is basically (shortened):

  primes := iter.take_while(iter.filter(prime_candidates(), is_prime), fn(p) { p < n })
  print(iter.sum(primes))

And of course the answer is: 142913828922 which took ~21.ss for the Go Vm to compuete.

⤋ Read More

prologic

twtxt.net

Fri, Aug 14 10:30PM (3w ago)

↳ In-reply-to » It is such a nice feeling that Mu is such a capable little language 😅 And I decided to write code code in Mu by hand 🤚 haha 🤣 and start solving Project Euler problems, like Problem 8 which works out to be a nice elegant solution in Mu:

Oh man wow 😮 Problem 9 was quite hard 😱 I had to build two new functions in the Mu stdlib for computing combinations and permutations, but then the combinations of range(1000) for triples such as a + b == c is enormous! So i had to write iterator versions of these to do lazy evaluation. Anyway solution follows:

#!/usr/bin/env mu

// Special Pythagorean Triplet

import "iter"

fn usage() {
  print("Usage:", args()[0], "<n>")
}

fn sqr(x) { x * x }

fn main() {
  if len(args()) < 2 {
    usage()
    exit(1)
  }

  n := must(int(args()[1]))
  print("n:", n)

  // For a < b < c and a + b + c == n, both a and b are strictly less
  // than n/2. Generate only (a,b) combinations and derive c directly. This
  // keeps the search lazy and reduces n=1000 from C(999,3) = 165,668,499
  // candidate triples to C(499,2) = 124,251 candidate pairs.
  pairs := iter.combinations(iter.range(1, n / 2), 2)

  triples := iter.map(pairs, fn(xs) {
    a := xs[0]
    b := xs[1]
    return [a, b, n - a - b]
  })

  // Enforce b < c; a < b is already guaranteed by combinations over an
  // increasing range, and a + b + c == n holds by construction.
  triples = iter.filter(triples, fn(xs) {
    return xs[1] < xs[2]
  })

  // Euler 9 has one answer for n=1000. find() stops the entire upstream
  // iterator chain as soon as the first Pythagorean triple is found.
  answer := iter.find(triples, fn(xs) {
    return sqr(xs[0]) + sqr(xs[1]) == sqr(xs[2])
  })

  print(answer)

  if answer != nil {
    print(answer[0] * answer[1] * answer[2])
  }
}

main()

⤋ Read More

thecanine

twtxt.net

Tue, Aug 11 8:32AM (4w ago)

↳ In-reply-to » The author of the Xfce Wayland compositor on LLMs:

Most of this insane power consumption happens during the models training, so the additional cost of prompting, is minimal. Facebook invented cool ways to exted the environmental damage far beyond the training stage, by dumping the dirty data center water into nearby water sources, but that’s more so a Facebook problem, than a fault of AI.

I would still never waste my own money, paying for any AI subscriptions that my employer doesn’t already pay for, no matter if we use them, or not.

I am also against aggressive scraping and taking everything as training data, disregarding the authors wishes and license. I allow it on my website, mostly as proof that no matter how much they scrape my stuff, the real me remains both a superior webdev and artist.

I am well aware this makes me come across really egotistical, especially after developers far better than me, embraced vibe coding, arguably way too much. To give an example, all attempts to get Microsoft Copilot, that’s embedded into SharePoint, to generate an extension/uBlock filter to make SharePoint revert to the old superior UI failed, but I was able to do that in three lines of code ( https://thecanine.smol.pub/ublock-filters ), in half an hour, despite me never seeing the code of that page before. After a few of these comparisons, it is increasingly hard to see most of these tools, as my replacement.

I am not even going to talk about the people trying to imitate my art style with AI, until they can show me a result where all the squares are the same size, in a grid and actually square.

⤋ Read More

movq

www.uninformativ.de

Sun, Jun 21 2:42AM (11w ago)

↳ In-reply-to » For this week's (slightly delayed) #caturday post, we catch Bowie as he contemplates whether to play Arkham City (again).

@itsericwoodward@itsericwoodward.com (I still hate that these photos look so good, despite being made with a telephone. 😝 I mean, the (probably built-in) filters are easy to spot, but it looks super convincing when you don’t pay attention.)

⤋ Read More

Searching yarn