As far as I understand, crossbar provides transparent connection between clients. So clients can communicate with each other. WebRTC does almost same thing, without a central server(apart from signalling server) with lower latency.
So why we need crossbar in the first place?
Am I missing something?
The two technologies are very different from one another.
WebRTC enables voice, video and data to be passed in real time directly between browsers. It has no notion of messaging, signaling or even a network architecture. It is a kind of a building block that got implemented inside a web browser and you can now use. When using it, you'll still be needing servers to get one browser/device connected to another browser/device - a signaling server.
From looking at the site of crossbar.io, it seems like this is an open source framework that implements signaling for IOT devices - it can connect devices to a centralized server that handles the control messages and signals that get sent between the devices and/or users.
Here's the interesting thing - if Crossbar were to handle voice/video or low latency data transmission - they'd probably end up integrating WebRTC. And on the other hand, you can use Crossbar as a signaling mechanism for WebRTC.
Apples to oranges.