BurnChat

Hey IH 👋 I just launched BurnChat (https://burnchat.io) — disposable chat rooms with zero logs. I want to share how I built it, what I learned, and where I'm at. ## What is it? BurnChat is a web-based chat room that stores nothing. No database. No log files. No user accounts. You create a room, share the link, chat, and hit the burn button when you're done. Every message is wiped from server memory in milliseconds. That's it. That's the whole product. ## Why I built it I kept running into the same problem: I needed to share something privately with someone — a password, a quick plan, sensitive info — and every option either required an account, stored the messages forever, or both. Then the Disney Slack breach happened. 44 million messages leaked. Not because of some sophisticated attack — because Slack stores everything permanently and someone got in. That was the push I needed. I wanted to prove you could build a useful chat tool that literally stores nothing. Not "we encrypt it." Not "we delete it after 30 days." Nothing. Messages exist in server RAM while the room is active and vanish when it closes. ## How I built it The whole backend is a single server.js file. About 300 lines of code. Stack: - Node.js + Express + Socket.IO - Zero storage dependencies (no database driver, because there's no database) - Vanilla HTML/CSS/JS frontend, no build step - Hosted on a single VPS with OpenLiteSpeed Messages are stored in a JavaScript Map object in memory. When someone hits burn, the array is zeroed, socket connections are severed, the room is deleted, and the garbage collector reclaims the memory. Under 10 milliseconds. What I had to actively turn off: - Web server access logs - Error logs (routed to /dev/null) - System journal - All analytics - All cookies and tracking Building something that stores nothing means fighting every default in the web dev ecosystem. Everything assumes you want to log, track, and collect. ## Features - 🔥 Burn button — any participant can destroy the room instantly - 🔒 Room passwords — lock/unlock on the fly - 🔗 Custom URLs — burnchat.io/your-room-name - 📷 Image sharing — compressed in browser, lives in RAM - ⚡ Auto-burn — room self-destructs when everyone leaves - 🌓 Day/night mode — auto-detects by time of day - 📊 Admin panel — aggregate stats only, no PII ## What's working - Site is live at burnchat.io - Blog with 5 technical articles driving organic traffic - BitcoinTalk announcement with a $1,500 USDT giveaway running (3 weeks, $500/week) - Twitter account (@burnchatio) with scheduled content via a custom bot - Submitted to privacy directories and awesome lists ## What I learned 1. "No database" is a feature, not a limitation. People hear "no database" and think it's a prototype. It's the product. The absence of storage IS the privacy. 2. Simple products are harder to explain. "It's a chat room that disappears" takes 10 minutes to explain because people keep looking for the catch. There is no catch. 3. Privacy tools have a trust problem. Everyone claims to be private. The difference is: do you ask users to trust your policy, or does your architecture make abuse impossible? We went with architecture. 4. Content marketing works for niche tools. Our blog posts about what "delete" actually means on Slack/Teams/Discord are the biggest traffic drivers. People share them because they're genuinely surprised. ## Numbers (being transparent) - Revenue: $0 (it's free, no monetization yet) - Users: I literally don't know (no analytics) - Rooms burned: visible on the homepage counter - Cost to run: ~$10/month (single VPS) - Time to build: about 2 weeks to v1, ongoing iterations ## What's next - Chrome extension (right-click → "Open BurnChat room") - Telegram bot (/burnchat generates a room link) - Possibly end-to-end encryption - Exploring monetization (premium features for teams? donations?) ## Try it https://burnchat.io Create a room. Send a message. Hit burn. Try to find any trace that the room existed. You won't. Happy to answer any questions about the architecture, marketing, or anything else. And if you have ideas for monetization that don't involve tracking users... I'm all ears.