W3C home > Mailing lists > Public > www-style@w3.org > December 2010

Re: [CSS21] 4.3.2 Lengths (reference pixel?)

From: David Singer <singer@apple.com>
Date: Wed, 15 Dec 2010 09:30:28 -0800
Cc: Ambrose LI <ambrose.li@gmail.com>, www-style@w3.org
Message-Id: <E751CEA0-89F4-4EFF-A2A2-453A09BCC8F7@apple.com>
To: Tab Atkins Jr. <jackalmage@gmail.com>

On Dec 15, 2010, at 1:32 , Tab Atkins Jr. wrote:

> On Tue, Dec 14, 2010 at 11:48 PM, Ambrose LI <ambrose.li@gmail.com> wrote:
>> I know everything is already decided and this is essentially a dead
>> thread. But might I say again that fixing px to 96dpi creates a px
>> unit that is not even useful. The only reason people use px is because
>> a lot of designs are constrained by sizes of images. A lot of pages
>> have JPEG banners and the text area needs to be exactly the same width
>> as the JPEG banner. If we cannot rely on 1 image pixel = 1px then I
>> don't see how the px unit serves any purpose.
> 
> Device pixels aren't the same thing as image pixels, luckily.  The
> relationship between image pixels and CSS px are defined by the
> image-resolution property, which defaults to 1dppx (1 dot per px).
> That is, each image pixel is a CSS px.


For good reason, I might add.  As screen resolution rises, and my text and vector graphics get crisper (but stays the same visual size), I do *not* want my images to shrink as they keep matching image pixel to device pixel.

David Singer
Multimedia and Software Standards, Apple Inc.
Received on Wednesday, 15 December 2010 17:31:02 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:35 GMT