Re: [becss] "Behavioral Extensions to CSS" computed value question

fantasai wrote:
> David Woolley wrote:

> 
> The BECSS draft already crosses this line by importing scripts through
> the 'binding' property. I haven't seen any serious discussion in the WG
> about the security implications of this.

Yes.  On a quick scan of it yesterday, that worried me.  I thought the 
position was that behaviours might use CSS selectors, but they would be 
segregated from CSS.

Could I suggest that bind be explicitly forbidden in style attributes, 
and user agents required to ignore it there.  Otherwise you are changing 
the rules under which content management/BBS systems make third party 
content safe and you will increase the pressure for methods of marking 
sections of web pages as unsafe, as recently proposed on www-html.

It is not a good idea to invalidate the presumption that CSS is 
relatively benign.



-- 
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.

Received on Saturday, 27 October 2007 20:03:42 UTC