BPMS Watch

The Portability Challenge

In my previous post, I discussed how the meaning of BPMN portability differs between BPDM, XPDL, and students in my BPMN training.  Today, BPMN serialization is essentially a two-horse race: it’s XPDL 2.0 (maybe 2.x), or BPDM.  Don’t tell me “these are not competing standards.”  As far as BPMN serialization goes, of course they are.  But that doesn’t mean they are fundamentally trying to do the same thing.

XPDL 2.0 has a head start in the marketplace, but BPDM (published only in draft form) is the “official” one from OMG.  So it could go either way.

Surprisingly, however, neither standard seems to place top priority on what seems to me fundamental in any portability standard: the ability to create a model in tool A and - except for non-standard elements - open and manipulate the model in tool B.  I mean, how basic can you get?  You can do that with BPEL, for instance.

Keith Swenson of WfMC, the home of XPDL 2.0, posted the beginnings of an “interoperability” compliance test last year.  This was a good start, but they didn’t really keep it going.  Also, they set the bar too low.  The test consisted of 6 Hello World-type diagrams: a process with a single task, an exclusive decision and merge, a parallel split and join, conditional split without merge, exclusive decision and merge with swimlanes, and a single subprocess activity.  I mean, if different tools can’t serialize those trivial things the same way, what was WfMC thinking?  Also, I would say that several of the diagrams themselves are incorrect BPMN, for example using an OR-gateway to merge an exclusive split or a parallel split.  But let’s not quibble about that stuff, which is easy to fix.  Let’s get on to the challenge.

The BPMN spec includes a Voting Process sample to illustrate diagram semantics and mapping to BPEL.  It includes most, if not all, of the essential elements of a BPMN diagram, including subprocesses, events and message flows. 

Here it is:

 (main process)

  (discussion subprocess)

Coincidentally, the BPDM developers have included the Voting Process example in their supporting materials for the serialization.  Here it is:

 (main process)

 (discussion subprocess)

I propose that XPDL and BPDM both publish their respective serializations of the Voting Process diagram in the BPMN spec, and that vendors use this diagram and serialization as a portability test for their own tools.  If they convert Voting Process to the same XML, I call that portable, whether the layout is preserved or not.  Lest you think that such a challenge is “wired” for BPDM, let me point out that the BPDM Voting Process sample diagram is not, in my view, valid BPMN 1.0 or 1.1.  Maybe it’s BPMN 2.0 (??), but any student in my BPMN training can point out several ways that OMG’s redrawing of the diagram violates what we call BPMN today.  (OMG claims that BPDM is the serialization for BPMN 1.0 and 1.1, not just BPMN 2.0, which is still 18 months off.) 

So I think this is a fair fight, and both OMG and WfMC should be motivated to win it.   Keith, Robert, Antoine, Phil?  Whaddya say?

2 comments July 30th, 2007

On BPMN Portability

There’s no denying that BPMN is gaining traction in the marketplace.  I see it in my training.  I see it in BPMS and BPA vendors getting on board.  But what’s amazing about this is that it’s happening without a standard way to store and interchange BPMN between tools.  It almost boggles the mind that the creators of BPMN “forgot” about this when they started, and its current owners place model interchange so far down the priority list (it’s still not in the draft BPMN 1.1 spec, not yet released).

At the OMG Think Tank last week, I had a small roundtable on “what should be the purpose of BPM standards?”  Not well attended, but it was the afternoon of the last day, and half the audience had left for home already.   Besides, the topic was sort of a subtext for the conference as a whole, already beaten to death.  But clearly there is no unanimity on the subject. 

BPDM is OMG’s long-promised metamodel (and derived serialization) for BPMN.  Its advocates, like Phil Gilbert of Lombardi and Antoine Lonjon of MEGA, are clearly driven by the need to make process models precise in their execution semantics.  Advocates for XPDL, such as Keith Swenson of Fujitsu (who was there) and Jon Pyke, now of Cordys (who was not), focus on the diagram as a picture of the process, i.e. the graphical layout, in a machine-consumable form.  I think both of those camps are wrong, or at least out of touch with the majority of BPMN users, who want nothing more than to be able to model - at a business level, not in full executable detail - their business process in tool A and be able to open it, edit it, or simulate it in tool B. 

At our table I took a poll on these 3 meanings of what a portability standard for BPMN should provide.  You could vote more than once:

• Business-level model in tool A can be opened, edited, etc in tool B.  100%
• Above, plus preserving the diagram layout (xy coordinates of shapes, fonts etc).  50%
• Model created in tool A runs identically on system B, C, etc.  0%

The first bullet (my definition) also implies (to me at least, although maybe not made explicit in our poll) that tool B should understand all of the standard BPMN elements from tool A, not just some of them.

The second bullet is essentially XPDL’s definition, with the understanding that tool B is free to support only a subset of BPMN.

The third bullet is essentially BPDM’s definition, in a sense the answer to a question that few people are asking.

If I can remember how to get the poll widget on the sidebar to work, I think I will let BPMS Watch readers vote on this one.

1 comment July 30th, 2007

Back Again

Maybe I bit off a bit more than I could chew.  I launched my BPMN training this spring and simultaneously the update of my BPMS Reports on BPMInstitute.org.  Plus actual paying work.  It was stuffing 10 pounds into a five-pound bag, and the BPMS Watch blog was what fell on the floor.  But now version 2.1 of the BPMN training is about to go live, and the first 7 of 12 or 13 BPMS Reports are done, and I can at least come up for air.

In the past 2-3 months, not only have I not had time to write my blog, I haven’t even had time to read the blogs.  It just made me feel guiltier.  I started to dip into it today, but my Bloglines notifier still lists over 1500 unread.  It will take me a few weeks to dig through it.  I start as always with Sandy (moved back to her own domain! who knew?) and Steinar Carlsen’s links - the best.  I’ll be updating the blogroll once I have a chance to work through it all, but I already detect new energy in the BPM blogosphere since I last tuned in.

OMG’s BPM Think Tank was last week, as usual my favorite BPM conference - great networking, absence of BPM newbies and hustling for leads - and a great inspiration for blogging.  Sandy covered the talks and panels in her usual excellent fashion, but the real info exchange at this event takes place in the hallways and bars.

I plan to post soon about the standards confusion - BPDM, BPMN 1.1, XPDL, BPEL4People - and how it’s actually making some progress, as well as about the new training - what I learned from version 1 - and something I promised when I started the blog but have never delivered in a consistent way: capsule reviews of leading BPMS offerings and what is the essential strength of each.  That’s the plan, anyway.

Add comment July 30th, 2007