> be NASA programmer.> excited to write Lisp.> code lookin like hieroglyphs.> boss walks by.> "wtf is that, alien code?".> no, just Lisp boss.> boss laughs, Lisp now forbidden.> mfw no cosmic parens.> mission to Mars now in JavaScript.> planet doomed.> feelsbadman.jpg.

a direct 1:1 port of this to haskell would clock in at 40 billion lines of code, some say it can't even be done with current technology.

I still have my Model 100 in the hard plastic briefcase. The foam has disintegrated and my tapes no longer read. 😔.

I wrote an arbitrary precision integer arithmetic library for this to compute pi. I wrote the multiplication routine when my dad was driving us to Taco Bell for lunch one summer. The bigints were stored as strings and I used STRING$, CHR$, STR$, and MID$ to manipulate them.

If your programming language website doesn't have a page like this, it's ngmi.

(i should have added a word to the tune of "eventually").

PSA: O(100) was never intended to be synonymous with ≈100 (and in rigorous math, they aren't synonymous).

PSA: O(1) does not mean. - approximately 1.- a constant function/a function whose value is independent of its input.- an efficient function. O(1) just says the absolute value of a function is never greater than _some_ constant. A good example to remember: sin(x) is in O(1).

More info:

Coalton now has a `derive` mechanism to automatically derive type class instances for your own data types. If you're adventurous enough you can even write your own derivers. More info 👇. (h/t @garlic0ne)

i just got tickets to 2 bach keyboard recitals 2 days in a row in 2 different states. not sure this was a good idea.

RT @JustDeezGuy: @run_guns @stylewarning Haskell and Common Lisp are byzantine lambda calculus wrappers, with the latter ram-rodding its wa….

Yale Haskell is kind of funny. They implement Haskell in a pseudo-Scheme, and the pseudo-Scheme is implemented as Common Lisp macros.

(it's impossible because it's such a good language).

Haskell people: Why is Foldable on * -> * preferable over, say, Foldable on a container (*) and eltType (*) (with a functional dependency)? Does this just mean non-parametric containers in practice don't get to be used for folding?.

I think there should be a computer language benchmark game, but adjudicated by people who have exceptionally good taste in what constitutes a reasonable entry. A separate category of "pull all the stops" would be fine.

If you're interested:

Wrote a short "glossary" on Coalton's naming conventions.

Long overdue, but we finally merged a PR someone submitted allowing constraints on type class methods in Coalton. It only took 6 months for us to get to. bless the hearts of open-source contributors' patience.

Exceptions and resumptions have landed in Coalton. After catching an exception that has bubbled up the stack, you can resume execution at demarcated points back down the stack.

(Bourbaki doesn't count.).