W3C home > Mailing lists > Public > www-style@w3.org > January 2015

Re: [css3-images][CSS3-UI] Default size of cursor images

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Tue, 13 Jan 2015 14:48:41 -0800
Message-ID: <CAAWBYDBjdx3VTvy5hjCHJ=CGVx-grQWm=pFrXeCAuheSuaGb=g@mail.gmail.com>
To: Florian Rivoal <florian@rivoal.net>
Cc: www-style list <www-style@w3.org>
On Tue, Jan 13, 2015 at 2:39 PM, Florian Rivoal <florian@rivoal.net> wrote:
>> On 13 Jan 2015, at 21:56, Tab Atkins Jr. <jackalmage@gmail.com> wrote:
>>
>> Actually, I misspoke, you don't need to explicitly invoke the
>> algorithm.  It's used by default by all objects unless otherwise
>> specified.
>>
>> I can update Images to point to UI rather than 2.1 for it.
>
> Right, but since 2.1 has an extra optional clause, I was a bit quick in saying you could point to css3-ui, without discussing that clause.
>
>>> As far as I can tell, the 'default sizing algorithm'[1] is compatible with the 2.1 definition for cursors, and more constraining/specific.
>>
>> Yes, 2.1 defines the same algo, except for the MAY about maintaining
>> aspect ratio.
>
> The MAY isn't about maintaining aspect ratio, it is about being allowed to shrink if you end up with something too big (and being required to maintain the aspect ratio if you do so).
>
> Or am I reading this wrong?

Nah, you're right.  I'm oversimplifying.

>>> Does any UA want to use the extra flexibility given by the 2.1 rule?
>>
>> I'd say do a quick survey of browsers, and if no one currently uses
>> it, consider it dropped.
>
> I suspect this was introduced to work around some possible OS level limitation on cursor sizes.
>
> Try it here:
> http://jsbin.com/vahora/1/watch
>
> None of FF (win8 and OS X), Chrome (Wind8 and OS X), Safari, IE 10/Win8 used the MAY clause.
>
> On WinXP & Win 7, FF, Chrome and IE9 failed to show the large (128x128) cursor entirely (despite being capable of showing smaller 16x16 cursors). IE6 did use the MAY clause, and showed a shrunk cursor.
>
> On Linux, FF displayed large cursors without problems (did not test chrome).
>
> So it looks like even though this MAY clause is not needed on latest generation systems, it is useful for legacy OSes, although few browsers take advantage of it. Keeping it seems reasonable, even if not critical. Should it go in CSS3-UI or CSS3-IMAGE?

I'm fine with keeping it, but rephrasing it to specifically reference
this corner case, and making it a MUST.  The "don't show the cursor at
all" behavior is unacceptable.

"If an operating system is <strong>incapable</strong> of rendering a
cursor above a given size, cursors larger than that size must be
shrunk to within the OS-supported size bounds, while maintaining the
cursor image's intrinsic ratio, if any."

~TJ
Received on Tuesday, 13 January 2015 22:49:28 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:51:56 UTC