Understanding the WebRTC architecture from a free software perspective

WebRTC is here, now.

Google launched Chrome 25 with built-in WebRTC support on Monday, 25 February. Within 24 hours, DruCall, an easy-to-use WebRTC module for Drupal was available Free as a shiny new GPL project.

While this technology offers a level of convenience for end-users that is unprecedented, just what is involved for the keen server administrator who wants to deploy this today?

With 20-30% of web users expected to be running WebRTC browsers within 3 months (due to automatic updates), and 60-70% by Christmas, will the free software community be able to seize this opportunity to redefine the way real-time communications is deployed?

I've whipped together a brief architecture diagram to answer those questions, demonstrating how all core components can be built with free software, today. All but one of these components is conveniently packaged in stable Linux distributions like Debian and Ubuntu. The full diagram is taken from the DruCall web site where you can find more details: