Re: Client policy processing

how would simple client know what is negated?

regards, Frederick

Frederick Hirsch
Nokia


On May 9, 2007, at 9:29 AM, ext Anthony Nadalin wrote:

> I would say this is all optional, as the client may not have assess  
> to its own policy or the ability to actually process the policy  
> (limited device). i get worried that if we don't have absence means  
> negation that we will wind up in spots where it will be hard or  
> impossible to know the actual policy that was in effect.
>
> Anthony Nadalin | Work 512.838.0085 | Cell 512.289.4122
> <graycol.gif>
> Frederick Hirsch <frederick.hirsch@nokia.com>
>
>
> Frederick Hirsch <frederick.hirsch@nokia.com>
> Sent by: public-ws-policy-request@w3.org
> 05/08/2007 05:03 PM
>
> <ecblank.gif>
>
> To
> <ecblank.gif>
>
> ws policy <public-ws-policy@w3.org>
> <ecblank.gif>
>
> cc
> <ecblank.gif>
>
> Hirsch Frederick <frederick.hirsch@nokia.com>
> <ecblank.gif>
>
> Subject
> <ecblank.gif>
>
> Client policy processing
> <ecblank.gif>
> <ecblank.gif>
>
>
> Is it correct to say:
>
> 1. Client has access to its own policy, the provider policy and the
> result of intersection which it performed
> 2. Result of intersection is a policy in its own right, and has no
> implicit meaning other than what is stated in that policy (with its
> own vocabulary)
> HOWEVER
> 3. Client can interpret that result-of-intersection policy together
> with provider policy to infer acceptable interactions with provider,
> based on vocabulary present in provider policy.
>
> Thus the policy that results from intersection itself does not say
> negation, but it can be inferred from that policy taken in
> conjunction with provider policy.
>
> Is this an approach toward making this less confusing?
>
> Thanks
>
> regards, Frederick
>
> Frederick Hirsch
> Nokia
>
>
>
>

Received on Wednesday, 9 May 2007 16:48:19 UTC