Re: Scaling icons
- From: Owen Taylor <otaylor redhat com>
- To: Bill Haneman <bill haneman sun com>
- Cc: gtk-devel-list gnome org
- Subject: Re: Scaling icons
- Date: 20 May 2003 14:10:27 -0400
On Mon, 2003-05-19 at 13:33, Bill Haneman wrote:
> Thanks for replying to my rather ill-formed concern.
>
> It sounds as though the new proposals are compatible with the
> scaling/theming needs of accessibility, and the burden of scalability
> lies with theme authors (i.e. they should provide scalable versions of
> icons even if they wish to include fixed-size icons for cosmetic
> reasons).
>
> One scenario still is unclear to me; if a theme designer has included
> both scalable and non-scalable icons, how can we facilitate both the
> high-sharpness display that the theme designer presumably desired and
> allow low-vision users to forcibly scale those icons in situations where
> size is more important than sharpness? It seems to me that fixed-size
> icons inherently defeat accessibility since in order for them to have
> meaning, there can't be equivalent scalable versions of the same icons
> in the theme.
>
> For instance, if I have a 24x24 icon that I don't want viewed in another
> size, it seems that I can't include scalable versions at, say, 48x48 and
> 18x18 to cover the full range of sizes, because the scalable version(s)
> would be then used for all sizes other than 24x24!
>
> If this is so - i.e. if fixed-size icons are desirable to theme authors,
> and using them effectively requires that scalable equivalents are not
> available, it seems to me that we have an accessibility problem unless
> we provide an option to "force" scaling of icons that the theme has
> suggested are "fixed size". What am I missing here?
I'm not sure I completely understand your concern, but something like:
size 24, threshold = 2 foo-24.png
scalable foo-48.png
will work fine to use the foo-48.png at all sizes other than 22-26.
It can get a lot more complicated than if desired; I really
can't think of a situation that you couldn't handle with the
current specification.
Regards,
Owen
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]