The 400ms rule is that if the user does something, the UI must do something, anything.
It can mean "display the linked page within 400ms" or "start the animation within 400ms". You can't expect 1 TB of data to be transferred within 400ms but you can display a progress bar within 400ms.
Apple used that to great effect. iPhones, at least in the early days always felt more responsive than Android phones. That because they hid latency using smooth animation. And it worked. It was also apparent when I compared Chrome and Firefox a few years ago. Chrome felt faster but mostly because Chrome was faster at showing something on screen. They took about the same time to fully load the page.
Also, this is a Very different scenario to the listed. The html loads in < 10ms, But I have to wait 400ms (!!!) to see it. It’s actually the exact opposite of the cases you mentioned above, they take a fast operation and dramatically reduce the perceived performance by making you wait 400ms (!!!) for it to complete.
Side note, it would be very interesting if comments were a DAG rather than a Tree... that way one could respond to several comments with the same thread.
It can mean "display the linked page within 400ms" or "start the animation within 400ms". You can't expect 1 TB of data to be transferred within 400ms but you can display a progress bar within 400ms.
Apple used that to great effect. iPhones, at least in the early days always felt more responsive than Android phones. That because they hid latency using smooth animation. And it worked. It was also apparent when I compared Chrome and Firefox a few years ago. Chrome felt faster but mostly because Chrome was faster at showing something on screen. They took about the same time to fully load the page.