Re: [wsbpel] Issue 24 - separate schemata for abstract and executable processes

["Jim Clune" <jim@parasoft.com>]

Yes, I agree with that. Here are my thoughts on this.

Separating the schemas is better for answering the questions:
 - Is this a valid executable BPEL process?
 - Is this a valid abstract BPEL process?
Using a single schema is better for answering the question:
 - Is this a valid BPEL process?

If the consensus is that the first two questions are more commonly asked in
practice than the last question, I'm in no position to dispute that. I do
think that determining whether or not a BPEL process is valid (independent
of whether it is abstract or executable) is more awkward in the separate
schemas case than answering either of the first questions with the aid of a
single schema.

Jim Clune
Parasoft Corporation          email: jim.clune@parasoft.com
101 E. Huntington Ave.      voice: (626) 256-3680
Monrovia, CA.  91016           fax  : (626) 305-9048
----- Original Message ----- 
From: "Danny van der Rijn" <dannyv@tibco.com>
To: <wsbpel@lists.oasis-open.org>
Sent: Tuesday, December 16, 2003 10:29 AM
Subject: Re: [wsbpel] Issue 24 - separate schemata for abstract and
executable processes


> while i don't have strong feelings about this, it seems to me that the
> advantage you point out is also a disadvantage - it makes it (slightly)
> harder to tell what kind of valid schema you have - is it abstract or
> executable?  the question i have is which is the more prevalent use case?
>
> will it be more often the case that i want to allow either abstract or
> executable, or will it be more often the case that i want one or the
other,
> but not both?
>
> comments?
>
> ----- Original Message ----- 
> From: "Jim Clune" <jim.clune@parasoft.com>
> To: "Danny van der Rijn" <dannyv@tibco.com>; "John Evdemon"
> <jevdemon@microsoft.com>; <wsbpel@lists.oasis-open.org>
> Sent: Monday, December 15, 2003 9:23 AM
> Subject: Re: [wsbpel] Issue 24 - separate schemata for abstract and
> executable processes
>
>
> > I like the idea of introducing tighter type-checking by distinguishing
> > between abstract and executable processes in the schema. However, I'm
> > wondering if this would be more elegantly solved by disambiguating the
> types
> > where appropriate within a single schema. The current schema has a
single
> > top-level element definition named process of type bpws:tProcess.
Perhaps
> > this could be replaced by an xs:choice definition that has as its
choices
> > xs:elements of type bpws:tProcessAbstract and bpws:tProcessExecutable. A
> > similar naming convention could be used for the activities, and elements
> > that have identical constraints for the executable and abstract versions
> > could use type names without the "Abstract" or "Executable" suffix.
> >
> > A potential advantage of this approach could be that validation can
always
> > be done against a single schema. This facilitates validation against a
> > schema without knowing or determining first which type of process it is.
A
> > secondary benefit could be that it encourages reusing a single type
> > definition for both versions where appropriate. Of course, the latter
> > benefit is a weaker one because it could also be achieved by having
three
> > schemas: an abstract version, an executable version, and a base version
> that
> > both the others import.
> >
> > Jim Clune
> > Parasoft Corporation          email: jim.clune@parasoft.com
> > 101 E. Huntington Ave.      voice: (626) 256-3680
> > Monrovia, CA.  91016           fax  : (626) 305-9048
> > ----- Original Message ----- 
> > From: "Danny van der Rijn" <dannyv@tibco.com>
> > To: "John Evdemon" <jevdemon@microsoft.com>;
<wsbpel@lists.oasis-open.org>
> > Sent: Friday, December 12, 2003 9:01 PM
> > Subject: Re: [wsbpel] Issue 24 - separate schemata for abstract and
> > executable processes
> >
> >
> > > yes.  that was the issue i raised.  and then yaron pointed out that
once
> > > separated, the schemata should be published as well.
> > >
> > > danny
> > >
> > > ----- Original Message ----- 
> > > From: "John Evdemon" <jevdemon@microsoft.com>
> > > To: "Danny van der Rijn" <dannyv@tibco.com>;
> <wsbpel@lists.oasis-open.org>
> > > Sent: Friday, December 12, 2003 5:21 PM
> > > Subject: RE: [wsbpel] Issue 24 - separate schemata for abstract and
> > > executable processes
> > >
> > >
> > > On Friday, December 12, 2003 4:32 PM, Danny van der Rijn wrote:
> > > >
> > > > while the current schemas were published, the current document still
> > > has the
> > > > schemas, i believe.  and the current schemas don't provide separate
> > > schemas
> > > > for abstract and executable processes.
> > > >
> > > > if i'm showing the effects of jet lag, please let  me know..
> > > >
> > > I think the schemas should remain in the spec (appendices) _and_ be
made
> > > available as separate files (as they are now).
> > >
> > > The current spec doesn't use separate schemas for abstract and
> > > executable processes - isn't this one of the reasons you raised issue
#
> > > 24?
> > >
> > > John
> > >
> > >
> > >
> > >
> > >
> > > To unsubscribe from this mailing list (and be removed from the roster
of
> > the
> > > OASIS TC), go to
> > >
> >
>
http://www.oasis-open.org/apps/org/workgroup/wsbpel/members/leave_workgroup.php.
> > >
> > >
> > > To unsubscribe from this mailing list (and be removed from the roster
of
> > the OASIS TC), go to
> >
>
http://www.oasis-open.org/apps/org/workgroup/wsbpel/members/leave_workgroup.php.
> > >
> >
>
>
> To unsubscribe from this mailing list (and be removed from the roster of
the OASIS TC), go to
http://www.oasis-open.org/apps/org/workgroup/wsbpel/members/leave_workgroup.php.
>
>