Death by open source

Microsoft open sourced VS Code. Now Cursor and Windsurf are the better products.

Google open sourced Chrome. Now Comet is emerging.

I’m biased and an early adopter of new tools, so take my predictions with a grain of salt.

Some say distribution is going to be hard for new tools. But arguably Cursor is sufficiently distributed to the people who will matter in the next 5 years. Nearly every startup founder I know uses Cursor. And Cursor already pushes the upper bound of usefulness for an IDE. But then, the early adopters I know have already moved on to Claude Code. But that’s arguable a different form factor and Cursor has already won the IDE form factor.

For Microsoft, a better VS Code isn’t existential. While Microsoft doesn’t make any revenue off of VS Code, it’s provides good network effects for Github and Azure. People who use Cursor still use GitHub, so Cursor doesn’t really harm Microsoft’s existing revenue; but rather Microsoft’s future revenue from coding AIs.

For Google, an open source Chromium based competitor hasn’t really reached critical mass yet. Currently, I can’t imagine a killer use case for AI in a browser. But I didn’t predict that cmd+[i,l,k] could sufficiently grow a new code editor to critical mass. Computer use would be an obvious prediction. Maybe a browser that tests changes to my codebase and then takes screenshots of the changes would be a killer feature for me (a dev). But I can’t currently think of a killer consumer use case.