Re: [css3-images] Features Overview

On 05/09/2011 03:01 PM, Silvia Pfeiffer wrote:
> Tab Atkins wrote:
>> The behavior of fragments should either be completely handled by CSS
>> or completely handled at a lower level.
>
> I think I agree (assuming I've understood the issue properly).
>
>> If we're using MF, we should
>> shoot for the latter, which means MF needs some additions to handle
>> all the relevant cases.
>
> So, can you propose what additional text needs to be added?

Something like:
  | For images containing multiple sizes, if a fragment identifier using pixel
  | coordinates, is first resolved into percentages using [some heuristic].

The heuristic in CSS3 Images for picking the right variant image is:

  # If an object (such as an icon) has multiple sizes, then the largest size
  # is used.
  # If it has multiple aspect ratios of that size (or of no size), then the
  # aspect ratio closest to the aspect ratio of the default object size is used.
  # http://dev.w3.org/csswg/css3-images/#default-object-size

The latter sentence is pretty arbitrary. You can pick a different arbitrary,
e.g. the ratio closest to a square is chosen, and portrait is preferred over
landscape. Or vice-versa.

The issue here is that CSS should, ideally, choose the variant that gives
the best resolution for the shape we're trying to squeeze the image into.
So we need the multiple sizes available, but they all have to represent
the same part of the image in the normal case (an icon image, of consistent
aspect ratio, with multiple resolutions packaged together).

~fantasai

Received on Monday, 9 May 2011 23:00:49 UTC