RE: [wsbpel] Example for how compensation relates to WS-T (BA)

["Satish Thatte" <satisht@microsoft.com>]

These are all great questions.  Let us start with interoperability,
before we sink into the morass of minimality ;-)

I believe that a combination of abstract processes with WSDL (1.1 or
1.2) should provide an elementary level of predictability of external
behavior that I would characterize as the focus of interoperability.
Here we need to find some way (including another dependency) to address
issues regarding policies for reliability and security, among others,
although the protocols required are defined elsewhere.

For both executable processes and abstract processes, we will need an
operational semantics that unambiguously defined behavior without
relying on interpretations of the English text in the spec.  This is the
other side of predictability, not of externally facing behavior but of
executable behavior.

Satish

-----Original Message-----
From: Assaf Arkin [mailto:arkin@intalio.com] 
Sent: Friday, June 20, 2003 1:33 PM
To: edwink@collaxa.com
Cc: Satish Thatte; 'Yuzo Fujishima'; wsbpel@lists.oasis-open.org
Subject: Re: [wsbpel] Example for how compensation relates to WS-T (BA)

Edwin Khodabakchian wrote:

>Assaf,
>
>I think that we all agree that although this might be clear in a few
>people's mind, there is room for interpretation and for the sake of
>interoperability the spec needs further clarification on that specific
>subject. 
>
>I am suggesting that we start by defining a simple example as a way to
>identifying the areas that need further clarification within the spec.
>
>Edwin
>  
>
I think that writing a simple example is a great idea. It would 
definitely help the discussion.

The point I'm trying to make is that we need a bit of focus here before 
we can act as gatekeepers to accept/reject changes to the specification.

This also goes back to <sequence> yes or no?

I assume we all agree that we're not going to boil the ocean and we're 
going to leverage other works as much as possible. But we still need to 
clarify a few points before we can start entertaining some issues and 
strive to offer resolutions:

1. Is interoperability a requirement, a goal or a nice to have?

2. Can interoperability be inferred from interface or strictly from 
service definition?

3. Aside from execution what other requirements do we want to meet 
(modeling, notation, XML authoring, etc)?

4. Do we have some criteria along the lines of 
least-amount-of-constructs or least-complexity-of-specification or 
least-constructs-to-do-X/Y/Z?

5. Are we interested in addressing coordination between services?

6. Are we interested in addressing reliable messaging?

7. WSDL 1.1 or WSDL 1.2?

8. Can we make certain technological assumptions (e.g. use of 
WS-Addressing for most usage of correlations)?

arkin