Here is WebKit's proposal for a new type of property and variable.

User Agent Properties

This specification defines an open-ended set of properties called User Agent properties, which, among other things, are used to define the substitution
value of constant() functions.

Name: (various)
Value: <declaration-value>
Initial: (nothing, see prose)
Applies to: all elements
Inherited: yes
Percentages: n/a
Media: all
Computed value: specified value with variables substituted (but see prose for "invalid variables")
Canonical order: per grammar
Animatable: no

A User Agent property is not specified in a style sheet. User Agent properties
instead define variables, referenced with the constant() notation.
For example, a page that wants to use the user's requested font sizing, and page
background:

:root {
  font-size: constant(user-font-size);
}

body {
  background-color: constant(user-background-color);
}

Unlike other CSS properties, User Agent property names are case-sensitive.

User Agent properties are not reset by the all property. If a style rule attempts to
define a value for a User Agent property, it is ignored.

Using Cascading User Agent Variables: the 'constant()' notation

The value of a User Agent property can be substituted into the value of another
property with the constant() function. The syntax of constant() is:

constant() = constant( <user-agent-property-name> [, <declaration-value> ]? )

The constant() function is used in the same manner, and follows the same rules,
as the var() function.

Defined User Agent Variables

• user-font-size: User's requested font size
• user-background-color: User's requested background color
• user-foreground-color: User's requested foreground color
• safe-area-inset-top: Inset, as a <length> from the top of the viewport to the title-safe content area.
• safe-area-inset-right: Inset, as a <length> from the right of the viewport to the title-safe content area.
• safe-area-inset-left: Inset, as a <length> from the left of the viewport to the title-safe content area.
• safe-area-inset-bottom: Inset, as a <length> from the bottom of the viewport to the title-safe content area.

The CSS Working Group just discussed dino suggestion on writing modes.

I was just talking to @heycam about this -- and one question he brought up is whether we should pull in most/all of the non-boolean media queries and allow them to be specified as variables. Do people want to write calc expressions based on things like the width media query, etc.?

Do people want to write calc expressions based on things like the width media query, etc.?

Isn’t this already possible with the vw and vh CSS units?

Would constant() work inside calc() so we could do things like this?

* {
  font-size: calc( (constant(user-font-size) * 0) + 12px);
}

Do people want to write calc expressions based on things like the width media query, etc.?

When you say width do you mean the same value as window.innerWidth? I experimented a little big with exposing various values as CSS properties to see what uses I could come up with. Things I experimented with included these global values:

Cursory experimented with exposing:

• innerWidth
• innerHeight
• cursor X and Y position

Scrollery experimented with exposing the following values:

• scrollWidth
• scrollHeight
• scrollTop
• scrollLeft

And then the most fun was exposing values related to HTML elements as CSS variables with Varsity like:

• offsetWidth
• offsetHeight
• offsetLeft
• offsetTop
• element's aspect-ratio
• number of characters of text
• number of child elements
• current value

In my experiments I would use JavaScript to supply, and keep updated, the values for these special CSS variables, and then tried to incorporate these variables in my CSS. Many of these techniques aren't things that these 'User Agent Properties' would enable, but I'm sharing some of these here because what they are, and the way they would be used seems so parallel. Perhaps the CSSplus experiments can validate the usefulness of these 'User Agent Properties' and these 'User Agent Properties' being suggested and add some merit to the idea of exposing more of these values from the browser to CSS using CSS Variables as an interface:

• Font-size based on an element's own offsetWidth
• Using mouse cursor X and Y positions as CSS variables
• Using an element's own scrollTop as a CSS variable
• Using an element's own scrollHeight to animate from height: 0; as a CSS variable
• Styling based on an element's own children.length as a CSS variable
• Scaling any HTML element ecbased on a wrapper's offsetWidth as a CSS variable
• Scaling an aspect ratio based on an element's own offsetWidth as a CSS variable
• and plenty more…

So hopefully that helps support the idea with some examples of something kind of similar, and gives an idea of what the CSS might look like when integrating these 'global CSS variables' into designs :D

I was just talking to @heycam about this -- and one question he brought up is whether we should pull in most/all of the non-boolean media queries and allow them to be specified as variables. Do people want to write calc expressions based on things like the width media query, etc.?

It's not a bad idea; I'm down for it. It would just mean we'd have to be a little careful about naming conflicts between MQs and other UA-vars, but I don't think that's a huge issue. (Alternately we could namespace, but I don't think that's necessary.)

Also: I really want to hook this functionality for custom JS things; the syntax is already super compatible. Just create a map hanging off of document or CSS or something, require the keys to be --foo form and the values to be TypedOM instances (so you get grammar-checking at parse-time, as intended), and boom, "global variables" that can be used in all locations in a stylesheet, including in MQ conditions, and which don't have unfortunate performance implications due to inheritance from root.

I had a little brainstorm about exposing JS properties of elements to CSS via CSS variables so authors can make use of a little more situational awareness when writing styles. I've been toying around with this idea for a while, but this thread about User Agent properties gave me the motivation I needed to get it published :D

There are a few fun code examples in the README and on the demo page: https://github.com/tomhodgins/varius#examples

I wonder if a value handoff between JS and CSS like this via an interface of CSS variables could help give CSS more power without needing to extend CSS features 🤔

Now that the iPhone X has been announced, the suggestion to add safe-area-insets is more clear. The idea is that a page needs to know how to avoid the rounded corners and notch at the top or side of the screen (depending on orientation).

The term "safe area" has been around for a fair while in TV, and we think it makes sense here. Since the value can change depending on orientation, it needs to be exposed directly to the style sheet - hence the proposal for CSS constants.

I think we had three major things to decide on:

1. The name: whether to use constant(name), var(name-without-leading-hyphen) or var(user name). I'd prefer constant because it makes it more clear that there is no way for the developer to set these properties.

2. A JS hook so that they can be easily queried and raise notifications on change.

3. A place to define the new properties/values. Maybe these just end up in the respective specifications clearly labelled as read-only.

a page needs to know how to avoid the rounded corners and notch at the top or side of the screen

Why would Apple deem it OK to place this burden on web developers?

@tobireif

Why would Apple deem it OK to place this burden on web developers?

We're not requiring them to do this. By default we'll avoid the notch by insetting everything. However, if the author wants their page to use the full area, they can opt in but then need to avoid the corners.

Why don't you just add a property to body or html that tells the browser that it's ok to render stuff that's normally cut off because it's outside the viewport?

render-safe-area: allow;

And then the UA can just fill up the space as needed with stuff that's normally cut off? This way the developer could just place elements outside the normally available viewport, e.g. render a header element that should extend to the edges of the screen with negative margins.

Potentially browsers that support this could add CSS units that are 1/100th's of the safe area (analogous to vw/vh).

This could be useful not only for iPhone X, but also for things like circular watch screens, etc.

Potentially browsers that support this could add CSS units that are 1/100th's of the safe area (analogous to vw/vh).

I think that is the whole point of the proposal for "Use Agent Variables", the value of the variables we are talking about would be exactly similar to variables containing the size of the safe/unsafe area.

I don't want to discourage talking about render-safe-area and other proposals at all, but in a different issue (such that the current issue can stay focused on ironing out the user-agent variables proposal).

I agree with @grorg reasoning about the default behaviour and the new proposed properties, but still, ideally neither the user, nor the developer should need to bother with this.

For colors there are already the so called System Colors, that do work in some browsers and OSes (namely IE and Firefox, never tested in Edge). The Background, Highlight and HighlightText colors also work in Windows Phone's IE and one can do a fully adaptive app theme, that mutates with the currently selected background and highlight in the OS - something not possible in iOS/Android.

If these constants are going to be implemented in all browsers, I'm all for that.

Could this help work around the non-tappable area at the bottom of the viewport in iOS Safari? As a designer of web apps I would love a way around that.

Now that the iPhone X has been announced, the suggestion to add safe-area-insets is more clear. The idea is that a page needs to know how to avoid the rounded corners and notch at the top or side of the screen (depending on orientation).

This has obvious parallels to the Round Display stuff; we should make sure these work in compatible ways. But like @FremyCompany said, this is a different issue and the specifics should be discussed in a different thread.

The name: whether to use constant(name), var(name-without-leading-hyphen) or var(user name). I'd prefer constant because it makes it more clear that there is no way for the developer to set these properties.

I also strongly prefer constant(). (I discussed in the meeting how this ties into the desire for cheaper "global variables", so authors could use constant(--name) while UA-defined ones are just constant(name))

A JS hook so that they can be easily queried and raise notifications on change.

👍 This is easy API-wise - just hang a maplike off of CSS, and make it an event emitter.

A place to define the new properties/values. Maybe these just end up in the respective specifications clearly labelled as read-only.

I'd prefer we make a new spec for this that's intended to be where most of them sit. Like Selectors is for selectors - most of them live there, but a few do live in other specs.

This could be useful not only for iPhone X, but also for things like circular watch screens, etc.

iOS 11 uses viewport-fit in the viewport meta tag to handle this, which is consistent with the CSS Round Displays working draft.

Why don't you just add a property to body or html that tells the browser that it's ok to render stuff that's normally cut off because it's outside the viewport?

And then the UA can just fill up the space as needed with stuff that's normally cut off? This way the developer could just place elements outside the normally available viewport, e.g. render a header element that should extend to the edges of the screen with negative margins.

We've done that, via a viewport tag. The problem is that we can't just extend objects - because it doesn't always look correct. Also, many objects are snapped to the edges anyway (think of a top or bottom bar), so they need to know how much padding to include in those objects in order to know they are showing content without being clipped.

After a lot of experimentation, we concluded that the best thing to do was to give authors the ability to extend to the full viewport and avoid edges if they want to, but we'd do the right thing otherwise.

Note that this discussion is orthogonal to the constant proposal. I think there is general support for that without finalizing the safe-area-inset topic.

/me dons fire retardant suit

Note that constant is partially supported in iOS 11/High Sierra, but we are very open to changing this ASAP to something everyone agrees on. The good news is that the values are always zero on existing hardware (and parsing errors mean they fallback easily on other engines)

This has obvious parallels to the Round Display stuff; we should make sure these work in compatible ways. But like @FremyCompany said, this is a different issue and the specifics should be discussed in a different thread.

Yep, agreed. As mentioned, we use the viewport-fit meta tag from round displays. One can imagine a display with rounded corners as just a version of a rounded display (or visa versa)… ignoring the notch.

Minor suggestion: const() might be a nicer name to parallel var() (since that didn't get named variable()).

For anyone curious about how this works in practice on the iPhone X (both the viewport-fit aspects from the CSS Round Displays spec and the proposed constant() values), I've written a blog post about it with screenshots showing some different cases: https://ayogo.com/blog/ios11-viewport/

I also strongly prefer constant(). (I discussed in the meeting how this ties into the desire for cheaper "global variables", so authors could use constant(--name) while UA-defined ones are just constant(name))

I have got a problem here, and my problem varies ;) More precisely, the values that we are discussing here are not going to be constants. The scrollbar width can be updated on Windows and it is very likely we will end up toggling them off for instance when you detach your keyboard and got to tablet mode.

Similarly, when you enable an high-contrast theme, "constant" colors are going to change.

Even the proposed notch-safe space will depend on the screen orientation, I assume.

All in all, I have not seen any argument to convince me we should invest in adding a new function instead of a namespace for existing value references.

TLDR: "global variables" != "constants"

Hi Dean 😃

We're not requiring them to do this.

You had written "a page needs to know how to avoid the rounded corners and notch" - that sure sounded as if each page is required to know how to avoid the rounded corners and the notch.

By default we'll avoid the notch by insetting everything. However, if the author wants their page to use the full area, they can opt in but then need to avoid the corners.

Thus on/beside each page that doesn't opt in (as the myriads of existing unmaintaned web pages ever created up to now) there will two thick white bars, correct?

https://twitter.com/thomasfuchs/status/907764896829452288

Although the white bars are stylable (albeit in a limited way it seems https://twitter.com/thomasfuchs/status/907770902338785280 ), and although though there might be CSS in the future for intentionally-fuller-area pages for dealing with the notch, ideally web developers wouldn't have to bother with Apple's notch or Apple's white bars - at least in my opinion.

@tobireif, I agree that in addition to giving authors explicit control over how the screen shape is handled, Safari on an iPhone X should by default display pages in a way that takes the notch and rounded corners into account in a reasonable fashion, so that authors are not required to make changes to have their pages properly displayed on the device.

You also seem to be making the argument that's Apple's current answer to this problem (letterboxing the page using the background color, to avoid unexpectedly clipping content in pages not designed to take into account the screen shape) is a completely unfit way to handle the problem. Why do you think that is the case, and what alternative do you suggest?

To me, clipping content by default—to avoid ungainly, unexpected letterboxing—has far more breaking potential, potentially hiding important areas of content or interactive elements. Without allowing any clipping to occur, it's hard not to end up with blank areas around the page; any ideas as to handle these better is appreciated.
Another nice point of the letterboxing-by-default approach is how nicely it folds in with the viewport-fit working draft, so that authors can work to target non-rectangle displays in general, not just iPhone X-shaped ones.

How should behave this UA Properties/Constants when writing direction is RTL instead of LTR?

safe-area-inset-right
safe-area-inset-left

How should behave this UA Properties/Constants when writing direction is RTL instead of LTR?

Those refer to the left and right of the physical device screen, regardless of writing direction

@Cykelero

You also seem to be making the argument that's Apple's current answer to this problem (letterboxing the page using the background color, to avoid unexpectedly clipping content in pages not designed to take into account the screen shape) is a completely unfit way to handle the problem. Why do you think that is the case, and what alternative do you suggest?

Ideally there wouldn't be a notch. That is my opinion. No notch, no issues 😀

The notch issue is not caused by a force of nature, but it has been added based on a decision by Apple. That is the issue.

Now that they did add the notch, it sure is sensible to limit the default viewport to the safe area (although as default color for the side-bars, black-like-the-notch-etc might be better than glowing bright white), and it also is sensible to also offer an option for using the full area (plus constants for handling the notch etc).

I'm not arguing anything. I'd like to have had a vote (or better a veto 😀 ) regarding whether Apple adds the problematic notch, but alas that decision has been set in stone already.

Now we can only deal with the issues, which I sure will do 😀
I'll probably (normally) stay inside the safe area (the default viewport) but try to style the side-bars/letterboxes in a sensible way.

For the benefit of anyone finding this looking for actionable advice on how to deal with the iPhone X viewport issues, here’s a great write-up on using viewport-fit: http://stephenradford.me/removing-the-white-bars-in-safari-on-iphone-x/

What @FremyCompany said. @tabatkins had to explain to me that there are two axes of variablilty: changing over tree context and changing over the lifetime of the document. constant() is about tree context, but “constant” connotes time. global() would be a better name.

I have got a problem here, and my problem varies ;) More precisely, the values that we are discussing here are not going to be constants.

Yeah, what eeeps said - they're constant over the page, as opposed to refs that can vary from element to element.

All in all, I have not seen any argument to convince me we should invest in adding a new function instead of a namespace for existing value references.

A big difference is that these are usable in places other than properties and animations - you should be able to use these "constants" in MQs and such. (Again, because they're constant across the page, and so don't care about what context they're being used in.) I'm loathe to say "var() is only allowed on elements normally, but you can use it wherever you want if you use this particular type of var name" - much better for everyone if it's a totally distinct name.

From http://stephenradford.me/removing-the-white-bars-in-safari-on-iphone-x/:

I added the following to the main container on the website.

padding: constant(safe-area-inset-right);  

Unless I'm mistaken, this will break when the phone is rotated. It also assumes that the right inset will be the largest, and that padding by the largest value on all sides will look fine.

To avoid this kind of error, should we have safe-area-inset that could be used in shorthand properties like margin, padding, border-width? It would be equivalent of constant(safe-area-inset-top) constant(safe-area-inset-right) constant(safe-area-inset-bottom) constant(safe-area-inset-left).

We might also consider ways to get the max/min inset value, via a generic max and min function.

It's worth pointing out that these do work in calc() expressions, so something like padding-top: calc(20px + constant(safe-area-inset-top)) works like a charm

Unless I'm mistaken, this will break when the phone is rotated. It also assumes that the right inset will be the largest, and that padding by the largest value on all sides will look fine.

The inset values adjust when you rotate, so it probably won't break on rotation, but it is indeed wrong.

To avoid this kind of error, should we have safe-area-inset that could be used in shorthand properties like margin, padding, border-width?

If it's possible to have variables/constants produce multiple values, then yes, it would help avoid wrongly over-applying a single edge's value.

We might also consider ways to get the max/min inset value, via a generic max and min function.

CSS calc(), min() and max() are super useful with this, they are already in the CSS Units and Values Level 4 Editor's Draft. https://drafts.csswg.org/css-values/#funcdef-min

Besides the use you describe, it's often useful to padding or margin to the minimum of the inset or a fixed padding for the case where there's no inset.

To me, this could be easily avoidable by having WebKit support a new meta tag that either 'respects full viewport' (so takes the notch into account) or instead on the WebKit/Safari side, just indent the corners to line up the corners and notch so there's no need to inset. Surely that's the most backwards compatible and on-breaking way forward?

This seems very overkill for supporting one platform by implementing more CSS properties that is only useful for iOS Safari (on iPhone X).

Something to the effect of:

<meta name="safe-viewport" content="render-viewport-spacing=full" />

@danieltj27 see the references to viewport-fit in this issue.

I'm really hoping this 'notch' thing is just a very short-lived phase that doesn't last. However, I do expect to see more rounded screens. It could be helpful to have the corner radius values as a relative length unit (like em and rem). Perhaps something like vcr, 'viewport corner radius' not a 'videocassette recorder' :)

margin-left: 0.5vcr;

But then again, the values for each corner could be different. Also, we should consider that a UI may want to respond differently to non-square corners than to a 'notch' protruding into the viewport.

Also, anyone know how Essential plans to treat the 'notch' in their phone's viewport?

As Dean mentioned, the concept of a "safe area" isn't specific to the iPhone X and also applies to things like TV screens. This issue is specifically about defining a way for user agents to expose those values (and others, such as user requested font size) to web pages.

Discussion around the idea of viewport-fit, rounded corners, and non-square screens is better targeted towards the CSS Round Devices spec.

Good to know. Safe area for a TV seems quite different than the iPhone X / Essential phone 'notch'.

I've filed a pull request for the constant() and UA properties part of this issue. There will be a separate pull request for the safe-area-insets (when I work out where they should go).

I'll note that the iPhone X notch isn't the only special thing that a user agent might want to tell the author to avoid. It could be an on-screen overlay button or keyboard, or a picture-in-picture video that is being played by another app on the system.

There are many reasons why the User Agent might want to tell the author of the page that content might be occluded and non-interactive in a particular region.

@othermaciej Yeah, custom properties support multiple values. Here's an ugly demo.

OK so I’ve been following this issue for a while and there are some details for which I’d like clarifications if possible. Sorry if some were addressed or implied (and cc-ing @dauwhe and @TzviyaSiegman since that will probably impact the Publishing Working Group and the personalization Task Force at some point).

The first question is quite simple: who will be able to create and set those constants?

For instance, I assume user-font-size could be set based on the iOS a11y preference or a browser setting, and user-background-color based on high contrast mode in Windows 10.

But what if you have an intermediary layer managing documents like, say iBooks or Google Play Books? They have sepia and night modes, their own font-size user setting, and a couple of extras options which OSs and browsers might not offer (text-align, hyphens, line-height, dyslexia typeface, etc.).

I guess that if the apps are using a web view, they might be able to create and set them. Correct? But what about cloud versions?

Sorry if that is unclear to me, I was kinda super happy when I discovered this proposal as it could help with user settings (both apps and authors) and I guess I need an answer as whether this could fit in there or not. To be fair, this should probably be global variables since they will change at runtime.

On a related note, PDF version 2 is using HTML and CSS to make documents more accessible, so PDF apps might offer such options at some point (and PDF is still a popular format).

Another question: how will this possibly relate to COGA Semantics to Enable Personalization?

People might not like Reading Systems (and I kinda get the reasons why), but if it proved anything, it’s that when user settings have a GUI in the app’s chrome, user will override a lot of authors’ styles (and expect those user settings to be available). On the other hand, authors tend to hate when their entire stylesheet is removed (reader modes, which sometimes fail spectacularly with more than basic documents).

Consequently, I can clearly see how User Agent properties could help, although the burden would be put on authors. Now, this proposal deals with user settings (but this section is not yet developed), so back to the first question.

@grorg IMHO unless Safari supports CSS Level 4 min/max at the same time, it will be tough to accommodate the complex positioning rules we have today.

There are many reasons why the User Agent might want to tell the author of the page that content might be occluded and non-interactive in a particular region.

I hope you mean varying safe-area-inset-* when Safari's bars come and go.

@diligiant

I hope you mean varying safe-area-inset-* when Safari's bars come and go.

Those don't tend to obscure content. They can be scrolled away, unlike the notch. What you're talking about is a better fit for visualViewport.

@jakearchibald the bottom bar does (try positioning something at 95vh). It disappears when you start to scroll but initially, you have no way to know where the visible content ends.

@diligiant this information is best exposed by visualViewport.

@jakearchibald I won’t contradict you :) Do you treat the keyboard overlay similarly?

The drawback being it can’t be handled with pure CSS.

@diligiant On Android & iOS, the keyboard just reduces the viewport size, you can still scroll down to the area that the keyboard may have obscured. This means visualViewport will tell you about it. This works in Chrome today.

When @grorg mentioned "overlay keyboards", I took that to mean something that overlays the viewport, so even if you scroll right to the bottom of the page, the keyboard will still be obscuring content. In this case constant(safe-area-inset-bottom) could give you the amount of padding you can apply to the bottom of the page to overcome this.

@jakearchibald I'll be very interested to read @grorg explanations as I have the feeling safe-area-inset-* and visualViewport are tackling different issues (highlighted by the fact that the former is css-oriented and the latter being js-oriented.)

@diligiant

I have the feeling safe-area-inset-* and visualViewport are tackling different issues

We agree!

If it's something that changes the size/position of the viewport, such as a keyboard, a URL bar, or a pinch-zoom, visualViewport is what you're looking for. It is very very rare to need to change layout as the result of a viewport change (details).

If it's something that obscures content but does not change the viewport, such as physical screen shape, overscan, a picture-in-picture overlay, or an overlaying keyboard, the proposed solution is viewport-fit and safe-area-inset-*.

See https://developers.google.com/web/updates/2017/09/visual-viewport-api#layout_viewport_vs_visual_viewport for a description of the layout vs visual viewport. The iOS & Android keyboard changes the visual viewport, whereas the notch requires the user to adjust layout so content moves out of its way.

Again, can we please move discussion about the new iPhone's notch/corners (which isn't specific to the iPhone anyway) to another issue. It's not directly germane to the general discussion of UA-defined constants/globals.

Should there be a standardized set of constant names with an expected meaning, or is that all up to the UA? I believe @grorg's pull request only defines the mechanism, but not any specific constant names.

@othermaciej I think the names Apple intends to use should be standardised. Otherwise, it's all the problems of vendor prefixes without the minor benefit of an actual vendor prefix.

FWIW the proposed naming seems good.

@diligiant

IMHO unless Safari supports CSS Level 4 min/max at the same time, it will be tough to accommodate the complex positioning rules we have today.

Yep, we agree. See https://bugs.webkit.org/show_bug.cgi?id=167000 😄

Replying to @tabatkins

I have got a problem here, and my problem varies ;) More precisely, the values that we are discussing here are not going to be constants.

Yeah, what eeeps said - they're constant over the page, as opposed to refs that can vary from element to element.

All in all, I have not seen any argument to convince me we should invest in adding a new function instead of a namespace for existing value references.

A big difference is that these are usable in places other than properties and animations - you should be able to use these "constants" in MQs and such. (Again, because they're constant across the page, and so don't care about what context they're being used in.) I'm loathe to say "var() is only allowed on elements normally, but you can use it wherever you want if you use this particular type of var name" - much better for everyone if it's a totally distinct name.

Ok, so I gave this another round of thoughts, and I agree that I can see the value in having a different function name to make it clear the places where the replacement will work are not the same. I still think this isn't needed, but I can see the value.

I really do not like the name constant(...) though. Really, there is no such thing as "constant at any given time". If you look at the definitions of "constant", you get either of these which both do not apply here (because the value of the variable can change over time):

adj: remaining the same over a period of time
noun: a situation or state of affairs that does not change

The other thing is that this is utterly verbose. If you have to use this multiple times in a calculation, having to type constant() is really annoying and space-filling, reducing gimpsability quite a lot.

I remember that when var(...) was introduced, we ruled out $... because it would conflict with css preprocessors, then also ruled out $(...) because Tab wanted to use it for things that would replace to the same thing globally.

I think this proposal matches that description: some value replacement that happens to be globally the same (but could be globally invalidated as a result of a value change). I would therefore propose we rename these variables to "globals" instead of "constants" and use $(...) to refer to them anywhere a css value is allowed.

Opinions?

Hmm... Are you at so worried that $ here could lead to even more confusion?

To me, the term constant() feels like a misnomer, especially if the rule (and property and value) can be overriden by a more specific selector.

body {
  color: constant(fontColor);
}
body.home {
  color: red;
}

It's not looking "constant" in any sense of the word, plus it's confusing on another level with var and const in JavaScript existing for similar uses, but different to each other and also different to variable() and the proposed constant().

The $() syntax doesn't feel quite like anything else in CSS. The first thing that pops into my mind would be something like global() or ua() standing for user-agent. Consider you had never heard of the functionality of exposing browser settings to CSS before, which of the following syntax examples conjures up the idea that that's what's going on in the code:

body {
  color: constant(fontColor);
  background: constant(backgroundColor);
}

or

body {
  color: $(fontColor);
  background $(backgroundColor);
}

or

body {
  color: browser(fontColor);
  background: browser(backgroundColor);
}

Given these examples, the presence of constant() without any other context makes it sound like an invariable…variable? What's that? It doesn't really make sense, if you wanted it constantly one colour couldn't you just set that in the CSS all the time? It's so wrong it feels nearly self-negating.

The second example, using $() reads to me like a layer on top of CSS - either a preprocessor variable or something that's going to be interpolated before that text becomes CSS. This is closer to what's really going on, but my first thought is not that it's coming from the browser. If I saw that in the wild my first thought would be that somebody accidentally published some kind of Sass/Less/Stylus-like preprocessor dialect without compiling it to CSS first. My initial thought would be to google what preprocessor uses $() as a syntax so I could compile it properly.

But given the third example, browser(), even if you have no idea that the functionality is possible, it's CSS-like enough that it could be a native feature, and the name browser() hints that it's nothing that could be rendered in advance…before the browser. For me at least, I think this is not only elegant enough to use while authoring CSS, but also provides the correct context to help people intuit what is going on in the code, as well as provide something 'googleable' they can look up to find out more.

That's my 2¢. This is a great idea for a feature, getting the naming right (or wrong) could be the difference between empowering authors and bewildering them.

Why not follow the idea of currentColor?

body {
    color: userForegroundColor;
}

currentColor isn't global, so it isn't really the same.

I get that, but it is a predefined value that the dev can't directly manipulate (indirectly he can in this case). The values proposed here can also change constantly (no pun intended) depending on device orientation, environment, settings, whatever, so it is merely an alias to some value defined in the UA.

The difference between currentColor and userForegroundColor is just the scope, as one uses this and the other window

EDIT: The newly proposed system-ui "font" or simply sans-serif would be another example that is, in my view, the same thing as we are discussing here. currentColor might have been the wrong example to use here because it's distracting.

Why not follow the idea of currentColor/system-ui/...?

Because a browser needs to track all these values one by one, and that is costly in terms of code to write and combinations to test. The advantage of reusing the infrastructure of var(...) is that it is generic and can be detected very easily at parse time.

The second example, using $() reads to me like a layer on top of CSS - either a preprocessor variable or something that's going to be interpolated before that text becomes CSS. This is closer to what's really going on, but my first thought is not that it's coming from the browser.

If I understood @tabatkins properly, while some of these values will come from the browser, some can also come from JavaScript using the proposed $(--my-global) syntax. If I misunderstood that then my point is moot, of course,

If I saw $(...) in the wild my first thought would be that somebody accidentally published some kind of Sass/Less/Stylus-like preprocessor dialect without compiling it to CSS first.

Well, it's because you don't know this is allowed in CSS. That's something you would learn as part of learning CSS in the future even before you start learning about SASS.

Should there be a standardized set of constant names with an expected meaning, or is that all up to the UA? I believe @grorg's pull request only defines the mechanism, but not any specific constant names.

Yes, this isn't even a question. Without standardization, this is just vendor-specific, and thus worthless. (And thus should be protected behind whatever your vendor-specific-code mechanism is: prefixes or flags or what-have-you.)

Why not follow the idea of currentColor?

Because keywords need to be tracked and monitored for collisions; there are places where we accept arbitrary keywords (like animation-name) where we explicitly handle the few built-in keywords (and thus implicitly disallow you from using them as animation names/etc). Every time we add a new built-in keyword, we run the risk of breaking author code that is currently using it as a custom keyword.

A function, on the other hand, can be handled globally without collision; we don't and will never allow user-defined functions with arbitrary names (they'll be -- prefixed like other custom things), so there's no worry about breaking author-level code.

I think this proposal matches that description: some value replacement that happens to be globally the same (but could be globally invalidated as a result of a value change). I would therefore propose we rename these variables to "globals" instead of "constants" and use $(...) to refer to them anywhere a css value is allowed.

I'm fine with this; it does indeed match the use-case I was trying to reserve $ for. It means some Syntax changes, tho. ^_^

Why the constant can't be used in calc? This does not meet expectations.

Yes, we can use height: 44px; padding-top: constant(safe-area-inset-top), but height: calc(constant(safe-area-inset-top)+44px) is conciser, and sometimes we have no choice.

Because i using calc() with no spaces around the +, constant can be used in calc.

Why the constant can't be used in calc? This does not meet expectations.

The constants do work in calc(), at least in my testing in Safari on the iPhone X simulator.

header {
  height: calc(44px + constant(safe-area-inset-top));
}

They also work in gradients:

linear-gradient(to bottom, black 0%, black constant(safe-area-inset-top), grey constant(safe-area-inset-top), grey 100%);

The Working Group just discussed User Agent properties and variables, and agreed to the following resolutions:

• RESOLVED: rename what was constant variables to environment variables using env()
• RESOLVED: Add this as an ED of variables L2
• RESOLVED: Add dino as an editor of variables L2 with TabAtkins
• RESOLVED: The initial set as safe area top, bottom, left and right

@zhouqicf Your screenshots show you using calc() with no spaces around the +; that's an error, resulting in an invalid property. constant() will have the same basic semantics as var(), so it's usable anywhere var().

I have a few questions.

First, I am presuming the request included making it clear “that there is no way for the developer to set these properties”. And now, I see interest in doing the opposite. which is to make these properties settable. I also see interest in making these properties exposable to @media.

So, if these properties are settable, how do we distinguish the non-settable ones? The ol’ double-dash?

@env --foo 0px;

body {
  margin: env(--foo);
  padding: env(safe-area-inset);
}

That makes me wonder. If env() is no longer the distinguishing quality of being un-settable, then could we modify var() to support the same functionality?

@var --foo 0px;

body {
  margin: var(--foo);
  padding: var(safe-area-inset);
}

And if that is possible, would we need @var at that point? Or would @var be needed to expose variables to @media?

Then, if these properties are assignable and also available to @media, might they create recursion?

@env --foo 0px;

@media( width > env(--foo) ) {
  @env --foo 200vw;
}

EDITED: That last example, which I accidentally left un-finished.

I notice the resolution didn't include the proposed multi-value env(safe-area-inset), is that intentional or just wasn't considered at the time?

@othermaciej #1819 (comment) – I agree with @grorg here. Although it works well for simple examples, it isn't as useful as I initially thought.

I guess we just have to watch out for anyone advocating patterns that would break on rotation, or future devices.

Perhaps it's too late, but I had a thought regarding the name choice. What about val()? It's close enough to var() that it's familiar, and has the impression that it's unsettable, and would therefore be set by some other process (the browser or what have you). It's also short.

My only issue with it would be that it is almost too similar to var(), and might be mistakenly interchanged, though I am not convinced that this would be a major problem.

We're locked and loaded to ship env(), and there was a WG resolution on that, so we'd really prefer that it not change again.

Ah, very good. Then don't mind me!

Can anyone give me guidance on how to get at these new constant/env() properties from javascript?

@jkuss This is actually quite interesting. I believe at this moment WebKit, being the only User Agent shipped with this functionality, offers no JS API to directly obtain envs in general. Interestingly, WebKit has listed some of the name strings (i.e., safe-area-inset-bottom) in CSSStyleDeclaration object at [340] ~ [343], but the values are inaccessible through .getPropertyValue function unlike any other CSS property.

Being UA-specific things, it sort of makes sense that envs belong to the unholy navigator object. But on the other hand, I could see them being an another set of custom properties; defined by the UA not by users, so there is no reason to treat them differently from vars.

I know this problem is right in the middle of a standardization minefield, and probably out-of-scope of CSSWG, but it doesn’t particularly seem in-scope anywhere else.