Re: [Obe-developer] OBE Process Lifecycle

[Adrian P. <ad...@qu...>]

Dear Mina,

Thank you for your interest in the Open Business Engine, and apologies for
the delayed reply (I've been out of the country). I've inserted answers
inline below.

Cheers,

Adrian.

Adrian Price
Architect/Lead
The Open Business Engine
http://obe.sourceforge.net/



-----Original Message-----
From: obe...@li...
[mailto:obe...@li...] On Behalf Of Mina R
Waheeb
Sent: 4 December 2007 23:02
To: obe...@li...
Subject: [Obe-developer] OBE Process Lifecycle

Hi all,

   I'm new to OBE project and trying to embed the engine into
already existed  protocol. I have read the available doucmentation and
examples which makes me believe it's possible to do that but I still
have some questions:

- In process instance lifecycle when the client should be invoked to
start application?
[AMP] In XPDL, a 'user task' is typically an <Activity> with a HUMAN
<Performer> and one or more <Tool ... Type="APPLICATION"> entries. When an
instance of that activity is started, it creates a work item for each HUMAN
participant. When the human participant executes their work item through
your application's worklist page/screen/feature/Portlet (or however you
decide to expose worklist functionality to your users), the corresponding
tool agent is invoked on the client side. Typically the tool agent will
launch another local application or browser page.

I can create my own ToolAgent which invokes the client through network
protocol and return RUNNING code (is that keep the activity thread
waiting?) or the ToolAgent is designed for execute APPLICATION in the
client side and PROCEDURE in the server side, What is the best choice
to do that?

[AMP] That depends on whether the user needs to interact with the
application or not. If yes, use an APPLICATION; if not, consider using a
PROCEDURE if it is technically feasible to wrap the application code as a
procedure invocation (depends upon parameterizability).

- When the client finish the process execution and send the needed
information to the server how can i resume the process?

[AMP] The tool agent detects when the application has terminated and informs
the server. However, there's no way to know whether the work that the user
was supposed to do with the aid of the tool has actually been done, so it's
not possible for OBE to automatically mark the work item as complete. The
user is responsible for marking the work item complete. Doing so will result
in the activity getting marked complete as well, depending on the activity
finish mode and completion strategy (see the XPDL and OBE docs on these
topics respectively).

- What is WorkItem used for?

[AMP] A work item represents the assignment of an activity instance to a
workflow participant. Work items are generally presented to the user via a
worklist handler, which allows work items to be executed, suspended,
resumed, marked complete, reassigned and so on. Work items provide the means
to launch software tools, as the aforegoing discussion describes. OBE
provides a command line worklist handler and a JSF-based web worklist
handler as examples.

Thanks for OBE project community, please continue.

Thank you,
Mina R Waheeb.

-------------------------------------------------------------------------
SF.Net email is sponsored by: The Future of Linux Business White Paper
from Novell.  From the desktop to the data center, Linux is going
mainstream.  Let it simplify your IT future.
http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
_______________________________________________
Obe-developer mailing list
Obe...@li...
https://lists.sourceforge.net/lists/listinfo/obe-developer