PeerJS is, according to its developers, designed to "complete WebRTC", since WebRTC in its current form doesn't say much about how to find other users with which to make connections. PeerJS looks to bridge that particular gap in functionality by offering up an API that allows for connections to be made with the addition of just three lines of code, and even establishes the handshaking process to make the necessary connections happen. PeerJS also offers PeerServer and PeerServer Cloud, an open source Web server and its cloud-based equivalent to give the clients the ability to make the rendezvous in the first place.
PeerJS is still a work in progress system, however, as it depends on WebRTC DataChannel and various other browser features to operate, but it's a system that has quite a bit of potential behind it. So far, only Chrome 26+, the Dev and Canary versions, can actually use PeerJS--no version of Firefox can yet use PeerJS--but there's quite a bit of development going on that should make this not only usable on more browsers, but also for more purposes when it does emerge .What's more, the developers of PeerJS believe that DataChannel will be found in Firefox and Chrome alike, in stable versions, within the next three months, which will give PeerJS a lot more room to run.
Admittedly, in the short term, PeerJS's ability to "complete WebRTC" is limited at best. When the product in question can't be found on most browsers in the field, it sort of makes the overall value a little on the light side. But PeerJS does have an interesting future ahead--if it can really start getting on browsers within the next three months it will be a very good sign indeed for the group--as well as a point about the ability to locate other users to establish a connection with in the first place.
So for the time being, PeerJS' impact on the overall picture should be a bit slim, but there's likely quite a bit that's going to follow from this system, and in reasonably short order besides.
Edited by Brooke Neuman