Comments on: ACM Workshop at BPM2012: BPMN Smackdown by @swensonkeith

By: BPMN vs ACM… Again « BPMS Watch

BPMN vs ACM… Again « BPMS Watch — Fri, 21 Sep 2012 21:06:18 +0000

[...] latest round started with a paper presented by Keith Swenson at a BPM conference, stirred up by Sandy Kemsley’s review, and kicked into full riot by an ebizQ comment thread. The supposedly winning argument from the [...]

By: Scott Francis

Scott Francis — Tue, 18 Sep 2012 22:09:33 +0000

“Is that really how this came across? ”
Yes Keith, it is really how it comes across in the body of writing you do. I’m obviously not the only one who interpreted it that way, though maybe it is just my lack of understanding

By: Sandy Kemsley

Sandy Kemsley — Tue, 18 Sep 2012 20:27:36 +0000

Keith, I seem to recall you making the statement that “if it has BPMN, it’s not ACM”. I believe that we will see a lot of hybrid systems that handle everything from fully unstructured case management to structured BPM, and the entire spectrum in between. In that case, the products/platforms will have BPMN as a modeling capability, but will not insist that the end users of ACM functionality use it.

By: Keith

Keith — Tue, 18 Sep 2012 17:54:40 +0000

This is not about systems “built with BPMN”. Of course case managers will use BPM systems regularly. They will also use software written in Java or in C++ or in COBOL. This is not about the technology that programmers use to create systems for other people.

I clearly don’t care what is “under the covers”. Is that really how this came across?

ACM is a technology where the case manager MAKES PLANS themselves, as part of what they do. This discussion is about how the case manager makes those plans. These plans ARE by definition things that the case manager sees and interacts with.

ACM is an approach for knowledge workers: people who think for a living. These are professionals who know their particular job in their particular situation better than anyone else. Their processes are unpredictable, and so they have to be elaborated as the work proceeds. The work is not repeatable so there is no point if spending a lot of effort making a plan that is elaborate. The ACM part is the part that they actually construct and manipulate those plans.

The only person I care about is the case manager, because that is the person responsible for the plan and for the outcome of the case. The case manager must be both create and modify processes. That is the representation of the process that I am talking about being incompatible with BPMN.

My abundant support for BPMN in the BPM realm should make it clear that for appropriate uses, BPMN is fine. But this was not a talk about BPMN is useless for anything. This was that it was incompatible with ACM.

By: Imperative vs. Declarative Processes Models | Collaborative Planning & Social Business

Mon, 17 Sep 2012 13:31:22 +0000

[...] prescribed course is illegal. In my talk on “BPMN is incompatible with ACM” (slides, review) I pointed out that when people draw a BPMN diagram, they sometimes decide to place actions in [...]

By: Sandy Kemsley

Sandy Kemsley — Tue, 11 Sep 2012 09:51:35 +0000

Keith – headlines are meant to grab attention

As Scott points out, just because there’s BPMN behind the covers, doesn’t mean that it has to manifest that way to all users.

By: Scott Francis

Scott Francis — Thu, 06 Sep 2012 04:33:54 +0000

Also, keeping in mind that solutions built “with BPMN that works for users” may not mean that users interact directly with BPMN – and yet they can still enable the user to specify all kinds of interesting things which the BPMN model will leverage

By: Patrick Lujan

Patrick Lujan — Thu, 06 Sep 2012 00:03:08 +0000

No, the issue is can a solution be built with BPMN that works for the users? Answer: sometimes yes, sometimes no, it depends. It’s not an all or nothing proposition. Again, as with structured versus unstructured, depends upon what you want to do. That is, sometimes BPMN is the right widget, sometimes not.

By: Keith Swenson

Keith Swenson — Tue, 04 Sep 2012 07:33:06 +0000

Sandy, as usual an excellent high level summary, but I really must object to the title. If I was smacking anything down, it was not BPMN. Instead, you might say I was trying to smack those people who feel there is one solution to everything, and tend to be indiscriminate about what they use BPMN for. Once they learn BPMN, they tend to want to use it for everything (including conference registration instructions).

A summary of the workshop paper can be seen at: http://social-biz.org/2012/07/21/bpmn-is-incompatible-with-acm/

The issue is not that the diagrams drawn by developers are overly complex. The example of process I used in the talk were very very simple: just a sequence of three activities. My point was that because this simple diagram is the RESULT of a planning/design process, and because the rationale behind that simple process is not included in the diagram, the runtime case manager can not have confidence in changing it.

Your example that people might be able to restrict themselves to patterns that are run time modifiable is like “Cooking Dinner in a Chem Lab”. Sure, a chemist can restrict themselves to only using edible ingredients in the dinner, but I will be much more comfortable eating dinner produced in a kitchen that is maintained properly to only contain edible substances.

I present this as a “discussion topic” not being 100% certain, and so I hope that this discussion can continue. My conclusion is decidedly “low tech”. There are times that a lower tech approach is more friendly to users. This might be one of those situations.