- From: Josh Tumath via GitHub <noreply@w3.org>
- Date: Wed, 21 Jan 2026 11:18:30 +0000
- To: public-css-archive@w3.org
> > Some sites have pages that are dark or light mode only. I.e. a video-streaming page or custom made pages or articles. > > My expectation here would be that those pages have `color-scheme: dark`/`color-scheme: light` set on `:root` in CSS, or have a `<meta name=color-scheme content=dark>`/`<meta name=color-scheme content=light>` in the markup. In that case, a user-requested override should have no effect here, as the author has indicated that that page only supports the one `color-scheme` they have defined. > > It is only when `light dark` is the value for `color-scheme` (in CSS on `:root` or via the meta tag) that a user can override the value passed down from the OS. Yes, but I don't think that was @johannesodland's point. 🙂 I think he's saying it would still be nice to set `<meta name=color-scheme content=dark>` and not have to change any existing CSS code like `(prefers-color-scheme)` MQs respond to the _used color scheme_ (which is Tab's original proposal). But my opinion on this now is: now that #10577 and #12681 are resolved, we have two new ways of getting the _used color scheme_. I think that we should discourage authors from using the `(prefers-color-scheme)` MQ in favour of `if(color-scheme(dark): foo, else: bar)` and `@container color-scheme(dark)`. (Even though, sadly, that means they will have to change their code.) -- GitHub Notification of comment by JoshTumath Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/10249#issuecomment-3777547060 using your GitHub account -- Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config
Received on Wednesday, 21 January 2026 11:18:31 UTC