raganwald.com
LightNews LightNews
banner
raganwald.functional.cafe.ap.brid.gy
raganwald.com
@raganwald.functional.cafe.ap.brid.gy
8 followers 0 following 70 posts
Reg "raganwald" Braithwaite's account for discussing functional programming and shamelessly plugging essays about programming.

Follows are NOT endorsements. This […]

🌉 bridged from ⁂ https://functional.cafe/@raganwald, follow @ap.brid.gy to interact
Posts Replies Media Videos
raganwald.functional.cafe.ap.brid.gy
JavaScript Allongé was written in this converted semi-detached cottage.
The living area of a small cottage that is heated by a stove.
January 16, 2026 at 3:55 PM
raganwald.functional.cafe.ap.brid.gy
Life, coffee, and blueberry pie.

https://conwaylife.com/book/
A book about Conway's Game of Life open to a new chapter about spaceships, a Mondrian cup of espresso, and a mostly eaten blueberry pie in a foil baking pan.
January 16, 2026 at 2:55 PM
raganwald.functional.cafe.ap.brid.gy
Although many people say that JavaScript Allongé is a book about functional programming, it’s actually a book about how functions as first-class entities facilitate composition.

So... Although Kleisli arrows allow for the composition of functions that work […]

[Original post on functional.cafe]
A man draws a green compound bow, overlaid with the question “Allonge?” (sic).
December 5, 2025 at 2:43 PM
raganwald.functional.cafe.ap.brid.gy
A personal favourite. And yes there *is* a Mister Braithwaite in “Enter the Dragon!”

https://raganwald.com/2011/11/01/williams-master-of-the-comefrom.html
A light-skinned black man with a full afro and mutton chops loos to our left, his right
December 3, 2025 at 3:03 AM
raganwald.functional.cafe.ap.brid.gy
At the turn of the century, OG blogger Joel Spolsky posited the "Joel Test," twelve simple yes/no questions that defined a reasonable-by-the-standards-of-the-day local optimum for shipping software. Some seem ridiculously obvious today, but weren't standard […]

[Original post on functional.cafe]
Do you use source control? Can you make a build in one step? Do you make daily builds? Do you have a bug database? Do you fix bugs before writing new code? Do you have an up-to-date schedule? Do you have a spec? Do programmers have quiet working conditions? Do you use the best tools money can buy? Do you have testers? Do new candidates write code during their interview? Do you do hallway usability testing?
November 24, 2025 at 3:52 PM
raganwald.functional.cafe.ap.brid.gy
source: @RandomHourlyDuckBot
A duckling in mid-squawk is overlaid with the words, "A either end of the strong typing scale, there lies a hacker class."
November 15, 2025 at 1:59 PM
raganwald.functional.cafe.ap.brid.gy
A duck-shaped novelty bottle of whiskey is overlaid with the words, "When you're days into refactoring a duck-typed codebase"
November 9, 2025 at 2:08 PM
raganwald.functional.cafe.ap.brid.gy
A Brutal Look at Balanced Parentheses, Computing Machines, and Pushdown Automata

https://raganwald.com/2019/02/14/i-love-programming-and-programmers.html
November 8, 2025 at 7:52 PM
raganwald.functional.cafe.ap.brid.gy
Forget algebraic types, the Type Lords use Gallifrean types.
A young woman looks at words written underneath old, peeling wallpaper. Overlaid are the words "Blink, and the weeping angels of strong typing will feed on your time. Ok, and duck type! Really duck type!"
November 8, 2025 at 5:11 PM
raganwald.functional.cafe.ap.brid.gy
There are two ways to write production #typescript. One is to use such advanced type features that if it compiles at all, you can be sure there are no obvious bugs. The other is to write such straightforward, extremely simple types that there are obviously no bugs.

Only a third method works.
November 6, 2025 at 2:29 PM
raganwald.functional.cafe.ap.brid.gy
Source: @RandomHourlyDuckBot
A pile of yellow and brown ducklings, overlaid with: There is no one right kind of duck-typed programming language. postscript: If concatinative languages Factor into your choice, then go FORTH and seek Joy.
November 5, 2025 at 3:52 PM
raganwald.functional.cafe.ap.brid.gy
Since 1951.
A cartoon rabbit and a cartoon duck stand on either side of a tree trunk. A sign says that it is DUCK SEASON, and someone has pasted the word "typing" between the two words.
November 4, 2025 at 7:17 PM
raganwald.functional.cafe.ap.brid.gy
Source: @RandomHourlyDuckBot
A duck stands on a city sidewalk wearing a green vest and a cardboard sign that says FREE HGS. Arrows point to the duck's spread wings. Overlaid are the words, "When you want to Keep It Simple & Ship, duck typing feels like a free hug."
November 3, 2025 at 8:08 PM
raganwald.functional.cafe.ap.brid.gy
JavaScript Allongé follows this pattern:

https://raganwald.com/assets/books/javascriptallongesix.pdf

The very first version was written before ES2015. Its purpose was to discuss programming with first-class functions and composition. To do that, it had to […]

[Original post on functional.cafe]
Cover of "JavaScript Allongé" featuring a small white dairy jug, a white espresso cup on a black saucer, and a spoon. Cover of "JavaScript Allongé, The Six Edition" featuring a small white dairy jug, a white espresso cup on a black saucer, and a spoon.
October 27, 2025 at 3:14 PM
raganwald.functional.cafe.ap.brid.gy
A long time ago, in a ruby-encrusted galaxy far, far away...

https://raganwald.com/assets/books/kestrels-quirky-birds-and-hopeless-egocentricity.pdf
A brass-toned mechanism featuring concentric gears is overlaid with the words "Kestrels, Quirky Birds, & Hopeless Egocentricity"
August 13, 2025 at 2:46 PM
raganwald.functional.cafe.ap.brid.gy
My first type-level TypeScript Y Combinator good enough to criticize:

https://gist.github.com/raganwald/6558ddfd76b16fe7b7213a80f54576e0
// ---------- The Why Bird as an HKT ---------- // // This is a nearly direct translation of a JS // implementation from "Why Y? Deriving the // Y Combinator in JavaScript:" // // https://raganwald.com/2018/09/10/why-y.html // // const Y = // fn => // M(m => a => fn(m(m))(a)); // A kind that encapsulates the "maker" function within the Why Brd type _$Maker<FN extends HigherOrderKind, M extends HigherOrderKind, A> = $< FN,_$cast<_$M<M>, _$inputOf<FN>> > extends infer InnerMaker extends Kind.Kind ? $< InnerMaker, _$cast<A, _$inputOf<InnerMaker>> > : never; interface Maker_T<FN extends HigherOrderKind, M extends HigherOrderKind> extends Kind.Kind { f( a: this[Kind._] ): _$Maker<FN, M, typeof a> } interface Maker<FN extends HigherOrderKind> extends Kind.Kind { f( m: _$cast<this[Kind._], HigherOrderKind> ): Maker_T<FN, typeof m> } // The Why Bird composes a decoupled recursive kind with itself, // converting a decoupled recursive kind to a recurive kind interface Y extends Kind.Kind { f( fn: _$cast<this[Kind._], Kind.Kind> ): _$M<Maker<typeof fn>> }
August 4, 2025 at 4:17 PM
raganwald.functional.cafe.ap.brid.gy
For the tenth time: https://en.wikipedia.org/wiki/Greenspun%27s_tenth_rule
A man with a coffee cup sits at a table with a banner reading, "All programming is either Lisp or Greenspunning, change my mind."
August 4, 2025 at 3:42 PM
raganwald.functional.cafe.ap.brid.gy
A type-level TypeScript implementation of the Mockingbird. The Why Bird and Jackson’s Widowbird are next...
type M<K extends Kind.Kind> = $<K, Type._$cast<K, _$inputOf<K>>>; type LastButLast<S extends string, AccumulatedButLast extends string> = $N<M<LastButLastKind>, [S, AccumulatedButLast, NewCounter]> extends infer Result ? Result extends { type: ContinuationType, parameters: [infer NewS extends string, infer NewAccumulatedButLast extends string] } ? LastButLast<NewS, NewAccumulatedButLast> : Result : never; type _$LastButLast<K extends LastButLastKind, S extends string, AccumulatedButLast extends string, RecursionCount extends Counter> = S extends `${infer Head}${infer ButHead}` ? ButHead extends "" ? [Head, AccumulatedButLast] : Count<RecursionCount> extends RecursionLimit ? ContinueWith<[S, AccumulatedButLast]> : $N<M<LastButLastKind>, [ButHead, `${AccumulatedButLast}${Head}`, Increment<RecursionCount>]> : ["", ""]
July 30, 2025 at 3:32 PM
raganwald.functional.cafe.ap.brid.gy
“[Structure and Interpretation of Computer Programs] wasn't really about Scheme; I think a lot of the point of using Scheme (as I understood it) is that the language is so minimalist and so beautiful that even this first intro course can be about fundamental […]

[Original post on functional.cafe]
Detail from the cover of "Structure and Interpretation of Computer Programs" by Abelson and Sussman
July 26, 2025 at 2:57 PM
raganwald.functional.cafe.ap.brid.gy
The SNOBOL4 Programming Language, also known by its fans as “The Green Book:”

https://raganwald.com/assets/books/snobol-green-book.pdf
The cover of “The SNOBOL4 Programming Language, second edition.” It is green, and there are no illustrations.
July 15, 2025 at 1:52 PM
raganwald.functional.cafe.ap.brid.gy
“...We’re done. And this is a win over the typical recursive or even iterative solution for large numbers, because while each operation is more expensive, we only perform log2n operations.”

—An ES6 function to compute the nth Fibonacci number (2015) […]

[Original post on functional.cafe]
A nautalis-type spiral created with "whirling squares" of sizes equal to the fibonacci sequence
June 23, 2025 at 4:35 PM
raganwald.functional.cafe.ap.brid.gy
In 1972, John Horton Conway invented the FRACTRAN programming language to show that a generalization of the Collatz problem is algorithmically undecidable. He showed that this generalized form is equivalent to FRACTRAN programs, and also that FRACTRAN […]

[Original post on functional.cafe]
The fractions (7/3, 99,98, 13/49, 36/91, 10/143, 49/13, 7/11, 1/2, 91/1) are displayed as a comma-separated list in parentheses.
June 19, 2025 at 12:55 PM
raganwald.functional.cafe.ap.brid.gy
The SNOBOL4 Programming Language, also known by its fans as, “The Green Book:”

https://raganwald.com/assets/books/snobol-green-book.pdf
The cover of “The SNOBOL4 Programming Language, second edition.” It is green, and there are no illustrations.
June 9, 2025 at 3:21 PM
raganwald.functional.cafe.ap.brid.gy
In 2004, Matthew Cook proved that the Rule 110 elementary cellular automaton is universal. His construction began with the existing proof that Post Tag Systems are universal, then introduced a variation, the Cyclic Tag System, which he showed is equivalent to […]

[Original post on functional.cafe]
Figure 2 from Cook's paper shows a diagram of a glider system in Rule 110 emulating a cyclic tag system which has a list of two appendants: YYY and N.
June 6, 2025 at 12:49 PM
raganwald.functional.cafe.ap.brid.gy
“DNA does not have one strand for animals, another for cats, and a third for jaguars. It does not compose these specializations together to produce the final specimen individual. DNA is a chaotic, unstructured mess. Einstein once said that “God does not play […]

[Original post on functional.cafe]
another way forward There is another way forward. What if objects exist as encapsulations, and the communicate via messages? What if code re-use has nothing to do with inheritance, but uses composition, delegation, even old-fashioned helper objects or any technique the programmer deems fit? The ontology does not go away, but it is decoupled from the implementation. Here’s how.
May 31, 2025 at 11:31 AM