JavaScript and mathematical computing across servers and clients

on YouTube ↗17m

" summary

Claim: JavaScript can run real scientific work across clients, servers and embedded devices, but the specifics matter. Libraries demonstrated include numbers.js and brain.js; compilation via mcrip 10 (emscripten) lets C/C++ numeric code run on Node (compiled linear‑algebra library runs on Node and crashes large browsers but oddly works in Safari). A drone cat classifier is built and tested and fails due to poor training data. Major caveat: 64‑bit numeric limits require native fallbacks or RPC tricks (Duo/Che). NaCl and d3.js are used for in‑browser models and visualization.

AI-written summary · about

" chapters (6)

1. 0:00Scientific computing with Javascript
2. 3:48Javascript: The PhoneGap for scientific computing
3. 9:44Demo: Some sample libraries and their applications.
4. 11:54Demo: Emscripten
5. 12:19Forging ahead with Emscripten
6. 14:15Questions Accost me.

" related talks

• 43mThe Compute Revolution You’re Ignoring: JavaScript in ScienceGunj Joshi · DevConf.IN 2026
• 40mJavaScript ... for Science!Angelina Fabbro, Bill Mills · Strange Loop 2013
• 12mThe Goodness of JavaScriptDave Geddes, Aaron Frost · Fluent 2014
• 24mMath In JavaScript Can Be AwesomeDominic Kramer · Node+JS Interactive 2018
• 38mJavascript Considered...UsefulJenn Schiffer · JSConf EU 2019
• 52mJavaScript at the Speed of LightDmitry Lomov · The Rolling Scopes Conference 2015
• 01h 08mHow JavaScript WorksDouglas Crockford · Silicon Valley Code Camp 2018
• 15mThe future of machine learning using JavaScriptDoni Rubiagatra · CityJS Singapore 2023