Transcript excerpt of Philipp Hancke answering the question “How do I get started with WebRTC?”. Hancke recommends learning concepts first (architecture courses), using a simple code lab to practice building a minimal peer-to-peer app, and reading free resources (e.g., “WebRTC for the Curious”) while warning that hands-on tinkering alone may not scale to multiparty or deep protocol understanding. He notes prior background (H.323, video conferencing) made learning easier and emphasizes ongoing learning and gaps in documentation.
Chapters
How do I get started with WebRTC? That's a tough one. I mean, for us, we've been doing WebTC basically since the early days of WebTRIC. So we learned it very organically. I mean, we are both coming from a related background. You from H323, I came from video conferencing and protocol stacks and media processing. So that was like, okay, so there's this new thing called WebRTC, but it's just the old thing with a new shiny API on top. And there are a few more things that you need to understand which are mostly concepts and not the whole thing. So it's a lot easier. Yes. I mean for me it was. I came from signaling and media servers, so it was like, okay, okay, this webcast, okay, I know the concepts, I applied them, but I still haven't stopped learning. But let's say you need to do it today. It's difficult. I mean, we've done the courses which give you a good framework. I mean, your architecture course is still super solid. Overview. We've done the networking courses if you are looking into the protocols. But for that I wouldn't say that's for getting started because you don't get started by implementing. We are producing yourself. You can't see, you can't see the forest from the trees. Exactly. So you need to know, you need to understand the concepts. First architecture course is great. For that we have the code lab, which is now free, teaching you how to actually write code. Because that was something that always annoyed me about the architecture because it didn't tell you how to write the code. And the code lab is basically, it's a very simple peer to peer app. Not something you would use in production because for multiparty you want to do a server. But that means you need to think about how you use APIs in a very simple, minimal way and then you can expand. That's great. We have courses, they cost money, I don't want to pay. What should I do? There is Web Procedure for the Curious, which is a free book. It's a good intro, but I'm not, I never, I wouldn't learn it that way. I read it, I think because you compare it to the protocols courses that we have that are just, let's say, a lot of details in them. Yes. Okay. And then I, what I see is a lot of people that start by building some application peer to peer, play with it, incorporate it, and then start moving and figuring out things from there. I think that it works well, but it doesn't scale into group calling or into understanding how the architecture works or how media actually flows over the network. And for that, you will need to go and learn that somewhere. So, yes, you'll have to figure that out. Yes. I mean, the problem is that you have what we also have in the codehead. You have these APIs. They work great. But. Oh, what if they don't work? Then there is no real manual there. Okay, thank you, thank you.
