W3C home > Mailing lists > Public > www-style@w3.org > April 2012

Re: [css-syntax, ideas-needed] Merge/inherit lists of values.

From: Kang-Hao (Kenny) Lu <kennyluck@csail.mit.edu>
Date: Wed, 11 Apr 2012 15:57:58 +0800
Message-ID: <4F853986.7070100@csail.mit.edu>
To: Simon Sapin <simon.sapin@kozea.fr>
CC: WWW Style <www-style@w3.org>
(12/04/11 12:35), Simon Sapin wrote:
> Le 11/04/2012 00:34, Andrew Fedoniouk a écrit :
>>> But this isn't about inheritance at all: this is about handling of
>>> >  multiple rules applying to the same element (cascading) rather than
>>> >  propagation of computed values from ancestor to descendant
>>> >  (inheritance).
>> That is debatable of course.
>> "In classical inheritance ... classes can inherit attributes and
>> behavior..."
>> src:http://en.wikipedia.org/wiki/Inheritance_(computer_science)
>>
>> Process of applying multiple matching rules can be seen as runtime or
>> dynamic or prototype inheritance - final/used style is a set of
>> properties and values
>>   inherited from all matched rules in order of selector specificity.
>>
>> But the term does not matter indeed as soon as concept is clear.
> 
> 
> Yes, you can argue that what you propose looks like the general concept
> of inheritance. I think that David’s point was that in CSS, we already
> have something named inheritance and an 'inherit' keyword, both with
> very precise meaning and behavior. You proposal is not about *this*
> inheritance, and the term is already taken. It would be confusing if
> "counter-increment: inherit" and "counter-increment: !inherit" were both
> valid but meant something completely different.
> 
> I think that this proposal addresses a real need, but it needs another
> term.

What about.... pseudo-inheritance?wwww (Just joking)
Received on Wednesday, 11 April 2012 07:58:30 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:52 GMT