Earlier quoted context omitted.
Why is it hard? Video streaming has been a thing for a couple decades. I am genuinely curious.
When you stream a Youtube or Netflix video, you can download several seconds ahead over boring old HTTP. It's a one-to-one link, you don't have to upload anything, and you don't need access the user's camera or microphone. To be competitive in videoconferencing these days, you need: * Low latency, so people don't talk over each other * Video and audio compression that doesn't get confused by dropped packets. * Setup…
This is not to say it is easy at all, of course; there a thousand things to do to implement it, but I think other fields face similar constraints, like videogame engines and distributed simulations.