Re: [csswg-drafts] [css-image-animation] Security Horizontal Review (#14363)

Thank you for this observation @simoneonofri. I suspect that if a global setting (if it exists) that affects image rendering were to not affect the pseudo-class, things would not be great: the expectation is that the primary use of this pseudo class is to enable authors to provide UI on to control animated images *only* on images that are actually animated.

Having images that aren't actually animated (because of a browser setting) claim to be would invite the creation of UI that does nothing, which isn't great.

Maybe the user is fine with that, because they're the one who asked for animations to be off, so they won't be confused by extra UI being made non-functional because of that, but it's not obvious that this is the case.

As far as I can tell, this has more to do with privacy than security as such, and it seems to be a tradeoff: either the pseudo class doesn't respond to the setting, in which case privacy is preserved, but there is a risk of non-functional/confusing UI, or it does, and we leak one bit of information (whether that setting exists and is on), but gain a more consistent user experience.

Unless we can find something better that preserves usability and privacy, I suggest updating the spec to make this an RFC2119 may, and explain the tradeoff?

Would that seem satisfactory?

-- 
GitHub Notification of comment by frivoal
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/14363#issuecomment-5392390092 using your GitHub account


-- 
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config

Received on Monday, 24 August 2026 08:07:38 UTC