W3C home > Mailing lists > Public > public-html@w3.org > May 2007

Re: 'role' should be property

From: Bruce Boughton <bruce@bruceboughton.me.uk>
Date: Wed, 23 May 2007 15:17:24 +0100
Message-ID: <46544CF4.6020205@bruceboughton.me.uk>
To: Dimitri Glazkov <dimitri.glazkov@gmail.com>
CC: mark.birbeck@x-port.net, public-html@w3.org

Dimitri Glazkov wrote:
> And you're mixing meaning into styling. How much worse could it get?
> If I don't get it, educate me. Or ignore me. Ah, the beauty of the
> email discussion.

There is no mixing of meaning and styling.  The original suggestion from 
Dmitry Turin did this but Mark's does not.

He suggests to borrow the mechanism of CSS and to apply it to other 
metadata, effectively creating the following hierachy of technologies:

HTML
 |
 +-- DOM
 |
 +-- CPS* (Cascading Property Sheets)
      |
      +-- CSS (Cascading Style Sheets)
      |
      +-- CRS* (Cascading Role Sheets)
      |
      +-- Some other cascading sheets language*, etc.


(NB: * denotes where I have filled in the gaps to illustrate the point, 
not concrete ideas)

I believe there is some merit to his idea that the selector mechanism 
for HTML/DOM should be separated from the semantics of the properties 
applied using it.  It remains to be seen whether other languages would 
make good use of CPS (or whatever you want to call this selector system, 
HPath, etc.).  I think this idea should be born in mind when discussing 
how to apply properties such as role to HTML.

Bruce
Received on Wednesday, 23 May 2007 14:17:55 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:44 UTC