Deserves a follow up article for “how software companies should actually be run to avoid promotion seeking behaviors like projects > all else”
Most software companies just need 1 (or at most a handful) of really amazing products.
Most products are composed of a handful of “projects”.
I’ve worked at big and small tech companies where we were so busy building projects for projects for projects with no correlation to customer value. Things like a custom UI snapshot testing suite.
I do think Google, Meta, and Microsoft have the budget to work on such projects that ultimately benefit the whole tech community. And it’s a tide that raises all ships (TypeScript, React, and Basel come to mind but there are many more examples). But if you’re anybody the size of Lyft or smaller, you probably just want a small team of engineers hyper-focused on building an amazing product without all the “projects”.
For example, what all the 4k+ engineers at Spotify are working on every day, I have no earthly idea. It’s not just a waste of money, it’s an obstacle to genuine progress and innovation.