W3C home > Mailing lists > Public > www-style@w3.org > February 2012

[css4-mediaqueries] Device units, viewing distance and occluded magnification

From: Charles Pritchard <chuck@jumis.com>
Date: Wed, 22 Feb 2012 20:08:41 -0800
Message-ID: <4F45BBC9.1000407@jumis.com>
To: "www-style@w3.org" <www-style@w3.org>
This list hit a bit of a dead-end on discussing how to represent 
physical dimensions: I'm going to widen my gaze. Now seems like a 
fantastic time to talk about device units and viewing distance.

Can we pull off a media query that says: (viewing-distance: 8ft)? When 
there's on person in the room, it's becoming feasible to judge how far 
the person is from the screen.
Does it make sense to consider "target-distance" ?

If we are considering overlays on contact lenses are steradians 
something we want to consider?

When I am animating, if I'm following WCAG, should I try to handle the 
flashing guideline: "the combined area of flashes occurring concurrently 
occupies no more than a total of .006 steradians within any 10 degree 
visual field on the screen (25% of any 10 degree visual field on the 
screen) at typical viewing distance"?
http://www.w3.org/TR/WCAG/#general-thresholddef

Is it reasonable to enable MyFancyWebApp (tm) to change its appearance 
as the user moves further away? The boundaries of UA and Web App will 
continue to shrink as we give more power to application authors. 
Usually, this kind of flexibility is left up to the UA.

viewing-distance and viewing-angle may have some interesting uses for 
motion parallax, and implicit "zoom".
logical and device pixel ratios are really important for bitmaps to meet 
the high quality of the display device.


1. We've got full screen now, an interesting semantic. When an element 
is now in full screen, it seems reasonable that CSS could do all sorts 
of strange and wonderful things.
2. We've got Microsoft's contextual zoom semantics. Pretty cool stuff 
for touch screens.
3. We've got Microsoft's logical and device pixel measurements in CSSOM 
(window.screen).
4. We have device-pixel-ratio working with CSS selectors in Mozilla's 
product.
Eh. At some point we may even have the height and width of the display 
device. It's frequently the case that the OS has this information.

There is no real consensus on devicePixelRatio. There's no real 
consensus on how to express browser zoom on the desktop.
As far as I can tell, we don't even have consensus on what (min-width: 
480px) should do.
http://www.w3.org/TR/css3-mediaqueries/

Bitmap images have a naturalHeight and naturalWidth. Canvas elements 
have a bitmap backing size, and it's usually 1:1 with the width/height 
attribute.
SVG can be rasterized onto a fixed height/width. Bandwidth matters.

Video looks better at higher resolution, but with visual media elements, 
a high resolution might cost too much in bandwidth or in cpu usage.
And it may be overkill if the screen is small.

I don't need to render my Canvas animation to meet the devicePixelRatio 
if the user is 8 feet from the screen. I could, but it's not an issue.
I don't need to render the full animation (I can use clip) when the user 
has occluded the screen through zoom, such as on a mobile phone.

Any chance of synthesizing all of this into a coherent proposal for 
media queries? I see no reason why I can't as an author, support 
projection displays, heads up displays, 3d focus tracking, and so 
forth.. I don't see why I couldn't do it efficiently, adjusting my pixel 
ratios and other such things. I'd like to gain consensus amongst vendors 
that we can setup standard names and a coherent system to fix these 
existing issues in exposing access to available information.

-Charles
Received on Thursday, 23 February 2012 04:09:04 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:51 GMT