You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by Warren Janssens <wh...@yahoo.ca> on 2001/11/16 23:32:57 UTC

workflow

I was just curious how the workflow in sandbox relates
(if at all) to jsr 94 and/or ruleml?

_______________________________________________________
Build your own website in minutes and for free at http://ca.geocities.com

--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>


Re: "Craig R. McClanahan"

Posted by "Craig R. McClanahan" <cr...@apache.org>.

On Tue, 20 Nov 2001, Warren Janssens wrote:

> Date: Tue, 20 Nov 2001 21:21:27 -0500 (EST)
> From: Warren Janssens <wh...@yahoo.ca>
> Reply-To: Jakarta Commons Developers List <co...@jakarta.apache.org>
> To: Jakarta Commons Developers List <co...@jakarta.apache.org>
> Subject: "Craig R. McClanahan" <cr...@apache.org>
>
> That's probably for the best that it is more
> procedural than declaritive -- I know that I don't
> really want to relearn prolog-style programming.
>
> Anyways, I guess my interest in a workflow engine
> would be for things like approval chains (you know --
> first this person or group approves, then the next
> person, etc...) since this thing tends to crop up in a
> lot of business applications.  Had you envisioned the
> non-web part of the framework being able to handle
> something like this.  If so then I think I wouldn't
> mind contributing since I know I'll have to write many
> more applications that need that during my career.
>

Although the use case you describe is very interesting, the current
Workflow implementation does not really address it yet (other than
reserving the interface name org.apache.commons.workflow.Process :-).  The
current stuff around "Activity" is mostly related to things that are
basicially synchronous, done in a short time frame, by a single individual
-- the "Process" notion is mostly a TODO.

That being said, I'd be very interested in collaborations to make the
Process notion come to life ..

Craig


> --- "Craig R. McClanahan" <cr...@apache.org> wrote:
> >
> >
> > On Fri, 16 Nov 2001, Warren Janssens wrote:
> >
> > > Date: Fri, 16 Nov 2001 17:32:57 -0500 (EST)
> > > From: Warren Janssens <wh...@yahoo.ca>
> > > Reply-To: Jakarta Commons Developers List
> > <co...@jakarta.apache.org>
> > > To: commons-dev@jakarta.apache.org
> > > Subject: workflow
> > >
> > > I was just curious how the workflow in sandbox
> > relates
> > > (if at all) to jsr 94 and/or ruleml?
> > >
> >
> > I don't know much of the detail behind JSR 94, but
> > from the description it
> > looks like this is focused on a "declarative" style
> > of locating rules that
> > match a particular business situation.  In a similar
> > vein, RuleML seems to
> > be more focused on the inference engine space.
> >
> > The Commons Workflow package is more "procedural" in
> > nature, defining the
> > actual steps that should take place and not worrying
> > so much about how you
> > decide which Activity is relevant at this point in
> > time.  My initial
> > primary use case was relatively low level
> > (automating the navigation of
> > wizard-style dialogs in web applications), and the
> > generic workflow design
> > grew out of those needs -- such as generalizing the
> > concept of Scopes
> > instead of hard-wiring it to the Servlet concepts of
> > request, session, and
> > applicaiton scope.
> >
> > Finally, it's worth noting that the Commons Workflow
> > package provides an
> > XML syntax for defining the Steps of an Activity,
> > but using it is not
> > required -- you can use any mechanism you wish to
> > construct the in-memory
> > object trees for each Activity.  If you use the XML
> > syntax, the concept of
> > using a namespace for rule sets is somewhat similar
> > to what RuleML does.
> >
> > Craig
> >
> >
>
>
> _______________________________________________________
> Build your own website in minutes and for free at http://ca.geocities.com
>
> --
> To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
> For additional commands, e-mail: <ma...@jakarta.apache.org>
>
>


--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>


"Craig R. McClanahan"

Posted by Warren Janssens <wh...@yahoo.ca>.
That's probably for the best that it is more
procedural than declaritive -- I know that I don't
really want to relearn prolog-style programming.

Anyways, I guess my interest in a workflow engine
would be for things like approval chains (you know --
first this person or group approves, then the next
person, etc...) since this thing tends to crop up in a
lot of business applications.  Had you envisioned the
non-web part of the framework being able to handle
something like this.  If so then I think I wouldn't
mind contributing since I know I'll have to write many
more applications that need that during my career.

--- "Craig R. McClanahan" <cr...@apache.org> wrote:
> 
> 
> On Fri, 16 Nov 2001, Warren Janssens wrote:
> 
> > Date: Fri, 16 Nov 2001 17:32:57 -0500 (EST)
> > From: Warren Janssens <wh...@yahoo.ca>
> > Reply-To: Jakarta Commons Developers List
> <co...@jakarta.apache.org>
> > To: commons-dev@jakarta.apache.org
> > Subject: workflow
> >
> > I was just curious how the workflow in sandbox
> relates
> > (if at all) to jsr 94 and/or ruleml?
> >
> 
> I don't know much of the detail behind JSR 94, but
> from the description it
> looks like this is focused on a "declarative" style
> of locating rules that
> match a particular business situation.  In a similar
> vein, RuleML seems to
> be more focused on the inference engine space.
> 
> The Commons Workflow package is more "procedural" in
> nature, defining the
> actual steps that should take place and not worrying
> so much about how you
> decide which Activity is relevant at this point in
> time.  My initial
> primary use case was relatively low level
> (automating the navigation of
> wizard-style dialogs in web applications), and the
> generic workflow design
> grew out of those needs -- such as generalizing the
> concept of Scopes
> instead of hard-wiring it to the Servlet concepts of
> request, session, and
> applicaiton scope.
> 
> Finally, it's worth noting that the Commons Workflow
> package provides an
> XML syntax for defining the Steps of an Activity,
> but using it is not
> required -- you can use any mechanism you wish to
> construct the in-memory
> object trees for each Activity.  If you use the XML
> syntax, the concept of
> using a namespace for rule sets is somewhat similar
> to what RuleML does.
> 
> Craig
> 
> 


_______________________________________________________
Build your own website in minutes and for free at http://ca.geocities.com

--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>


(Digester)[PATCH] Bug fix and enhacement

Posted by robert burrell donkin <ro...@mac.com>.
this patch contains a bug fix and an enhancement.

(bug fix) the addFactoryCreate method now sets the digester property

(enhancement) i've added a property which gives public visibility to the 
name of the xml element currently being processed.


Re: workflow

Posted by "Craig R. McClanahan" <cr...@apache.org>.

On Fri, 16 Nov 2001, Warren Janssens wrote:

> Date: Fri, 16 Nov 2001 17:32:57 -0500 (EST)
> From: Warren Janssens <wh...@yahoo.ca>
> Reply-To: Jakarta Commons Developers List <co...@jakarta.apache.org>
> To: commons-dev@jakarta.apache.org
> Subject: workflow
>
> I was just curious how the workflow in sandbox relates
> (if at all) to jsr 94 and/or ruleml?
>

I don't know much of the detail behind JSR 94, but from the description it
looks like this is focused on a "declarative" style of locating rules that
match a particular business situation.  In a similar vein, RuleML seems to
be more focused on the inference engine space.

The Commons Workflow package is more "procedural" in nature, defining the
actual steps that should take place and not worrying so much about how you
decide which Activity is relevant at this point in time.  My initial
primary use case was relatively low level (automating the navigation of
wizard-style dialogs in web applications), and the generic workflow design
grew out of those needs -- such as generalizing the concept of Scopes
instead of hard-wiring it to the Servlet concepts of request, session, and
applicaiton scope.

Finally, it's worth noting that the Commons Workflow package provides an
XML syntax for defining the Steps of an Activity, but using it is not
required -- you can use any mechanism you wish to construct the in-memory
object trees for each Activity.  If you use the XML syntax, the concept of
using a namespace for rule sets is somewhat similar to what RuleML does.

Craig



--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>