Though I guess in the extreme case a webpage could have a huge grid of 1x1px divs that change color in a requestAnimationFrame callback based on data retrieved over XHR.
So I can kinda see their point of view. It still seems a bit too far in the other direction if you ask me though.
... But enough grumbling. I wanted to point out the following documents for more context:
https://blog.chromium.org/2017/09/unified-autoplay.html (Somewhat wordsmithed...)
https://sites.google.com/a/chromium.org/dev/audio-video/auto...
https://docs.google.com/document/d/1EH7qZatVnTXsBGvQc_53R97Z...
...and it turns out I was right! "It doesn't work all the time, let's remove it." Frankly, I really really hate that attitude.
At least the option would have some effect, now it has none. That is clearly a worse situation than before.
Fuck this idiotic slow boiling of the users.
Asking on HN for a random chrome engineer using a fresh throwaway (created for this specific post) to look at your 11 hour old low priority performance bug for a specific piece of hardware seems misguided. There are hundreds of engineers working on Chrome -- it is extremely unlikely this person is the right SME for this bug, and even if they were, how could they possibly do anything to your bug internally without revealing their identity?
According to https://bugs.chromium.org/p/chromium/issues/detail?id=106364..., it was removed because it didn't actually stop sites from autoplaying video:
> full autoplay blocking is counter productive as images and <canvas> can do "video" playback just fine. We had this issue on mobile and ended up enabling muted autoplay there to avoid that issues.We found many websites having 100MB gifs that could be one order of magnitude smaller when implemented as <video autoplay muted>.
(Disclaimer: I work on Chrome at Google, but not on media.)
[1]: https://bugs.chromium.org/p/chromium/issues/list?q=media%20s...