[css-viewport] [css-device-adapt?] Really create a css-viewport spec

[emilio]

In #4766 (comment), we resolved to create a new CSS Viewport spec, to host the meta viewport stuff at least, but it doesn't seem to have happened yet.

Given there's ongoing effort in improving interoperability in this area, and have ICB / visual / layout viewports more precisely defined (see issues in https://github.com/web-platform-tests/interop-2022-viewport etc), it seems like finally doing this (even if it's mostly placeholders / issue references for now) would give us a good place to put the results of that interop effort.

Is there any official resolution / process we'd need to do to get this started?

@bramus has been leading a lot of this investigation, but I'd be happy to co-edit or so if he or some of the other folks in the cc list are not available.

cc: @smfr @dcrousso @dlibby- @skobes @hiikezoe @theres-waldo

[css-meeting-bot]

The CSS Working Group just discussed [css-viewport] [css-device-adapt?] Really create a css-viewport spec, and agreed to the following:

• RESOLVED: rename css-device-adapt css-viewport; Emilio will become editor and figur eout the details of renaming/updating

The full IRC log of that discussion

<emeyer> Topic: [css-viewport] [css-device-adapt?] Really create a css-viewport spec
<emeyer> github: https://github.com//issues/7590
<emeyer> emilio: We resolved on this when removing @Viewport, but it never happened
<emeyer> …Now that we’re trying to define interactions between various orders, I was talking with Bramus and we don’t have a great place to put this sort of thing
<emeyer> …A viewport spec seems like a natural place to do it
<emeyer> fantasai: I was thinking that reusing css-device-adapt makes a lot of sense to me
<emeyer> emilio: Should we just rename css-device-adapt to css-viewport?
<emeyer> fantasai: I would just run with what we have, but we can rename it, that’s fine.
<emeyer> emilio: I think css-viewport is a more findable name
<smfr> +1
<emeyer> Rossen_: proposal is to name viewport spec css-viewport and let go of css-device-adapt
<bradk> +1 to css-viewport
<emeyer> fantasai: We could take everything we don’t want out of css-device-adapt and continue with that
<fantasai> s/and continue/, rename it, and continue/
<emeyer> Rossen_: I don’t have a strong preference on how we start the new one, whether we rename an old spec or start a new one
<fantasai> fantasai: or we could kill css-device-adapt and start a new spec called css-viewport
<emeyer> smfr: Do we think this would ever become viewport-plus-other-stuff?
<fantasai> fantasai: depends if we want the continuity, or having a dead-end spec
<emeyer> Rossen_: I don’t want to force a resolution but it sounded like we were getting ready to resolve on css-viewport
<emeyer> smfr: I’m fine with viewport, just worried about the future
<emeyer> Rossen_: We can keep this one viewport-specific
<emeyer> RESOLVED: rename css-device-adapt css-viewport; Emilio will become editor and figur eout the details of renaming/updating

[tobiasschmidt89]

Hello,

is @viewport at-rule in CSS part of this spec/topic?

For users that author websites in HTML this would be amazing to avoid setting the <meta> tag on each HTML file.

Is there anything I can contribute to this as someone with no education on browser engine level code/spec/functionality?

Thank you and kind regards
Tobi

[cvrebert]

@tobiasschmidt89 The at-rule has already removed from CSS per #4766.
I'm doubtful that removing 1 line of boilerplate from non-templated HTML would be a persuasive argument.

[tobiasschmidt89]

Thank you for the link @cvrebert. Super interesting discussions on this topic.

Now I feel 50/50 about this.

On the one hand, it sounds like a small gain for the effort, especially since the last attempt was canceled.

On the other hand it feels like an opportunity to improve a fundamental configuration many modern websites have. I guess it could help with following use cases:

• separation of concerns between HTML/content and CSS/design.
• keeping HTML DRY and help users that author websites in HTML.
• avoid a human error where developers create responsive CSS, but forget to add the meta tag in a HTML template (also being able to add this to a normalise/reset CSS file.)
• flexibility for developers and users to enable user level stylesheets and developers that do not have access to HTML templates to fix the viewport.

I feel like the "correct" thing would be to have a CSS at-rule, but it might not be pragmatic.

I am super interested in the opinion of others on this.

Is this issue the right place to discuss this or should I use another one/create a new one?

Thanks
Tobi

[emilio]

This has been done (https://drafts.csswg.org/css-viewport-1/ exists) so this issue should be closed. Let's discuss that in a new issue :)