- From: fantasai <fantasai.lists@inkedblade.net>
- Date: Tue, 21 Aug 2007 16:55:33 -0400
- To: www-style@w3.org
Today we discussed how imprecision in device hardware results in
display widths and heights that are not *exactly* in line with the
nominal width and height. It was suggested that we should allow
some fuzz factor in media query matching for this reason. Therefore
I propose adding the following text to the spec:
(I included rules for 'aspect-ratio' in case we add them.)
Fuzzy Matching
--------------
Because devices are often not *exactly* their nominal size, this
spec allows inexact matches as follows:
* The UA should use the device's nominal width, height, and/or
aspect ratio instead of its actual width/height/aspect ratio
when matching the device-*, min-device-*, and max-device-*
media queries, if such information is available.
* The UA may match the 'resolution', 'width', and 'height'
media queries (but not the min-* or max-* variations on these
media queries) within 5% error.
* The UA may match the 'aspect-ratio' media query within 5%
error, where the error check is applied independently to the
numerator and denominator.
* For 'min-aspect-ratio' and 'max-aspect-ratio', if the actual
aspect ratio is between 1/12 and 12/1 then the UA may round
the actual aspect ratio to the nearest of:
* the nominal ratio of any paper size listed in tables 3-8
of [PWGMSN] in either portrait or landscape orientation
* any ratio {1..(2N-1)}/N where N ∈ {1..12}[1]
* the inverse of the above
before matching against the specified value. Additionally the
UA may choose to only perform such rounding in certain window
configurations (e.g. only when maximized).
* In all cases the UA may round values as appropriate to accommodate
limitations in floating-point storage and arithmetic.
[PWGMSN]
Ron Bergman; Tom Hastings. Media Standardized Names. 26 February 2002.
IEEE ISTO Printer Working Group 5101.1-2002.
URL: ftp://ftp.pwg.org/pub/pwg/candidates/cs-pwgmsn10-20020226-5101.1.pdf
[1] any ratio N/D where D ∈ {x∈ℤ|1≤x≤12} and N ∈ {x∈ℤ|1≤x≤2D}
Some changes to consider:
- Changing 5% to some other %
- Tweaking the formula in the second-to-last rule.
~fantasai
Received on Tuesday, 21 August 2007 20:55:45 UTC