Re: Exposing the critical ACTUAL style values?

At 12:35 PM 12/19/2002 -0500, TonyT wrote:
>I understand this better now, thanks.
>
>.  It was already proposed as Views and Formatting spec in the DOM
>> section at W3C.org.  But they had proposed a new abstract model where
you can
>> ask questions about entities which do not exist in markup, such as
>> characters.  I argued that instead best to expose rendered values of the
1:1 corresponding
>> properties of markeup else the logic below applies...
>
>
>I forsee such an abstract model opening the door to browser security issues,
>or possibly conflicting w/ existing security precautions/methodology used by
>browser.  It seems that this type of model would have to be specially
>sandboxed?

Interesting point.  I do not visual the security vulnerability, but that is
a point.  When you add a new layer, you have to rethink the security
ramifications.

I have cc: the mailing list.

-Shelby Moore