W3C home > Mailing lists > Public > www-style@w3.org > April 2008

RE: [css3-mediaqueries] Is 'resolution' in terms of device pixels or CSS pixels?

From: Bonner, Matt (IPG) <matt.bonner@hp.com>
Date: Tue, 8 Apr 2008 22:58:14 +0000
To: fantasai <fantasai.lists@inkedblade.net>, Brad Kemper <brkemper@comcast.net>
CC: "www-style@w3.org" <www-style@w3.org>
Message-ID: <368F79A511563D43ADADF8B99EB82F1B40A740AF@G3W0637.americas.hpqcorp.net>
> you're better off selecting font sizes based on information other
> than the screen resolution. Size of the screen, for instance, or
> distance-to-viewers-eyes (billboards vs. handhelds).

How likely are servers going to be to know the nature of 
the UA to that extent?

For example, the IBM T221 LCD has over 200 PPI. A server could 
assume that 200 PPI meant the display must be a handheld device, 
but noticing the 3840x2200 pixel format and 22" diagonal sizes
should cause the server to think otherwise.

Similarly, the Amazon Kindle 167 PPI might suggest the same 
viewing distance as an iPhone, but the 6" screen might suggest 
something slightly farther. And if the server doesn't notice that
the display is electronic paper (i.e. reflective, bitonal) it 
still might make poor decisions about what content to send.

This problem gets perhaps worse as you consider printing. What's
the viewing distance for a B (metric A3) size print?  Depends on
the application. Figure 1200 DPI for modern printers, only a few
bits/pixel.

It seems likely that deducing viewing conditions from technical 
specs will get harder over time as more display devices hit the
market. To that extent, Boris Zbarsky's comment "let the device 
manufacturer, who knows something about the viewing distance, 
handle the conversion to device pixels..." makes the most sense.

But I suspect that still leaves web page authors wanting. It
seems like just as css3-color describes color rendering intent,
[1] it might be helpful to offer a "spatial rendering intent" 
using similar values (e.g. absolute, relative, etc) to give the 
UA guidance.

Oh, and I second the call for adding "um" for microns as a unit.

To be clear, all of these my thoughts and opinions, not HP's 
official stance, by any means.

regards,
Matt

[1] http://www.w3.org/TR/css3-color/#renderingintent
--
Matt Bonner
Hewlett-Packard Company


-----Original Message-----
From: www-style-request@w3.org [mailto:www-style-request@w3.org] On Behalf
Of fantasai
Sent: Tuesday, April 08, 2008 9:19 AM
To: Brad Kemper
Cc: www-style@w3.org
Subject: Re: [css3-mediaqueries] Is 'resolution' in terms of device pixels
or CSS pixels?


Brad Kemper wrote:
>
>
> On Apr 7, 2008, at 2:24 PM, fantasai wrote:
>
>>> Using media queries on resolution to pick font sizes is probably a
>>> bad idea, though, but it could be useful for things like deciding
>>> whether to use higher-resolution or lower-resolution (and lower
>>> bandwidth) images.
>>
>> I second that.
>
> Gee, thanks for the design advice everyone. I'm not going to argue about
> that. You roll your way; I'll roll mine.

I was referring mostly to David's other comment, actually.

I can see some cases where you might choose font sizes differently on
low-res devices, where the font size /has/ to be larger to be readable.
But I agree that in cases where the resolution is high enough that that
doesn't matter, you're better off selecting font sizes based on
information other than the screen resolution. Size of the screen, for
instance, or distance-to-viewers-eyes (billboards vs. handhelds).

~fantasai



Received on Tuesday, 8 April 2008 22:59:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:55:04 GMT