Hi all, newbie here. Found this on the web recently, and I thought that this community might enjoy it:

This talk is the reason the community is named "Future of Coding" rather than "Future of Programming" — we didn't want to have a collision. Bret is one of the most-cited figures here, and a handful of folks from CDG and Dynamicland are here too. You've come to the right place :)

This is probably my least favourite of his talks (high bar admittedly). He doesn't really seem to dig into why these designs weren't developed. It's sort of just taken as read that we made a massive wrong turn for no particular reason

I did do a quick search for the URL before posting it 😁 , it hadn't been posted before. @Chris Knott do you have some resources I can brush up on regarding the reasons that these designs weren't developed?

Sorry I wasn't saying I knew better, just disappointed it isn't explored in the talk

Part of this video talks about how we shouldn't depend on APIs and computers should instead 'learn to talk to each other'. I've always found that statement particularly difficult to wrap my head around. Does anyone have a sense of what he's alluding to there?

Protocol discovery?

My feeling is that he's suggesting more than just a discovery protocol. Once both machines are running a common protocol, even if that protocol is used to bootstrap more sophisticated protocols, I feel like we've moved beyond talking to aliens. I don't know where to draw the line, though. Is it fair to assume both machines speak tcp? ip? I don't think Brett is suggesting computers should organically re-discover ethernet based on analog probing, but I do think he's suggesting more than just bonjour. My favorite take on it, which I don't think is an indented reading, is that computers need to share a common protocol with their neighbors, but not necessarily their neighbors neighbors.

@Chris Knott I agree about the why. Why did things turn out so differently? If we don't know, then things are liable to continue turning out differently.