W3C home > Mailing lists > Public > www-svg@w3.org > August 2012

Re: SVG masking, maskType attribute still necessary?

From: Dirk Schulze <dschulze@adobe.com>
Date: Fri, 17 Aug 2012 13:47:21 -0700
To: "Tab Atkins Jr." <jackalmage@gmail.com>
CC: "www-svg@w3.org list" <www-svg@w3.org>
Message-ID: <277FC5EE-5A70-46E1-9F90-5F8FCBD3BF4B@adobe.com>

On Aug 17, 2012, at 11:13 AM, Tab Atkins Jr. <jackalmage@gmail.com> wrote:

> On Fri, Aug 17, 2012 at 11:04 AM, Dirk Schulze <dschulze@adobe.com> wrote:
>> Hi SVG WG,
>> 
>> We added the 'maskType' attribute to differ between alpha and luminance masks on the SVG 'mask' element. IIRC we didn't thought about extending the 'mask' property. Since we added the keywords 'alpha' and 'luminance' to the 'mask' property, it seems unnecessary to add a new attribute to the 'mask' element.
>> 
>> Do we want to revert the decision on adding this attribute?
> 
> No.  The purpose of the maskType attribute is to signal what the mask
> is designed for.  Using the alpha of a mask designed for luminance is
> almost certainly a mistake and will likely produce bad and confusing
> results.
> 
> The purpose of the keyword on the property is to signal whether to use
> alpha or luminance on things *other than* <mask> elements, because
> they can't signal what they were designed for.
> 
Ah that makes sense.

> If I recall correctly, the keyword on the property is meant to have no
> effect on a <mask> element, because the maskType attribute is the
> source of truth there.
No, the property can override the attribute value. That is one reason why the default of the property is 'auto' [1].

Greetings,
Dirk

[1] http://dvcs.w3.org/hg/FXTF/raw-file/tip/masking/index.html#the-mask-type

> 
> ~TJ
Received on Friday, 17 August 2012 20:47:45 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:52 GMT