June 7, 2026
Fast app, furious comments
How's Linear so fast? A technical breakdown
People came for the speed secret and stayed for the jokes, nostalgia, and flexes
TLDR: Linear feels fast because it updates on your device first and syncs later, cutting out the usual wait. Commenters loved the slickness, joked about expecting math content instead, and argued the real story is how brutally hard this kind of speed is to build.
The big reveal in Dennis Brotzky’s breakdown is surprisingly simple: Linear feels insanely fast because it does as much work as possible on your device first, instead of making you wait for the internet every time you click something. In plain English, the app updates immediately, then quietly syncs in the background. That means fewer loading screens, fewer spinning wheels, and a whole lot less "why is this taking forever?"
But the comments? That’s where the real party started. One reader showed up expecting a totally different kind of “Linear” and cracked, “I was expecting a post on how hardware accelerate linear algebra lol” — instantly setting the tone for a thread full of nerdy whiplash and comedy. Others got weirdly emotional in the best way, with one commenter basically declaring this a victory lap for old-school ideas, saying seeing browser-based data and instant updates again made them happy and reminded them of MeteorJS. Yes, the tech nostalgia hit hard.
Then came the subtle flexes. One commenter casually dropped that they’d rebuilt this kind of system themselves and, oh by the way, it took months. That turned the discussion from “cool article” into “okay, so this is actually very hard.” The overall mood was admiration with a side of envy: people aren’t just impressed that Linear is slick — they’re realizing that speed like this isn’t magic, it’s a mountain of smart decisions. And that may be the spiciest part of all.
Key Points
- •The article argues that Linear’s speed comes from architectural decisions rather than a single optimization.
- •It says Linear stores and reads user-facing data from the browser using IndexedDB instead of depending on immediate server round trips.
- •According to the article, updates are applied locally first and then synchronized asynchronously to the server, which sends deltas to other clients over WebSocket.
- •The article contrasts a traditional spinner-based PATCH request flow with Linear’s local state update plus queued save operation.
- •A quoted 2024 remark from co-founder Tuomas Artman is used to show that Linear built its sync engine very early in the product’s development.