What an amazing selling point!
It requires lots of independent thinking (which I value) from founders to refrain from something everyone else is hitting the gas pedal on.
It requires lots of independent thinking (which I value) from founders to refrain from something everyone else is hitting the gas pedal on.
Just look at the amount of excitement in their industry right now. New IDEs are springing up practically by the week. Are any of these new IDEs based on Zed?
No, of course not. GPL closes the door on that, and who would want to own a fork of a tool that's so painful to develop? Developers working on VSCode can probably do an average dev-test cycle in 1-5 seconds. I've no idea how long the Zed dev-test cycle is, but I hear Rust builds are notoriously slow so I assume it's more like 1-5 minutes.
Even without hotpatching, Rust incremental compile times are on the order of 10 seconds, not multiple minutes. Clean compiles still take minutes, but that's a "once-a-week" workflow usually.
Not sure if Zed has gotten it hooked up yet, but things are changing rapidly in the Rust GUI (and games) space because of Dioxus's innovation here.
It doesn't really change my multitude of problems with Rust as a language for a project like Zed. Zed feels like a cathedral to me in the Eric S Raymond sense. Rust feels like a language for building cathedrals. Javascript is the language of the commons, which is why the idea of a hackable IDE in JS was so brilliant. Somewhere along the line the team forgot what made Atom great (despite its flaws).