Tilting at Windmills by

There’s nothing like putting up a straw man to win an argument.  First, you can propose arguments taken out of context, and secondly, there’s really no one on the other end arguing, you’re just arguing with yourself.

Still, Keith Swenson’s post makes for interesting reading. Unfortunately, there are a few real problems with his post, from my point of view:

First, his post ignores or impugns honest miscommunication between Keith and other members of the BPM or ACM field, in my judgment.  Worse, he fails to take responsibility for his own misunderstandings and misstatements, without doing so in a manner that makes it seem insincere.  It is as if everyone else is crazy, and Keith is the sole voice of reason.

There’s no reason for this, when the list of people who have misunderstood Keith, quite honestly, include esteemed members of our community like Bruce Silver and Sandy Kemsley (and yours truly!).  Not that I’d expect him to acknowledge these issues generally, but if you’re going to write a post alleging to represent both sides of an argument, I think this sort of gesture would be helpful.

I’ll just point out one such example:  Keith has many times asserted that BPMN is incompatible with ACM.  He’s back at it:

My position paper for the Adaptive Case Management Workshop was to propose that “BPMN is incompatible with ACM.”  I got a lot of flack from the hard core BPM disciples

For the record, I wouldn’t call Sandy Kemsley a “hard core BPM disciple” and her criticism of his talk was tougher than mine (though his response to my post was much harsher than his response to hers!).  Then he continues:

In spite of clearly stating that ACM is designed for knowledge workers to create their own process plans, many many still believe that there will be a process professional creating plans for others.

Hm.  Apparently this was, at that talk, and in several other places, not so clearly stated or understood. This falls under the heading of “everyone misheard me” or “I misspoke”.  When one group out numbers the other, I think it might be appropriate to give your audience the benefit of the doubt…

Second, more minor issue: the post still fails to communicate understanding of how BPMN can be used to describe generic processes. This may just reflect a shortcoming in the BPMS products that Keith is familiar with, or a difference in interpretation.  Update: Read the discussion he and I had in the comments section; his position is that using BPMN behind the scenes isn’t relevant to his statement that “BPMN is incompatible with ACM”, and that BPMN isn’t an “appropriate” technology for doing the behind the scenes work.

Keith concludes his post thus:

Note: most of the “Doubting Thomas” statements are taken from actual statements, either spoken to me, or written in exchanges, and I have used them here, if not word-for-word identically, at least representing the sentiment accurately as it was meant.  I did not identify or reference the individual, not wanting to put anyone on the spot, but I remain indebted to these individuals for their contribution to this blog post.

I don’t think the post serves the author’s goals.  To me, it conveyed a disdain or disrespect for the arguments of “Doubting Thomas” rather than accurately capturing sentiment or respectfully reproducing the arguments in a favorable light.  Not to mention the insincerity of being indebted to these unnamed individuals that you have spent the whole post expressing apparent massive frustration toward.  Some of the arguments presented by his fictional Doubting Thomas aren’t necessarily worthy of respect, but by mixing the absurd with the reasonable, you end up with an impression that the author is disdainful of all of the arguments.

I think the first step in understanding someone you’re arguing with is to truly think about it from their point of view, in their shoes, with their likely world-view – and using the best possible arguments that form a consistent logical framework.  If you can’t win the argument or take it to an impasse from both sides, then you’re not really understanding the other person’s point of view yet.  You’re not normally responsible for presenting both arguments in a debate, but if you choose to take it on, you owe it to yourself and others to do a really good job of representing the opposition.

 

  • keith

    Scott, Thanks! Your reflection that this post feels like an “artificial” strawman debate is an aspect of the post I had not anticipated. You are absolutely correct in pointing out that the serious, reasonable arguments have been left out. The only things I included there are the rather obviously flawed arguments which contradict the definition of ACM. If we are to have a reasonable discussion, we need to agree on the meanings of a few things. When the basic definition of ACM is to support unpredictable work, I do find it frustrating when people make statements that it SHOULD be for predictable work. This simply contradicts the definition, and gets us nowhere.

    And, as you point out, many of the luminaries in the field have at times made these flawed arguments. I don’t mean to imply that this is their current position, but instead these are arguments encountered over time, and in some cases not corrected. Take for instance Bruce Silver’s blog post still makes that statement that “ACM should really be about the repeatable logic” (http://www.brsilver.com/2012/09/21/bpmn-vs-acm-again/) but this is contradicts the DEFINITION of ACM. Below the post is a series of comments where Bruce eventually agrees that BPMN is not suitable for knowledge workers. What I found out is that still people think I am talking about the “developers” when I am talking about the actual end-user knowledge workers. I hope that is clear now.

    My post only included those simple illogical arguments, and I had not realized that this might imply that it the only arguments that exist. Clearly “winning” these arguments is not intended to convince anyone about the entire subject — only to please try an avoid arguments that contradict the definition. I added a note at the top to clarify.

    Your position that one should be able to use BPMN as a language of implementation underneath the ACM is a much more serious discussion. My latest post is an attempt to discuss one problem if you expect to have two representations of the process plans. This does not preclude BPMN being used as an implementation language, but simply the idea that process plans can be available in both. I will be very interested in your comments on that.

    http://social-biz.org/2012/11/19/two-languages-divide-but-dont-conquer/

    • http://www.bp-3.com/blogs sfrancis

      “When the basic definition of ACM is to support unpredictable work, I do
      find it frustrating when people make statements that it SHOULD be for
      predictable work.” – totally understand this frustration :)

      It is also true that sometimes other people are talking about developers, and you think they’re talking about end users. So the confusion I think goes both ways at times. This happens any time you talk about how a platform should work for end users and how it should be built – occasionally the pronouns/etc. get confused in the process.

      Thanks for clarifying the intent of the post – obviously part of it was lost on me ;) But the discussion below was helpful because there were a couple points that weren’t clear (to me) as to what your intent was, and after clarification it makes more sense, obviously.

      BTW, read your latest post- I like it. I agree with 80-90% of it…. :) But there’s one little thing I disagree with. Will post a bit later.

  • keith

    One more thing: which of the Doubting Thomas arguments did you think were reasonable?

    • http://www.bp-3.com/blogs sfrancis

      By reasonable, let me clarify- reasonable doesn’t mean “correct” just means that a reasonable person could see where they’re coming from and why they feel/think that way and there’s a logical basis for the statement or question.

      For example, whether you agree or not, it is reasonable for someone to say they could use BPMN as part of their implementation of an ACM system (under the hood). Some of the others come close – like the point about whether companies care if it is ACM or normal BPM – the argument that I recall seeing was they don’t care what it is called (which I think is true, as well as reasonable).

      scott