Re: Error in diagram?

Umit,

The diagrams look correct to me. Sanjiva replied already in 
http://lists.w3.org/Archives/Public/www-ws-desc/2003Jun/0073.html but I'll 
add a few comments.

The diagram does not reflect the syntax of WSDL. For example, a resource 
may use an interface, but the <interface> does not point to the resource. 
The <interface> is abstract and can be used by many resources. The line 
from interface to resource is a "uses" relation. 

Similarly, you should not infer that because <service> has an 
@targetResource attribute that there should be a line from service to 
resource. However, the diagram does show that if you start with a service 
you can trace back to the resource so service and resource are related 
functionally, hence the @targetResource attribute.

As I mentioned at the f2f, the diagram is inspired by COM. In COM, you 
access objects through interfaces. Diagrammatically, you draw an object as 
a box and add a "lollipop" for each interface available on that object. 



The object is a black box, just like a Web resource. All you can do is 
send messages to it. You have no idea how it's implemented. The circles on 
the lollipops act like terminals that receive signals on a device. The 
main difference here is that we have the further distributed computing 
complications of protocols and URIs (bindings and endpoints).

Arthur Ryman





Umit Yalcinalp <umit.yalcinalp@oracle.com>
Sent by: www-ws-desc-request@w3.org
06/19/2003 04:02 PM

 
        To:     David Booth <dbooth@w3.org>
        cc:     www-ws-desc@w3.org
        Subject:        Re: Error in diagram?

 




David Booth wrote:

>
> It occurs to me that figures 1-1 and 1-2 in our draft[1] that indicate 
> "service" and "resource" are incorrect.  They show the _interfaces_ 
> pointing to the "resource", whereas the targetResource attribute 
> allows the WSDL _service_ to point to a resource.  So the line should 
> be drawn from the _service_ to the resource -- not from the 
> _interface_ to the resource.
>
> Did I miss something?
>
> 1. 
> http://dev.w3.org/cvsweb/~checkout~/2002/ws/desc/wsdl12/wsdl12.html#intro_ws 
>
>

Nope, this is exactly what I was commenting about on the concall today. 
We first had the picture as you described but I am no longer clear why 
that is "inverted" as it looks like interfaces are sharing the resource 
but the resource concept is not available at the abstract level.

Arthur, since you drew this final diagram, perhaps you could clarify why 
we did this.


--umit


-- 
Umit Yalcinalp 
Consulting Member of Technical Staff
ORACLE
Phone: +1 650 607 6154 
Email: umit.yalcinalp@oracle.com