W3C home > Mailing lists > Public > www-style@w3.org > April 2013

Re: [css3-background] "Margins are always transparent" missing

From: Brad Kemper <brad.kemper@gmail.com>
Date: Wed, 10 Apr 2013 16:51:41 -0700
Message-Id: <CCAE686F-72B8-4D5D-B3D4-870F878B63DF@gmail.com>
Cc: W3C www-style mailing list <www-style@w3.org>
To: "www-style@gtalbot.org" <www-style@gtalbot.org>
On Apr 10, 2013, at 3:39 PM, "Gérard Talbot" <www-style@gtalbot.org> wrote:

> 
> Le Mer 10 avril 2013 17:58, Brad Kemper a écrit :
>> On Apr 10, 2013, at 12:40 PM, "G¨¦rard Talbot" <www-style@gtalbot.org>
>> wrote:
>> 
>>> I still do not see where this is explicitly stated. As presented,
>>> background-clip defines (can set) the background painting area; but
>>> nowhere does it say that background (background color and background
>>> image) - by definition - of an element is never painted into its own
>>> margin area.
>> 
>> So you want to list all the places where background DOESN'T paint?
> 
> Brad,
> 
> Where exactly have I made such request? to list all the place where
> background DOESN'T paint??

You have suggested that we explicitly state that the background does not paint into the margin or change its color. But you have not presented any logic for why the background spec needs to mention this, when it also doesn't mention an infinity of other things that it doesn't paint or color. So, where does it stop, I wonder? If we need to mention that, why shouldn't we also mention all the other places where background DOESN'T paint?

> Right here, I do propose to indicate in CSS3 box model module that
> background (background color and background image) of an element never
> paints into its own margin area.

You should change the subject line then. I thought this was about the backgrounds and borders module. The B&B spec currently explains EXACTLY where the background paints. There is nothing unclear about it. It is precise in its definition of the background printing area, a term it uses to define just that. 


> But, on the other hand, we can read that
>>> "
>>> The color/pattern of the element's background extends into the padding.
>>> See the 'background' property.
>>> "
>>> CSS3 box model module, 5. The 'padding' properties
>>> http://www.w3.org/TR/2001/WD-css3-box-20010726/#the-padding
>>> 
>>> and
>>> 
>>> "
>>> The background of the element extends into the border area. See the
>>> 'background' property.
>>> "
>>> CSS3 box model module, section 6. The 'border' properties
>>> http://www.w3.org/TR/2001/WD-css3-box-20010726/#the-border
>> 
>> Right. So there is no expectation or implication ANYWHERE that backgrounds
>> paint into the margins instead of into the background painting area, as
>> defined in the latest stable background spec, currently in CR.
> 
> Your reasoning is noted. Silence and textual absence

There is not an absence of text explaining where backgrounds are painted. 

> are sufficiently
> clear to you and (your assumption) should be sufficiently clear for the
> spec readers. If it's not mentioned, then it's clear enough that it does
> not apply, it does not paint. If it does not paint the margin area, then
> it is *_not_* necessary to mention so in CSS3 Backgrounds and Borders
> module and/or to mention it in CSS3 box model module. I disagree.
> 
> 
>>> I think the clear
>>>> definitions of where things in this spec do paint are sufficient.
>>> 
>>> I disagree. Everywhere (CSS 2.1, CSS 3 Backgrounds and Borders, CSS 3
>>> Box
>>> model) I read, I see a statement missing or silence or a weakness in the
>>> spec or a perfectible diagram. The specs (CSS 3 Backgrounds and Borders,
>>> CSS 3 Box model) should promote understanding, reduce sources of
>>> confusion, misunderstanding.
>> 
>> CSS 3 Backgrounds and Borders is about Backgrounds and Borders, not about
>> margins or box-model. It really has nothing to do with defining how
>> margins work, or what gets painted or not by other properties, and it is
>> not a proper place to reiterate random things in other specs.
> 
> There is one thing about my posts in this mailing list that you ought to
> know. If test(s) submitted for inclusion into one of the several CSS test
> suites at CSSWG show a misunderstanding of how a property or property
> value or CSS feature work according to spec, then I will be checking
> carefully to see how such a property or property value or CSS feature is
> defined, explained, worded, schematically illustrated, exemplified in the
> spec. And I will often follow-up and try to propose a correction or an
> editorial insertion or better wording or corrected example when suitable.
> I have done this many times in this mailing list.
> 
> I did not start this thread about margin area and background painting
> without a triggering reason, just like that, without a
> misunderstanding/confusion somewhere.

Specs are detailed things, and should be read by anyone writing tests for them. They don't get easier to read by including more and more words explaining what has already been explained clearly and with high precision.

> 
> Gérard
> -- 
> CSS 2.1 Test suite RC6, March 23rd 2011
> http://test.csswg.org/suites/css2.1/20110323/html4/toc.html
> 
> Contributions to CSS 2.1 test suite
> http://www.gtalbot.org/BrowserBugsSection/css21testsuite/
> 
> Web authors' contributions to CSS 2.1 test suite
> http://www.gtalbot.org/BrowserBugsSection/css21testsuite/web-authors-contributions-css21-testsuite.html
> 
Received on Wednesday, 10 April 2013 23:52:20 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 10 April 2013 23:52:20 UTC