Towards a server-less Web
summary
WebRTC provides browser-to-browser peer-to-peer channels (encrypted DTLS/SRTP) and a JavaScript dataChannel API; the prototype called Bish (browser based open publishing) uses a chord DHT for location‑independent URIs, stores only name→host records, and lets the client drop the signaling server after initial fetch. Measured limits: browsers hold ~64–100 simultaneous connections, initial connection setup can take ~1s per hop, and NAT traversal fails in many NAT setups (roughly 40% in small tests). Identity, privacy and robust replication remain unresolved.
AI-written summary · about
chapters (20)
- 0:00Start
- 3:53Take back Control
- 5:20What is WebRTC?
- 8:41Networking in Browsers
- 10:50This is WebRTC
- 14:45The Protocol Stack
- 20:06The API: Basics
- 23:12The API: Data Channels
- 28:03The API (on the wire)
- 29:52BOPlish: Overview
- 31:10BOPlish: What do we need?
- 32:38Routing
- 34:38Content Names
- 37:28Name Resolution: DHT a distributed key/value store
- 39:55Name Resolution: API
- 40:26Simple API
- 47:09Use Cases: Document Sharing
- 48:42Use Cases: Content Search
- 49:03Use Cases: Real-Time Chat
- 50:43Recap/Outlook
related talks
33mWebRTC Change Communications Forever
31mWebRTC: Real-time communication without plugins
32mExploring the P2P world with WebRTC & JavaScript
35mExploring the world of decentralized networks with WebRTC
27mWebRTC: Real-Time Communications
53mThe State of WebRTC
43mThe unstoppable secret superpower of WebRTC
20mThe disintermediated web