Next in thread → Next in month →

RE: [wsbpel-uc] Continuing the scenario/use case discussion

From
Harvey Reed <>
Date
2003-07-21T21:23:17+00:00
ID
007501c34fd0$b73e2140$
Thread
RE: [wsbpel-uc] Continuing the scenario/use case discussion
Comment#1: I forgot to mention that the Simpl-eb
doc is a good effort to reuse simple use cases, and they are already Supply
Chain based. I think the feeling with those, is that we should convince
ourselves that BPEL can at least tackle the simple stuff, before trying to
tackle the very complicated. We didn’t have time to discuss leveraging
this work during the con-call. Rajeev contributed this doc, and John posted the
doc in the use case area.

 

++harvey

 

-----Original Message-----

From: Harvey Reed
[mailto:] 

Sent: Monday, July 21, 2003 5:34
PM

To: 

Subject: [wsbpel-uc] Continuing
the scenario/use case discussion

 

All,

 

I posted the updated proposal, with
people’s cautions and observations, to the use case doc area (version 2).

 

Below are points that I believe we
all agreed to. However, before updating the proposal, I wanted to put this to
the group for discussion in this email thread:

 

1.     
Scenarios contain technical use cases, and (I forget
who) will propose a definition of both (current action item from con-call).

2.     
The Supply Chain metaphor is a good starting point
for a set of scenarios, as long as we assure that we don’t forget the
Enterprise itself.

3.     
Scenarios will start at the abstract process level,
and continue to the executable process level. This may necessitate two related
artifacts.

4.     
The scenarios (especially starting at the abstract
process level) will provide a broad context for embedding specific technical
use cases.

5.     
We can choose which specific technical use cases to
embed. We don’t need to cover everything, however we do have the
opportunity to choose use cases based on:

a.      
Illustrate suspected gaps, or awkwardness in the BPEL
spec

b.     
Illustrate appropriate boundary of the BPEL spec, in
terms of

                                                              
i.     
Capability

                                                            
ii.     
Interface to related spec (does this cause unintended co-dependency?)

6.     
The scenarios can be made vertical when we need them
to be.

 

Let the comments, critiques, and
discussions begin!

 

++Harvey
Next in thread → Next in month →