Re: [soa-rm-ra] Policy Discussion

[Ken Laskey <klaskey@mitre.org>]

Duane,

I'm troubled.  Given

Layered Policy Architecture – an approach to place policies within 

multiple layers of an architecture.  For example, a policy may be 

placed on the access point (service interface) from which a resource 

may be retrieved, yet the resource itself may be also protected with 

a policy which must be adhered to in order to interact with the 

policy. 

How do I describe something for use when it may have hidden policies embedded anywhere in its structure?  Can I ever describe a service or must I describe every operation/action/endpoint (o/a/e)?  Does this say every o/a/e should be its own service?  Does it say the service description has to be structured in such a way that it describes details of its every o/a/e?  Does a service description basically become the structure that holds all the o/a/e descriptions?  What does this say about discovery?  Is there any value in discovering a service or do I discover o/a/e?  Does this get me back to every o/a/e is a service?

Still reading so maybe you provide some insight later in the paper, but let me get out these questions and let everyone chew on them.

Ken

On Apr 14, 2007, at 9:00 AM, Duane Nickull wrote:

Did anyone have any comments on the policy contribution?

D

<Policies ­ general concepts.pdf>

------------------------------------------------------------------------------------------

Ken Laskey

MITRE Corporation, M/S H305     phone:  703-983-7934

7515 Colshire Drive                        fax:        703-983-1379

McLean VA 22102-7508