News for hackers. Stuff that matters.
Re:Show HN: WebRTC Data Channels Example by Hydraulix989 on Saturday December 16, 2017 @09:22AM (#15939642)
The signaling server is where a lot of the real complexity comes from [1]. I even had to write my own from scratch for our production implementation.

[1] https://www.sococo.com/blog/webrtc-signaling-here-be-dragons

Re:Show HN: WebRTC Data Channels Example by joantune on Saturday December 16, 2017 @09:59AM (#15939791)
Did you look into any of these:

https://www.html5rocks.com/en/tutorials/webrtc/infrastructur... before implementing that one?

Re:Show HN: WebRTC Data Channels Example by Hydraulix989 on Saturday December 16, 2017 @03:08PM (#15941700)
Yes, that's the first Google result, and they suggested building your own using socket.io -- read my article, if you haven't already (but some of their other _stack_ suggestions are decent).

Their ready-made signaling server suggestions are complete garbage (Signalmaster is buggy and incorrect).

Re:Show HN: WebRTC Data Channels Example by moron4hire on Saturday December 16, 2017 @10:41AM (#15939981)
I've not found signalling to be complex at all. You're just forwarding messages from one user to another. The server in the middle doesn't need to know anything other than to whom the messages go.

The one thing that is not handled in any material I've ever read on WebRTC: how to sequence more-than-two users joining a session at the same time, and how to recover from errors and stalled connection attempts (where the latter is a significant impediment to the former).

Re:Show HN: WebRTC Data Channels Example by Hydraulix989 on Saturday December 16, 2017 @08:36PM (#15943423)
Let me test your implementation then. ;-)
Re:Show HN: WebRTC Data Channels Example by GroSacASacs on Monday December 18, 2017 @09:35AM (#15951834)
How did the situation evolve since 2014