You want a controversial opinion (on HN), how about mine:
"It's not always a good tradeoff to throw something under the bus to save developer time."
So many important things are routinely sacrificed at the altar of developer speed and comfort, app runtime performance being the big one. It's used to justify all sorts of crazy things we do in modern software to ship faster, but so hard to push back on because developers are expensive and software needs to be done yesterday.
Some things are worth giving up so that developers can move faster, but I argue not everything. Need to evaluate case by case.