W3C home > Mailing lists > Public > www-style@w3.org > August 2009

Re: Shadows vs. layout

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Mon, 3 Aug 2009 17:18:22 -0500
Message-ID: <dd0fbad0908031518o5c94fc27i57635f633445dc31@mail.gmail.com>
To: Andrew Fedoniouk <news@terrainformatica.com>
Cc: David Hyatt <hyatt@apple.com>, fantasai <fantasai.lists@inkedblade.net>, Brad Kemper <brad.kemper@gmail.com>, W3C style mailing list <www-style@w3.org>
On Mon, Aug 3, 2009 at 5:09 PM, Andrew
Fedoniouk<news@terrainformatica.com> wrote:
> Tab Atkins Jr. wrote:
>>
>> On Mon, Aug 3, 2009 at 4:22 PM, Andrew
>> Fedoniouk<news@terrainformatica.com> wrote:
>>>
>>> Tab Atkins Jr. wrote:
>>>>
>>>> To be fair, on the GUI systems you know about, the elements are
>>>> purposely placed so that the shadow doesn't try to 'spill out' of the
>>>> container.  So the question of overflow behavior doesn't come up.
>>>
>>> Window rectangle does not include shadow/outline.
>>>
>>> At least on Windows it is a bit challenging to get outline rectangle
>>> of the window that includes shadow. So pardon me but I am not buying
>>> "purposely placed so that the...". Haven't seen such purposeful attempts.
>>
>> I'm talking about within a window - windows themselves live on the
>> desktop, which is overflow:hidden.  ^_^
>
> It is *usually* overflow:hidden but sometimes it could be scrollable.
> E.g. virtual desktops on notebooks.
>
> And yet any window can be a child of some other window. You easily
> can place any top level window inside some other scrollable container
> (window).
>
> That part of discussion is really out of scope. Just trust me that
> any window can be as a child of desktop window as a child of some
> other window. Usually is a matter of single system function call.

Fair enough.

>>> The only case when the shadow is just such a background - is a part of
>>> window (read: DOM element) background itself as here
>>> http://www.terrainformatica.com/htmlayout/images/tooltip-balloon.jpg for
>>> example.
>>
>> Hmm, HTMLayout lets elements jump out of the application's window?
>> That's... weird.
>
> Jumps as in any other UA. tooltips (titles) and drop lists of <select>s for
> example. Why do you think they are not weird?

Well, first, I had forgotten that <select> and @title and such jump
out of the window.

Second, that seems like a big security risk.  The current "jumps out
of the window" items are very strictly controlled, and very distinct.
Letting arbitrary styleable elements jump out of the window seems
problematic - it could overlay important things.  That's not an issue
in an installed program, but in a webpage it may be.

> It is a matter of declaring something like
> select:active > popup { position:popup; display:block; }
>
> Usually it not so straightforward but principle is like that.
> Why tooltips or say popup menus should be anyhow different from other
> DOM elements? Styleable, manageable by scripts, etc. Quite useful and
> popular feature if you would ask people who use them now.
>
> Everything here
> http://www.softpedia.com/screenshots/Norton-Internet-Security_2.png
> including that popup window is pure markup + CSS + behaviors.
> (CSS++ I would say but still).

That is quite cool, I'll admit.  ^_^

~TJ
Received on Monday, 3 August 2009 22:19:23 GMT

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