Systems: an ontological viewpoint
Before getting back to the more detailed subjects that I mentioned in this post on a business architecture viewpoint, I thought I should say a few words about the concept of a system. This is fundamental, not only to my thinking about business, to my thinking about everything. What I am doing here is making a statement about my personal ontology — the way that I view the world.
A reasonable working definition of system is “a set of interacting or interdependent entities, real or abstract, forming an integrated whole.” This is very broad, and I intend to use it in the broadest possible sense. From this perspective anything we can call out for consideration can be considered a system.
By no means does this mean that all systems are the same, just because we perceive them as systems. In fact, this opens up the opportunity to classify and characterize systems in an infinite variety of ways. Elsewhere I have pointed out a long list of types of systems, and that is the merest tip of the iceberg, once we start to think of a molecule as a system, the solar system, a computing system, and an enterprise or business system. I also point to that earlier discussion to call attention to some of the background of what I mean when I talk about the importance of system in my personal ontology.
A key point in my personal ontology is the interrelatedness of the concepts of system, purpose, and observer. I resonate with the thinking of Ross Ashby, who defined a system as a set of variables to be considered, and said, “every material object contains no less than an infinity of variables and therefore of possible systems.” I also resonate with the thinking of Maturana and Varela, who said, “everything said is said by an observer“. These points converge to say to me that anything can be considered as a system if there is a purpose that causes a person as an observer to want to treat some set of physical or conceptual entities as a system.
Having said that, it is important to balance this with what I like to call the Korzybski dictum: “Whatever you say a thing is, it is not.” The implication for our purposes is that simply naming something as an entity, and saying that it is a certain kind of system, may be very useful, but in no way exhausts the things that could be said about that phenomenon.
This is all background for one of my main and continuing points, which is that it is useful to think about businesses and other enterprises as living systems. This allows us to think about how they might be autopoietic, and whether they can be killed, whether they have have health, whether they evolve, reproduce, mature, etc. I hope to say more about this, in particular, very soon!
February 11th, 2008 09:39
One of the challenges of thinking in systems is the definition of the boundary. You and I had previously discussed that you make the effort, when drawing diagrams of systems, of making the boundary a dotted line. This is a subtlety that is often lost when we get buried in the details of looking into a system more deeply.
I’m mindful of this when I’m working through philosophy, particularly when I think about the phrase being-in-the-world. (This comes from Heidegger, which I haven’t actually read — I don’t read German! — and rely on the interpretations of Hubert Dreyfus). Being-in-the-world suggests a connectness between a human being, other human beings, tools that are at hand, and other things are present but not immediately at hand.
Being-in-the-world works in a context of situatedness. (Think about your favourite concert or sporting event, which could lead you to say “you had to be there”).
When thinking about human systems and information systems, it’s sometimes easy to draw the boundary around each one independently. If we look at the world as one where human beings use tools and interact with other human beings, we’re in a different system.
February 12th, 2008 06:05
Yep, couldn’t agree more.
By the way, when it comes to situatedness, have you looked into the situation theory work of Keith Devlin? I have found Devlin’s notion of situation (based on work by Barwise and Perry) is a very useful concept.