W3C home > Mailing lists > Public > www-style@w3.org > November 2005

Re: Why Binding Scripting in Style Layer Conflates Semantics

From: Shelby Moore <shelby@coolpage.com>
Date: Thu, 24 Nov 2005 02:48:18 -0500 (EST)
Message-ID: <4827.203.168.9.127.1132818498.squirrel@webmail6.pair.com>
To: www-style@w3.org

Andrew Fedoniouk wrote:
> I didn't find anything there about "graceful fallback" .
[snip]
> To be able to do any fallback UA must be told about class and base class
> of the <mapselect>.

I gave the answer already in my previous post:

<select
  xmlns="http://schemas.microsoft.com/winfx/avalon/2005"
  xmlns:x="http://schemas.microsoft.com/winfx/xaml/2005"
  x:Class="MyNamespace.MapSelect">
</select>

Above you have <select> that has been subclassed into
x:Class="MyNamespace.MapSelect".

The code is implemented in .Net (C# example) as:

namespace MyNamespace
{
  class MapSelect : select
  {...}
}

I am not sure if .Net has classes for (x)HTML tags yet, but this above is
for architectural point.


>  This can be accomplished by a) supplying
> namespace specific to the domain (a.k.a. type/style sheet) or b)
> by using element attributes.
>
> So <mapselect> will get something like
>
> <mapselect type="select">


That would break legacy UA.


>
> At this point I would like to know why it is better in principle than:
>
> <select type="mapselect" />


Yes that is better.  And that is analgously what we have above in XAML
example.



> -or-
>
> mapselect { behavior: select;  }
> <mapselect />


No.  If that is style layer binding, then you are conflating style and
semantics again.  This group seems to have a habit of wanting to put
everything in the style layer.  CSS is cool (selectors, cascade, etc, but
it is not the semantic layer.

Whereas, if that is just a new syntax for binding in semantic markup
layer, then I would argue it is inferior because of breaking legacy UA and
for other reasons.  If you want to get into a debate about which syntax is
best for binding subclassed semantics, then please take it off list to me,
and we can come back to list later to summarize our debate.  But in short,
an XML schema is best.


-- 

Kind Regards,
Shelby Moore
http://coolpage.com
Received on Thursday, 24 November 2005 07:48:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:41 GMT