Full Width 
Shortcuts 
Sign Up 
Log In 
Articles 
Reading List More from dthompson
I'm happy to announce that guile-websocket 0.2.0 has been released! Guile-websocket is an implementation of the WebSocket protocol, both the client and server sides, for Guile Scheme. This release introduces breaking changes that overhaul the client and server implementations in order to support non-blocking I/O and TLS encrypted connections. source tarball: https://files.dthompson.us/guile-websocket/guile-websocket-0.2.0.tar.gz signature: https://files.dthompson.us/guile-websocket/guile-websocket-0.2.0.tar.gz.asc See the guile-websocket project page for more information. Bug reports, bug fixes, feature requests, and patches are welcomed.
Wasm GC is a wonderful thing that is now available in all major web browsers since slowpoke Safari/WebKit finally shipped it in December. It provides a hierarchy of heap allocated reference types and a set of instructions to operate on them. Wasm GC enables managed memory languages to take advantage of the advanced garbage collectors inside web browser engines. It’s now possible to implement a managed memory language without having to ship a GC inside the binary. The result is smaller binaries, better performance, and better integration with the host runtime. However, Wasm GC has some serious drawbacks when compared to linear memory. I enjoy playing around with realtime graphics programming in my free time, but I was disappointed to discover that Wasm GC just isn’t a good fit for that right now. I decided to write this post because I’d like to see Wasm GC on more or less equal footing with linear memory when it comes to binary data manipulation. Hello triangle For starters, let's take a look at what a “hello triangle” WebGL demo looks like with Wasm GC. I’ll use Hoot, the Scheme to Wasm compiler that I work on, to build it. Below is a Scheme program that declares imports for the subset of the WebGL, HTML5 Canvas, etc. APIs that are necessary and then renders a single triangle: (use-modules (hoot ffi)) ;; Document (define-foreign get-element-by-id "document" "getElementById" (ref string) -> (ref null extern)) ;; Element (define-foreign element-width "element" "width" (ref extern) -> i32) (define-foreign element-height "element" "height" (ref extern) -> i32) ;; Canvas (define-foreign get-canvas-context "canvas" "getContext" (ref extern) (ref string) -> (ref null extern)) ;; WebGL (define GL_VERTEX_SHADER 35633) (define GL_FRAGMENT_SHADER 35632) (define GL_COMPILE_STATUS 35713) (define GL_LINK_STATUS 35714) (define GL_ARRAY_BUFFER 34962) (define GL_STATIC_DRAW 35044) (define GL_COLOR_BUFFER_BIT 16384) (define GL_TRIANGLES 4) (define GL_FLOAT 5126) (define-foreign gl-create-shader "gl" "createShader" (ref extern) i32 -> (ref extern)) (define-foreign gl-delete-shader "gl" "deleteShader" (ref extern) (ref extern) -> none) (define-foreign gl-shader-source "gl" "shaderSource" (ref extern) (ref extern) (ref string) -> none) (define-foreign gl-compile-shader "gl" "compileShader" (ref extern) (ref extern) -> none) (define-foreign gl-get-shader-parameter "gl" "getShaderParameter" (ref extern) (ref extern) i32 -> i32) (define-foreign gl-get-shader-info-log "gl" "getShaderInfoLog" (ref extern) (ref extern) -> (ref string)) (define-foreign gl-create-program "gl" "createProgram" (ref extern) -> (ref extern)) (define-foreign gl-delete-program "gl" "deleteProgram" (ref extern) (ref extern) -> none) (define-foreign gl-attach-shader "gl" "attachShader" (ref extern) (ref extern) (ref extern) -> none) (define-foreign gl-link-program "gl" "linkProgram" (ref extern) (ref extern) -> none) (define-foreign gl-use-program "gl" "useProgram" (ref extern) (ref extern) -> none) (define-foreign gl-get-program-parameter "gl" "getProgramParameter" (ref extern) (ref extern) i32 -> i32) (define-foreign gl-get-program-info-log "gl" "getProgramInfoLog" (ref extern) (ref extern) -> (ref string)) (define-foreign gl-create-buffer "gl" "createBuffer" (ref extern) -> (ref extern)) (define-foreign gl-delete-buffer "gl" "deleteBuffer" (ref extern) (ref extern) -> (ref extern)) (define-foreign gl-bind-buffer "gl" "bindBuffer" (ref extern) i32 (ref extern) -> none) (define-foreign gl-buffer-data "gl" "bufferData" (ref extern) i32 (ref eq) i32 -> none) (define-foreign gl-enable-vertex-attrib-array "gl" "enableVertexAttribArray" (ref extern) i32 -> none) (define-foreign gl-vertex-attrib-pointer "gl" "vertexAttribPointer" (ref extern) i32 i32 i32 i32 i32 i32 -> none) (define-foreign gl-draw-arrays "gl" "drawArrays" (ref extern) i32 i32 i32 -> none) (define-foreign gl-viewport "gl" "viewport" (ref extern) i32 i32 i32 i32 -> none) (define-foreign gl-clear-color "gl" "clearColor" (ref extern) f64 f64 f64 f64 -> none) (define-foreign gl-clear "gl" "clear" (ref extern) i32 -> none) (define (compile-shader gl type source) (let ((shader (gl-create-shader gl type))) (gl-shader-source gl shader source) (gl-compile-shader gl shader) (unless (= (gl-get-shader-parameter gl shader GL_COMPILE_STATUS) 1) (let ((info (gl-get-shader-info-log gl shader))) (gl-delete-shader gl shader) (error "shader compilation failed" info))) shader)) (define (link-shader gl vertex-shader fragment-shader) (let ((program (gl-create-program gl))) (gl-attach-shader gl program vertex-shader) (gl-attach-shader gl program fragment-shader) (gl-link-program gl program) (unless (= (gl-get-program-parameter gl program GL_LINK_STATUS) 1) (let ((info (gl-get-program-info-log gl program))) (gl-delete-program gl program) (error "program linking failed" info))) program)) ;; Setup GL context (define canvas (get-element-by-id "canvas")) (define gl (get-canvas-context canvas "webgl")) (when (external-null? gl) (error "unable to create WebGL context")) ;; Compile shader (define vertex-shader-source "attribute vec2 position; attribute vec3 color; varying vec3 fragColor; void main() { gl_Position = vec4(position, 0.0, 1.0); fragColor = color; }") (define fragment-shader-source "precision mediump float; varying vec3 fragColor; void main() { gl_FragColor = vec4(fragColor, 1); }") (define vertex-shader (compile-shader gl GL_VERTEX_SHADER vertex-shader-source)) (define fragment-shader (compile-shader gl GL_FRAGMENT_SHADER fragment-shader-source)) (define shader (link-shader gl vertex-shader fragment-shader)) ;; Create vertex buffer (define stride (* 4 5)) (define buffer (gl-create-buffer gl)) (gl-bind-buffer gl GL_ARRAY_BUFFER buffer) (gl-buffer-data gl GL_ARRAY_BUFFER #f32(-1.0 -1.0 1.0 0.0 0.0 1.0 -1.0 0.0 1.0 0.0 0.0 1.0 0.0 0.0 1.0) GL_STATIC_DRAW) ;; Draw (gl-viewport gl 0 0 (element-width canvas) (element-height canvas)) (gl-clear gl GL_COLOR_BUFFER_BIT) (gl-use-program gl shader) (gl-enable-vertex-attrib-array gl 0) (gl-vertex-attrib-pointer gl 0 2 GL_FLOAT 0 stride 0) (gl-enable-vertex-attrib-array gl 1) (gl-vertex-attrib-pointer gl 1 3 GL_FLOAT 0 stride 8) (gl-draw-arrays gl GL_TRIANGLES 0 3) Note that in Scheme, the equivalent of a Uint8Array is a bytevector. Hoot uses a packed array, an (array i8) specifically, for the contents of a bytevector. And here is the JavaScript code necessary to boot the resulting Wasm binary: window.addEventListener("load", async () => { function bytevectorToUint8Array(bv) { let len = reflect.bytevector_length(bv); let array = new Uint8Array(len); for (let i = 0; i < len; i++) { array[i] = reflect.bytevector_ref(bv, i); } return array; } let mod = await SchemeModule.fetch_and_instantiate("triangle.wasm", { reflect_wasm_dir: 'reflect-wasm', user_imports: { document: { getElementById: (id) => document.getElementById(id) }, element: { width: (elem) => elem.width, height: (elem) => elem.height }, canvas: { getContext: (elem, type) => elem.getContext(type) }, gl: { createShader: (gl, type) => gl.createShader(type), deleteShader: (gl, shader) => gl.deleteShader(shader), shaderSource: (gl, shader, source) => gl.shaderSource(shader, source), compileShader: (gl, shader) => gl.compileShader(shader), getShaderParameter: (gl, shader, param) => gl.getShaderParameter(shader, param), getShaderInfoLog: (gl, shader) => gl.getShaderInfoLog(shader), createProgram: (gl, type) => gl.createProgram(type), deleteProgram: (gl, program) => gl.deleteProgram(program), attachShader: (gl, program, shader) => gl.attachShader(program, shader), linkProgram: (gl, program) => gl.linkProgram(program), useProgram: (gl, program) => gl.useProgram(program), getProgramParameter: (gl, program, param) => gl.getProgramParameter(program, param), getProgramInfoLog: (gl, program) => gl.getProgramInfoLog(program), createBuffer: (gl) => gl.createBuffer(), deleteBuffer: (gl, buffer) => gl.deleteBuffer(buffer), bindBuffer: (gl, target, buffer) => gl.bindBuffer(target, buffer), bufferData: (gl, buffer, data, usage) => { let bv = new Bytevector(reflect, data); gl.bufferData(buffer, bytevectorToUint8Array(bv), usage); }, enableVertexAttribArray: (gl, index) => gl.enableVertexAttribArray(index), vertexAttribPointer: (gl, index, size, type, normalized, stride, offset) => { gl.vertexAttribPointer(index, size, type, normalized, stride, offset); }, drawArrays: (gl, mode, first, count) => gl.drawArrays(mode, first, count), viewport: (gl, x, y, w, h) => gl.viewport(x, y, w, h), clearColor: (gl, r, g, b, a) => gl.clearColor(r, g, b, a), clear: (gl, mask) => gl.clear(mask) } } }); let reflect = await mod.reflect({ reflect_wasm_dir: 'reflect-wasm' }); let proc = new Procedure(reflect, mod.get_export("$load").value); proc.call(); }); Hello problems There are two major performance issues with this program. One is visible in the source above, the other is hidden in the language implementation. Heap objects are opaque on the other side Wasm GC heap objects are opaque to the host. Likewise, heap objects from the host are opaque to the Wasm guest. Thus the contents of an (array i8) object are not visible from JavaScript and the contents of a Uint8Array are not visible from Wasm. This is a good security property in the general case, but it’s a hinderance in this specific case. Let’s say we have an (array i8) full of vertex data we want to put into a WebGL buffer. To do this, we must make one JS->Wasm call for each byte in the array and store it into a Uint8Array. This is what the bytevectorToUint8Array function above is doing. Copying any significant amount of data per frame is going to tank performance. Hope you aren’t trying to stream vertex data! Contrast the previous paragraph with Wasm linear memory. A WebAssembly.Memory object can be easily accessed from JavaScript as an ArrayBuffer. To get a blob of vertex data out of a memory object, you just need to know the byte offset and length and you’re good to go. There are many Wasm linear memory applications using WebGL successfully. Manipulating multi-byte binary data is inefficient To read a multi-byte number such as an unsigned 32-bit integer from an (array i8), you have to fetch each individual byte and combine them together. Here’s a self-contained example that uses Guile-flavored WAT format: (module (type $bytevector (array i8)) (data $init #u32(123456789)) (func (export "main") (result i32) (local $a (ref $bytevector)) (local.set $a (array.new_data $bytevector $init (i32.const 0) (i32.const 4))) (array.get_u $bytevector (local.get $a) (i32.const 0)) (i32.shl (array.get_u $bytevector (local.get $a) (i32.const 1)) (i32.const 8)) (i32.or) (i32.shl (array.get_u $bytevector (local.get $a) (i32.const 2)) (i32.const 16)) (i32.or) (i32.shl (array.get_u $bytevector (local.get $a) (i32.const 3)) (i32.const 24)) (i32.or))) By contrast, Wasm linear memory needs but a single i32.load instruction: (module (memory 1) (func (export "main") (result i32) (i32.store (i32.const 0) (i32.const 123456789)) (i32.load (i32.const 0)))) Easy peasy. Not only is it less code, it's a lot more efficient. Unsatisfying workarounds There’s no way around the multi-byte problem at the moment, but for byte access from JavaScript there are some things we could try to work with what we have been given. Spoiler alert: None of them are pleasant. Use Uint8Array from the host This approach makes all binary operations from within the Wasm binary slow since we’d have to cross the Wasm->JS bridge for each read/write. Since most of the binary data manipulation is happening in the Wasm module, this approach will just make things slower overall. Use linear memory for bytevectors This would require a little malloc/free implementation and a way to reclaim memory for GC'd bytevectors. You could register every bytevector in a FinalizationRegistry in order to be notified upon GC and free the memory. Now you have to deal with memory fragmentation. This is Wasm GC, we shouldn’t have to do any of this! Use linear memory as a scratch space This avoids crossing the Wasm/JS boundary for each byte, but still involves a byte-by-byte copy from (array i8) to linear memory within the Wasm module. So far this feels like the least worst option, but the extra copy is still going to greatly reduce throughput. Wasm GC needs some fixin' I’ve used realtime graphics as an example because it’s a use case that is very sensitive to performance issues, but this unfortunate need to copy binary data byte-by-byte is also the reason why strings are trash on Wasm GC right now. Stringref is a good proposal and the Wasm community group made a mistake by rejecting it. Anyway, there has been some discussion about both multi-byte and ArrayBuffer access on GitHub, but as far as I can tell neither issue is anywhere close to a resolution. Can these things be implemented efficiently? How can the need for direct access to packed arrays from JS be reconciled with Wasm heap object opaqueness? I hope the Wasm community group can arrive at solutions sooner than later because it will take a long time to get the proposal(s) to phase 4 and shipped in all browsers, perhaps years. It would be a shame to be effectively shut out from using WebGPU when it finally reaches stable browser releases.
I'm pleased to announce that the very first release of guile-bstructs, version 0.1.0, has been released! This is a library I've been working on for quite some time and after more than one rewrite and many smaller refactors I think it's finally ready to release publicly. Let's hope I'm not wrong about that! About guile-bstructs Guile-bstructs is a library that provides structured read/write access to binary data for Guile. A bstruct (short for “binary structure”) is a data type that encapsulates a bytevector and a byte offset which interprets that bytevector based on a specified layout. Some use cases for bstructs are: manipulating C structs when using the foreign function interface packing GPU vertex buffers when using graphics APIs such as OpenGL implementing data types that benefit from Guile's unboxed math optimizations such as vectors and matrices This library was initially inspired by guile-opengl's define-packed-struct syntax but is heavily based on "Ftypes: Structured foreign types" by Andy Keep and R. Kent Dybvig. The resulting interface is quite similar but the implementation is completely original. This library provides a syntax-heavy interface; nearly all of the public API is syntax. This is done to ensure that bstruct types are static and well-known at compile time resulting in efficient bytecode and minimal runtime overhead. A subset of the interface deals in raw bytevector access for accessing structured data in bytevectors directly without going through an intermediary bstruct wrapper. This low-level interface is useful for certain batch processing situations where the overhead of creating wrapper bstructs would hinder throughput. Example Here are some example type definitions to give you an idea of what it’s like to use guile-bstructs: ;; Struct (define-bstruct <vec2> (padded (struct (x float) (y float)))) ;; Type group with a union (define-bstruct (<mouse-move-event> (struct (type uint8) (x int32) (y int32))) (<mouse-button-event> (struct (type uint8) (button uint8) (state uint8) (x int32) (y int32))) (<event> (union (type uint8) (mouse-move <mouse-move-event>) (mouse-button <mouse-button-event>)))) ;; Array (define-bstruct <matrix4> (array 16 float)) ;; Bit fields (define-bstruct <date> (bits (year 32 s) (month 4 u) (day 5 u))) ;; Pointer (define-bstruct (<item> (struct (type int))) (<chest> (struct (opened? uint8) (item (* <item>))))) ;; Packed struct modifier (define-bstruct <enemy> (packed (struct (type uint8) (health uint32)))) ;; Endianness modifier (define-bstruct <big-float> (endian big float)) ;; Recursive type (define-bstruct <node> (struct (item int) (next (* <node>)))) ;; Mutually recursive type group (define-bstruct (<forest> (struct (children (* <tree>)))) (<tree> (struct (value int) (forest (* <forest>)) (next (* <tree>))))) ;; Opaque type (define-bstruct SDL_GPUTexture) Download Source tarball: guile-bstructs-0.1.0.tar.gz GPG signature: guile-bstructs-0.1.0.tar.gz.asc This release was signed with this GPG key. See the guile-bstructs project page for more information.
The Spring Lisp Game Jam 2024 ended one week ago. 48 games were submitted, a new record for the jam! This past week has been a time for participants to play and rate each other’s games. As I explored the entries, I noticed two distinct meta-patterns in how people approached building games with Lisp. I think these patterns apply more broadly to all applications of Lisp. Let’s talk about these patterns in some detail, with examples. But first! Here’s the breakdown of the jam submissions by language: lang entries % (rounded) ---- ------- ----------- guile 15 31 fennel 10 21 clojure 5 10 cl 5 10 racket 4 8 elisp 4 8 s7 3 6 kawa 1 2 owl 1 2 I haven’t rolled up the various Schemes (Guile, Racket, S7, Kawa) into a general scheme category because Scheme is so minimally specified and they are all very distinct implementations for different purposes, not to mention that Racket has a lot more going on than just Scheme. For the first time ever, Guile came out on top with the most submissions! There’s a very specific reason for this outcome. 11 out of the 15 Guile games were built for the web with Hoot, a Scheme-to-WebAssembly compiler that I work on at the Spritely Institute. 2 of those 11 were official Spritely projects. We put out a call for people to try making games with Hoot before the jam started, and a lot of people took us up on it! Very cool! The next most popular language, which is typically the most popular language in these jams, is Fennel. Fennel is a Lisp that compiles to Lua. It’s very cool, too! Also of note, at least to me as a Schemer, is that three games used S7. Hmm, there might be something relevant to this post going on there. The patterns I’m about to talk about could sort of be framed as “The Guile Way vs. The Fennel Way”, but I don’t want to do that. It's not an “us vs. them” thing. It’s wonderful that there are so many flavors of Lisp these days that anyone can find a great implementation that suits their preferences. Not only that, but many of these implementations can be used to make games that anyone can easily play in their web browser! That was not the case several years ago. Incredible! I want to preface the rest of this post by saying that both patterns are valid, and while I prefer one over the other, that is not to say that the other is inferior. I'll also show how these patterns can be thought of as two ends of a spectrum and how, in the end, compromises must be made. Okay, let’s get into it! Lisp as icing The icing pattern is using Lisp as a “scripting” language on top of a cake that is made from C, Rust, and other static languages. The typical way to do this is by embedding a Lisp interpreter into the larger program. If you’re most interested in writing the high-level parts of an application in Lisp then this pattern is the fastest way to get there. All you need is a suitable interpreter/compiler and a way to add the necessary hooks into your application. Since the program is mainly C/Rust/whatever, you can then use emscripten to compile it to WebAssembly and deploy to the web. Instant gratification, but strongly tied to static languages and their toolchains. S7 is an example of an embeddable Scheme. Guile is also used for extending C programs, though typically that involves dynamically linking to libguile rather than embedding the interpreter into the program’s executable. Fennel takes a different approach, recognizing that there are many existing applications that are already extensible through Lua, and provides a lispy language that compiles to Lua. Lisp as cake The cake pattern is using Lisp to implement as much of the software stack as possible. It’s Lisp all the way down... sorta. Rather than embedding Lisp into a non-Lisp program, the cake pattern does the inverse: the majority of the program is written in Lisp. When necessary, shared libraries can be called via a foreign function interface, but this should be kept to a minimum. This approach takes longer to yield results. Time is spent implementing missing libraries for your Lisp of choice and writing wrappers around the C shared libraries you can’t avoid using. Web deployment gets trickier, too, since the project is not so easily emscriptenable. (You may recognize this as the classic embed vs. extend debate. You’re correct! I'm just adding my own thoughts and applying it specifically to some real-world Lisp projects.) I mentioned Guile as an option for icing, but Guile really shines best as cake. The initial vision for Guile was to Emacsify other programs by adding a Scheme interpreter to them. These days, the best practice is to write your program in Scheme to begin with. Common Lisp is probably the best example, though. Implementations like SBCL have good C FFIs and can compile efficient native executables, minimizing the desire to use some C for performance reasons. Case studies Let’s take a look at some of the languages and libraries used for the Lisp Game Jam and evaluate their icing/cake-ness. Fennel + love2d love2d has been a popular choice for solo or small team game development for many years. It is a C++ program that embeds a Lua interpreter, which means it’s a perfect target for Fennel. Most Linux distributions package love2d, so it’s easy to run .love files natively. Additionally, thanks to emscripten, love2d games can be deployed to the web. Thus most Fennel games use love2d. ./soko.bin and Gnomic Vengeance are two games that use this stack. Fennel + love2d is a perfect example of Lisp as icing. Fennel sits at the very top of the stack, but there’s not really a path to spread Lisp into the layers below. It is also the most successful Lisp game development stack to date. S7 + raylib This stack is new to me, but two games used it this time around: GhostHop and Life Predictor. (You really gotta play GhostHop, btw. It’s a great little puzzle game and it is playable on mobile devices.) Raylib is a C library with bindings for many higher-level languages that has become quite popular in recent years. S7 is also implemented in C and is easily embeddable. This makes the combination easy to deploy on the web with emscripten. S7 + raylib is another example of Lisp as icing. I’m curious to see if this stack becomes more popular in future jams. Guile + Chickadee This is the stack that I helped build. Chickadee is a game library for Guile that implements almost all of the interesting parts in Scheme, including rendering. Two games were built with Chickadee in the most recent jam: Turbo Racer 3000 and Bloatrunner. Guile + Chickadee is an example of Lisp as cake. Chickadee wraps some C libraries for low-level tasks such as loading images, audio, and fonts, but it is written in pure Scheme. All the matrix and vector math is in Scheme. Chickadee comes with a set of rendering primitives comparable to love2d and raylib but they’re all implemented in Scheme. I’ve even made progress on rendering vector graphics with Scheme, whereas most other Lisp game libraries use a C library such as nanosvg. Chickadee has pushed the limits of Guile’s compiler and virtual machine, and Guile has been improved as a result. But it’s the long road. Chickadee is mostly developed by me, alone, in my very limited spare time. It is taking a long time to reach feature parity with more popular game development libraries, but it works quite well for what it is. Hoot + HTML5 canvas I also helped build this one. Hoot is a Scheme-to-WebAssembly compiler. Rather than compile the Guile VM (written in C) to Wasm using emscripten, Hoot implements a complete Wasm toolchain and a new backend for Guile’s compiler that emits Wasm directly. Hoot is written entirely in Scheme. Unlike C programs compiled with emscripten that target Wasm 1.0 with linear memory, Hoot targets Wasm 2.0 with GC managed heap types. This gives Hoot a significant advantage: Hoot binaries do not ship a garbage collector and thus are much smaller than Lisp runtimes compiled via emscripten. The Wasm binary for my game weighs in at < 2MiB whereas the love2d game I checked had a nearly 6MiB love.wasm. Hoot programs can also easily interoperate with JavaScript. Scheme objects can easily be passed to JavaScript, and vice versa, as they are managed in the same heap. With all of the browser APIs just a Wasm import away, an obvious choice for games was the built-in HTML5 canvas API for easy 2D rendering. 11 games used Hoot in the jam, including (shameless plug) Cirkoban and Lambda Dungeon. Hoot + HTML5 canvas is mostly dense cake with a bit of icing. On one hand, it took a year and significant funding to boot Hoot. We said “no” to emscripten, built our own toolchain, and extended Guile’s compiler. It's Lisp all the way until you hit the browser runtime! We even have a Wasm interpreter that runs on the Guile VM! Hoot rules! It was a risk but it paid off. On the other hand, the canvas API is very high-level. The more cake thing to do would be to use Hoot’s JS FFI to call WebGL and/or WebGPU. Indeed, this is the plan for the future! Wasm GC needs some improvements to make this feasible, but my personal goal is to get Chickadee ported to Hoot. I want Chickadee games to be easy to play natively and in browsers, just like love2d games. The cake/icing spectrum I must acknowledge the limitations of the cake approach. We’re not living in a world of Lisp machines, but a world of glorified PDP-11s. Even the tallest of Lisp cakes sits atop an even larger cake made mostly of C. All modern Lisp systems bottom out at some point. Emacs rests on a C core. Guile’s VM is written in C. Hoot runs on mammoth JavaScript engines written in C++ like V8. Games on Hoot currently render with HTML5 canvas rather than WebGL/WebGPU. Good luck using OpenGL without libGL; Chickadee uses guile-opengl which uses the C FFI to call into libGL. Then there’s libpng, FreeType, and more. Who the heck wants to rewrite all this in Lisp? Who even has the resources? Does spending all this time taking the scenic route matter at all, or are we just deluding ourselves because we have fun writing Lisp code? I think it does matter. Every piece of the stack that can be reclaimed from the likes of C is a small victory. The parts written in Lisp are much easier to hack on, and some of those things become live hackable while our programs are running. They are also memory safe, typically, thanks to GC managed runtimes. Less FFI calls means less overhead from traversing the Lisp/C boundary and more safety. As more of the stack becomes Lisp, it starts looking less like icing and more like cake. Moving beyond games, we can look to the Guix project as a great example of just how tasty the cake can get. Guix took the functional packaging model from the Nix project and made a fresh implementation, replacing the Nix language with Guile. Why? For code staging, code sharing, and improved hackability. Guix also uses an init system written in Guile rather than systemd. Why? For code staging, code sharing, and improved hackability. These are real advantages that make the trade-off of not using the industry-standard thing worth it. I’ve been using Guix since the early days, and back then it was easy to make the argument that Guix was just reinventing wheels for no reason. But now, over 10 years later, the insistence on maximizing the usage of Lisp has been key to the success of the project. As a user, once you learn the Guix idioms and a bit of Guile, you unlock extraordinary power to craft your OS to your liking. It’s the closest thing you can get to a Lisp machine on modern hardware. The cake approach paid off for Guix, and it could pay off for other projects, too. If Common Lisp is more your thing, and even if it isn’t, you’ll be amazed by the Trial game engine and how much of it is implemented in Common Lisp rather than wrapping C libraries. There’s also projects like Pre-Scheme that give me hope that one day the layers below the managed GC runtime can be implemented in Lisp. Pre-Scheme was developed and successfully used for Scheme 48 and I am looking forward to a modern revival of it thanks to an NLnet grant. I'm a cake boy That’s right, I said it: I’m a cake boy. I want to see projects continue to push the boundaries of what Lisp can do. When it comes to the Lisp Game Jam, what excites me most are not the games themselves, but the small advances made to reclaim another little slice of the cake from stale, dry C. I intend to keep pushing the limits for Guile game development with my Chickadee project. It’s not a piece of cake to bake a lispy cake, and the way is often hazy, but I know we can’t be lazy and just do the cooking by the book. Rewrite it in Rust? No way! Rewrite it in Lisp!
More in programming
The world is waking to the fact that talk therapy is neither the only nor the best way to cure a garden-variety petite depression. Something many people will encounter at some point in their lives. Studies have shown that exercise, for example, is a more effective treatment than talk therapy (and pharmaceuticals!) when dealing with such episodes. But I'm just as interested in the role building competence can have in warding off the demons. And partly because of this meme: I've talked about it before, but I keep coming back to the fact that it's exactly backwards. That signing up for an educational quest into Linux, history, or motorcycle repair actually is an incredibly effective alternative to therapy! At least for men who'd prefer to feel useful over being listened to, which, in my experience, is most of them. This is why I find it so misguided when people who undertake those quests sell their journey short with self-effacing jibes about how much an unattractive nerd it makes them to care about their hobby. Mihaly Csikszentmihalyi detailed back in 1990 how peak human happiness arrives exactly in these moments of flow when your competence is stretched by a difficult-but-doable challenge. Don't tell me those endorphins don't also help counter the darkness. But it's just as much about the fact that these pursuits of competence usually offer a great opportunity for community as well that seals the deal. I've found time and again that people are starved for the kind of topic-based connections that, say, learning about Linux offers in spades. You're not just learning, you're learning with others. That is a time-tested antidote to depression: Forming and cultivating meaningful human connections. Yes, doing so over the internet isn't as powerful as doing it in person, but it's still powerful. It still offers community, involvement, and plenty of invitation to carry a meaningful burden. Open source nails this trifecta of motivations to a T. There are endless paths of discovery and mastery available. There are tons of fellow travelers with whom to connect and collaborate. And you'll find an unlimited number of meaningful burdens in maintainerships open for the taking. So next time you see that meme, you should cheer that the talk therapy table is empty. Leave it available for the severe, pathological cases that exercise and the pursuit of competence can't cure. Most people just don't need therapy, they need purpose, they need competence, they need exercise, and they need community.
The excellent-but-defunct blog Programming in the 21st Century defines "puzzle languages" as languages were part of the appeal is in figuring out how to express a program idiomatically, like a puzzle. As examples, he lists Haskell, Erlang, and J. All puzzle languages, the author says, have an "escape" out of the puzzle model that is pragmatic but stigmatized. But many mainstream languages have escape hatches, too. Languages have a lot of properties. One of these properties is the language's capabilities, roughly the set of things you can do in the language. Capability is desirable but comes into conflicts with a lot of other desirable properties, like simplicity or efficiency. In particular, reducing the capability of a language means that all remaining programs share more in common, meaning there's more assumptions the compiler and programmer can make ("tractability"). Assumptions are generally used to reason about correctness, but can also be about things like optimization: J's assumption that everything is an array leads to high-performance "special combinations". Rust is the most famous example of mainstream language that trades capability for tractability.1 Rust has a lot of rules designed to prevent common memory errors, like keeping a reference to deallocated memory or modifying memory while something else is reading it. As a consequence, there's a lot of things that cannot be done in (safe) Rust, like interface with an external C function (as it doesn't have these guarantees). To do this, you need to use unsafe Rust, which lets you do additional things forbidden by safe Rust, such as deference a raw pointer. Everybody tells you not to use unsafe unless you absolutely 100% know what you're doing, and possibly not even then. Sounds like an escape hatch to me! To extrapolate, an escape hatch is a feature (either in the language itself or a particular implementation) that deliberately breaks core assumptions about the language in order to add capabilities. This explains both Rust and most of the so-called "puzzle languages": they need escape hatches because they have very strong conceptual models of the language which leads to lots of assumptions about programs. But plenty of "kitchen sink" mainstream languages have escape hatches, too: Some compilers let C++ code embed inline assembly. Languages built on .NET or the JVM has some sort of interop with C# or Java, and many of those languages make assumptions about programs that C#/Java do not. The SQL language has stored procedures as an escape hatch and vendors create a second escape hatch of user-defined functions. Ruby lets you bypass any form of encapsulation with send. Frameworks have escape hatches, too! React has an entire page on them. (Does eval in interpreted languages count as an escape hatch? It feels different, but it does add a lot of capability. Maybe they don't "break assumptions" in the same way?) The problem with escape hatches In all languages with escape hatches, the rule is "use this as carefully and sparingly as possible", to the point where a messy solution without an escape hatch is preferable to a clean solution with one. Breaking a core assumption is a big deal! If the language is operating as if its still true, it's going to do incorrect things. I recently had this problem in a TLA+ contract. TLA+ is a language for modeling complicated systems, and assumes that the model is a self-contained universe. The client wanted to use the TLA+ to test a real system. The model checker should send commands to a test device and check the next states were the same. This is straightforward to set up with the IOExec escape hatch.2 But the model checker assumed that state exploration was pure and it could skip around the state randomly, meaning it would do things like set x = 10, then skip to set x = 1, then skip back to inc x; assert x == 11. Oops! We eventually found workarounds but it took a lot of clever tricks to pull off. I'll probably write up the technique when I'm less busy with The Book. The other problem with escape hatches is the rest of the language is designed around not having said capabilities, meaning it can't support the feature as well as a language designed for them from the start. Even if your escape hatch code is clean, it might not cleanly integrate with the rest of your code. This is why people complain about unsafe Rust so often. It should be noted though that all languages with automatic memory management are trading capability for tractability, too. If you can't deference pointers, you can't deference null pointers. ↩ From the Community Modules (which come default with the VSCode extension). ↩
I wrote a lot of blog posts over my time at Parse, but they all evaporated after Facebook killed the product. Most of them I didn’t care about (there were, ahem, a lot of status updates and “service reliability announcements”, but I was mad about losing this one in particular, a deceptively casual retrospective of […]
It's only been two months since I discovered the power and joy of this new generation of mini PCs. My journey started out with a Minisforum UM870, which is a lovely machine, but since then, I've come to really appreciate the work of Beelink. In a crowded market for mini PCs, Beelink stands out with their superior build quality, their class-leading cooling and silent operation, and their use of fully Linux-compatible components (the UM870 shipped with a MediaTek bluetooth/wifi card that doesn't work with Linux!). It's the complete package at three super compelling price points. For $289, you can get the EQR5, which runs an 8-core AMD Zen3 5825U that puts out 1723/6419 in Geekbench, and comes with 16GB RAM and 500GB NVMe. I've run Omarchy on it, and it flies. For me, the main drawback was the lack of a DisplayPort, which kept me from using it with an Apple display, and the fact that the SER8 exists. But if you're on a budget, and you're fine with HDMI only, it's a wild bargain. For $499, you can get the SER8. That's the price-to-performance sweet spot in the range. It uses the excellent 8-core AMD Zen4 8745HS that puts out 2595/12985 in Geekbench (~M4 multi-core numbers!), and runs our HEY test suite with 30,000 assertions almost as fast as an M4 Max! At that price, you get 32GB RAM + 1TB NVMe, as well as a DisplayPort, so it works with both the Apple 5K Studio Display and the Apple 6K XDR Display (you just need the right cable). Main drawback is limited wifi/bluetooth range, but Beelink tells me there's a fix on the way for that. For $929, you can get the SER9 HX370. This is the top dog in this form factor. It uses the incredible 12-core AMD Zen5 HX370 that hits 2990/15611 in Geekbench, and runs our HEY test suite faster than any Apple M chip I've ever tested. The built-in graphics are also very capable. Enough to play a ton of games at 1080p. It also sorted the SER8's current wifi/bluetooth range issue. I ran the SER8 as my main computer for a while, but now I'm using the SER9, and I just about never feel like I need anything more. Yes, the Framework Desktop, with its insane AMD Max 395+ chip, is even more bonkers. It almost cuts the HEY test suite time in half(!), but it's also $1,795, and not yet generally available. (But preorders are open for the ballers!). Whichever machine fits your budget, it's frankly incredible that we have this kind of performance and efficiency available at these prices with all of these Beelinks drawing less than 10 watt at idle and no more than 100 watt at peak! So it's no wonder that Beelink has been selling these units like hotcakes since I started talking about them on X as the ideal, cheap Omarchy desktop computers. It's such a symbiotic relationship. There are a ton of programmers who have become Linux curious, and Beelink offers no-brainer options to give that a try at a bargain. I just love when that happens. The perfect intersection of hardware, software, and timing. That's what we got here. It's a Beelink, baby! (And no, before you ask, I don't get any royalties, there's no affiliate link, and I don't own any shares in Beelink. I just love discovering great technology and seeing people start their Linux journey with an awesome, affordable computer!)
“One of the comments that sparked this article,” our founder Paul McMahon told me, “was someone saying, ‘I don’t really want to do networking because it seems kind of sleazy. I’m not that kind of person.’” I guess that’s the key misconception people have when they hear ‘networking.’ They think it’s like some used car salesman kind of approach where you have to go and get something out of the person. That’s a serious error, according to Paul, and it worries him that so many developers share that mindset. Instead, Paul considers networking a mix of making new friends, growing a community, and enjoying serendipitous connections that might not bear fruit until years later, but which could prove to be make-or-break career moments. It’s something that you don’t get quick results on and that doesn’t make a difference at all until it does. And it’s just because of the one connection you happen to make at an event you went to once, this rainy Tuesday night when you didn’t really feel like going, but told yourself you have to go—and that can make all the difference. As Paul has previously shared, he can attribute much of his own career success—and, interestingly enough, his peace of mind—to the huge amount of networking he’s done over the years. This is despite the fact that Paul is, in his own words, “not such a talkative person when it comes to small talk or whatever.” Recently I sat down with Paul to discuss exactly how developers are networking “wrong,” and how they can get it right instead. In our conversation, we covered: What networking really is, and why you need to start ASAP Paul’s top tip for anyone who wants to network Advice for networking as an introvert Online vs offline networking—which is more effective? And how to network in Japan, even when you don’t speak Japanese What is networking, really, and why should you start now? “Sometimes,” Paul explained, “people think of hiring fairs and various exhibitions as the way to network, but that’s not networking to me. It’s purely transactional. Job seekers are focused on getting interviews, recruiters on making hires. There’s no chance to make friends or help people outside of your defined role.” Networking is getting to know other people, understanding how maybe you can help them and how they can help you. And sometime down the road, maybe something comes out of it, maybe it doesn’t, but it’s just expanding your connections to people. One reason developers often avoid or delay networking is that, at its core, networking is a long game. Dramatic impacts on your business or career are possible—even probable—but they don’t come to fruition immediately. “A very specific example would be TokyoDev,” said Paul. “One of our initial clients that posted to the list came through networking.” Sounds like a straightforward result? It’s a bit more complicated than that. “There was a Belgian guy, Peter, whom I had known through the Ruby and tech community in Japan for a while,” Paul explained. “We knew each other, and Peter had met another Canadian guy, Jack, who [was] looking to hire a Ruby developer. “So Peter knew about me and TokyoDev and introduced me to Jack, and that was the founder of Degica, who became one of our first clients. . . . And that just happened because I had known Peter through attending events over the years.” Another example is how Paul’s connection to the Ruby community helped him launch Doorkeeper. His participation in Ruby events played a critical role in helping the product succeed, but only because he’d already volunteered at them for years. “Because I knew those people,” he said, “they wanted to support me, and I guess they also saw that I was genuine about this stuff, and I wasn’t participating in these events with some big plan about, ‘If I do this, then they’re going to use my system,’ or whatever. Again, it was people helping each other out.” These delayed and indirect impacts are why Paul thinks you should start networking right now. “You need to do it in advance of when you actually need it,” he said. “People say they’re looking for a job, and they’re told ‘You could network!’ Yeah, that could potentially help, but it’s almost too late.” You should have been networking a couple years ago when you didn’t need to be doing it, because then you’ve already built up the relationships. You can have this karma you’re building over time. . . . Networking has given me a lot of wealth. I don’t mean so much in money per se, but more it’s given me a safety net. “Now I’m confident,” he said, “that if tomorrow TokyoDev disappeared, I could easily find something just through my connections. I don’t think I’ll, at least in Japan, ever have to apply for a job again.” “I think my success with networking is something that anyone can replicate,” Paul went on, “provided they put in the time. I don’t consider myself to be especially skilled in networking, it’s just that I’ve spent over a decade making connections with people.” How to network (the non-sleazy way) Paul has a fair amount of advice for those who want to network in an effective, yet genuine fashion. His first and most important tip: Be interested in other people. Asking questions rather than delivering your own talking points is Paul’s number one method for forging connections. It also helps avoid those “used car salesman” vibes. “ That’s why, at TokyoDev,” Paul explained, “we typically bar recruiters from attending our developer events. Because there are these kinds of people who are just going around wanting to get business cards from everyone, wanting to get their contact information, wanting to then sell them on something later. It’s quite obvious that they’re like that, and that leads to a bad environment, [if] someone’s trying to sell you on something.” Networking for introverts The other reason Paul likes asking questions is that it helps him to network as an introvert. “That’s actually one of the things that makes networking easier for someone who isn’t naturally so talkative. . . . When you meet new people, there are some standard questions you can ask them, and it’s like a blank slate where you’re filling in the details about this person.” He explained further that going to events and being social can be fun for him, but he doesn’t exactly find it relaxing. “When it comes to talking about something I’m really interested in, I can do it, but I stumble in these social situations. Despite that, I think I have been pretty successful when it comes to networking.” “What has worked well for me,” he went on, “has been putting myself in those situations that require me to do some networking, like going to an event.” Even if you aren’t that proactive, you’re going to meet a couple of people there. You’re making more connections than you would if you stayed home and played video games. The more often you do it, the easier it gets, and not just because of practice: there’s a cumulative effect to making connections. “Say you’re going to an event, and maybe last time you met a couple of people, you could just say ‘Hi’ to those people again. And maybe they are talking with someone else they can introduce you to.” Or, you can be the one making the introductions. “What has also worked well for me, is that I like to introduce other people,” Paul said. It’s always a great feeling when I’m talking to someone at an event, and I hear about what they’re doing or what they’re wanting to do, and then I can introduce someone else who maybe matches that. “And it’s also good for me, then I can just be kind of passive there,” Paul joked. “I don’t have to be out there myself so much, if they’re talking to each other.” His last piece of advice for introverts is somewhat counterintuitive. “Paradoxically,” he told me, “it helps if you’re in some sort of leadership position.” If you’re an introvert, my advice would be one, just do it, but then also look for opportunities for helping in some more formal capacity, whether it’s organizing an event yourself, volunteering at an event . . . [or] making presentations. “Like for me, when I’ve organized a Tokyo Rubyist Meetup,” Paul said, “[then] naturally as the organizer there people come to talk to me and ask me questions. . . . And it’s been similar when I’ve presented at an event, because then people have something that they know that you know something about, and maybe they want to know more about it, and so then they can ask you more questions and lead the conversation that way.” Offline vs online networking When it comes to offline vs online networking, Paul prefers offline. In-person events are great for networking because they create serendipity. You meet people through events you wouldn’t meet otherwise just because you’re in the same physical space as them. Those time and space constraints add pressure to make conversation—in a good way. “It’s natural when you are meeting someone, you ask about what they’re doing, and you make that small connection there. Then, after seeing them at multiple different events, you get a bit of a stronger connection to them.” “Physical events are [also] much more constrained in the number of people, so it’s easier to help people,” he added. “Like with TokyoDev, I can’t help every single person online there, but if someone meets me at the event [and is] asking for advice or something like that, of course I’ve got to answer them. And I have more time for them there, because we’re in the same place at the same time.” As humans, we’re more likely to help other people we have met in person, I think just because that’s how our brains work. That being said, Paul’s also found success with online networking. For example, several TokyoDev contributors—myself included—started working with Paul after interacting with him online. I commented on TokyoDev’s Dungeons and Dragons article, which led to Paul checking my profile and asking to chat about my experience. Scott, our community moderator and editor, joined TokyoDev in a paid position after being active on the TokyoDev Discord. Michelle was also active on the Discord, and Paul initially asked her to write an article for TokyoDev on being a woman software engineer in Japan, before later bringing her onto the team. Key to these results was that they involved no stereotypical “networking” strategies on either side: we all connected simply by playing a role in a shared, online community. Consistent interactions with others, particularly over a longer period of time, builds mutual trust and understanding. Your online presence can help with offline networking. As TokyoDev became bigger and more people knew about me through my blog, it became a lot easier to network with people at events because they’re like, ‘Hey, you’re Paul from TokyoDev. I like that site.’ “It just leads to more opportunities,” he continued. “If you’ve interacted with someone before online, and then you meet them offline, you already do have a bit of a relationship with them, so you’re more likely to have a place to start the conversation. [And] if you’re someone who is struggling with doing in-person networking, the more you can produce or put out there [online], the more opportunities that can lead to.” Networking in Japanese While there are a number of events throughout Japan that are primarily in English, for best networking results, developers should take advantage of Japanese events as well—even if your Japanese isn’t that good. In 2010, Paul created the Tokyo Rubyist Meetup, with the intention of bringing together Japanese and international developers. To ensure it succeeded, he knew he needed more connections to the Japanese development community. “So I started attending a lot of Japanese developer events where I was the only non-Japanese person there,” said Paul. “I didn’t have such great Japanese skills. I couldn’t understand all the presentations. But it still gave me a chance to make lots of connections, both with people who would later present at [Tokyo Rubyist Meetup], but also with other Japanese developers whom I would work with either on my own products or also on other client projects.” I think it helped being kind of a visible minority. People were curious about me, about why I was attending these events. Their curiosity not only helped him network, but also gave him a helping hand when it came to Japanese conversation. “It’s a lot easier for me in Japanese to be asked questions and answer them,” he admitted. But Paul wasn’t just attending those seminars and events in a passive manner. He soon started delivering presentations himself, usually as part of Lightning Talks—again, despite his relatively low level of Japanese. “It doesn’t matter if you do a bad job of it,” he said. Japanese people I think are really receptive to people trying to speak in Japanese and making an effort. I think they’re happy to have someone who isn’t Japanese present, even if they don’t do a great job. He also quickly learned that the most important networking doesn’t take place at the meetup itself. “At least in the past,” he explained, “it was really split . . . [there’s the] seminar time where everyone goes and watches someone present. Everyone’s pretty passive there and there isn’t much conversation going on between attendees. “Then afterwards—and maybe less than half of the people attend—but they go to a restaurant and have drinks after the event. And that’s where all the real socialization happens, and so that’s where I was able to really make the most connections.” That said, Paul noted that the actual “drinking” part of the process has noticeably diminished. “Drinking culture in Japan is changing a lot,” he told me. “I noticed that even when hosting the Tokyo Rubyist Meetup. When we were first hosting it, we [had] an average of 2.5 beers per participant. And more recently, the average is one or less per participant there. “I think there is not so much of an expectation for people to drink a lot. Young Japanese people don’t drink at the same rate, so don’t feel like you actually have to get drunk at these events. You probably shouldn’t,” he added with a laugh. What you should do is be persistent, and patient. It took Paul about a year of very regularly attending events before he felt he was treated as a member of the community. “Literally I was attending more than the typical Japanese person,” he said. “At the peak, there were a couple events per week.” His hard work paid off, though. “I think one thing about Japanese culture,” he said, “is that it’s really group based.” Initially, as foreigners, we see ourselves in the foreign group versus the Japanese group, and there’s kind of a barrier there. But if you can find some other connection, like in my case Ruby, then with these developers I became part of the “Ruby developer group,” and then I felt much more accepted. Eventually he experienced another benefit. “I think it was after a year of volunteering, maybe two years. . . . RubyKaigi, the biggest Ruby conference in Japan and one of the biggest developer conferences in Japan [in general], used Doorkeeper, the event registration system [I created], to manage their event. “That was a big win for us because it showed that we were a serious system to lots of people there. It exposed us to lots of potential users and was one of the things that I think led to us, for a time, being the most popular event registration system among the tech community in Japan.” Based on his experiences, Paul would urge more developers to try attending Japanese dev events. “Because I think a lot of non-Japanese people are still too intimidated to go to these events, even if they have better Japanese ability than I did. “If you look at most of the Japanese developer events happening now, I think the participants are almost exclusively Japanese, but still, that doesn’t need to be the case.” Takeaways What Paul hopes other developers will take away from this article is that networking shouldn’t feel sleazy. Instead, good networking looks like: Being interested in other people. Asking them questions is the easiest way to start a conversation and make a genuine connection. Occasionally just making yourself go to that in-person event. Serendipity can’t happen if you don’t create opportunities for it. Introducing people to each other—it’s a fast and stress-free way to make more connections. Volunteering for events or organizing your own. Supporting offline events with a solid online presence as well. Not being afraid to attend Japanese events, even if your Japanese isn’t good. Above all, Paul stressed, don’t overcomplicate what networking is at its core. Really what networking comes down to is learning about what other people are doing, and how you can help them or how they can help you. Whether you’re online, offline, or doing it in Japanese, that mindset can turn networking from an awkward, sleazy-feeling experience into something you actually enjoy—even on a rainy Tuesday night.
