[austin-cheney]

There are people who do care about that, but it’s astonishingly rare. Just consider it abandoned knowledge. Most of the people who continue to care about these things do so in open source contributions far away from their employment.

As a proof occasionally somebody will post something on HN about performance (asking for guidance, showing off a refactor, claiming performance is a critical must, whatever). I show them how to achieve load times of less than 100ms on a OS like GUI with full state restoration and the feedback is always the same. Either they can’t hire anybody who is capable of following that guidance or the juice isn’t worth the squeeze. Nothing in the guidance is challenging. It’s just not framework bullshit.

[rescbr]

> they can’t hire anybody who is capable of following that guidance

This is the main constraint when designing some solutions. Oftentimes I know the best path (or that something deemed impossible is completely doable), but it might be arcane knowledge and I would be the only one to know how it would work and be responsible to keep it working.

Bus factor of one strikes this right away, unless there are no alternatives.

[austin-cheney]

Again, it’s an economics problem. More specifically what you are addressing is not an availability problem, as everyone most commonly believes, but a selection problem.

Seriously, think about this logically. The compile target of the browser is the DOM. What is it these developers are so deathly afraid of: the DOM. Yes, it is raw emotional fear processed in the amygdala, qualified with poorly formed bullshit excuses. So, what about this induces fear? It’s not the technical challenge, because it’s not that challenging and easily taught to non-developers. I know this from experience. That’s how I know it’s a selection problem.