Ostlund & Revay "Systems Thinking in the Virtual Realm"
Posted by: daviding
Date: June 20, 2004 06:42AM
This is a very full paper, with references to Boulding's "Skeleton of Science", application of SSM, and references to Object-Oriented Modeling. Most systems scientists may be familiar with the first two, and I may be an exception at understanding the latter (since I work in information technology).
(Just as a note, Figure 3 didn't appear for me in my version of Word, but that shouldn't have too big an impact). I have two comments, one on the Boulding reference, and one on the SSM/Object-Oriented Modeling crossover.
On Boulding's skeleton of science .... Although it's a foundational article, I wonder a little bit about why you were referencing it. Is this because you were trying to get a handle on the VR-tual Pharmacy as a system, and were trying to build one gigantic system? An alternative approach would be to recognize the VR-tual Pharmacy as a mechanistic system, in the midst of a human activity system. (There must already be an existing system, because pharmacies work, even without this new technology).
The above view may draw us into a discussion as to the differences between a system of systems (that has at least one emergent property that doesn't exist in the parts), as opposed to a collection of systems (that interact, but can be recognized distinctly.
On SSM/Object-Oriented Modeling .... In my work with Ian Simmonds (at IBM Research), there seems to be a general confusion in the OO world about things, and information about things. (The most famous early IBM example, in OOUI design, was about dragging and dropping cars on people, which would mean cars "selling themselves". Actually, there's nothing physical happening to cars when they get sold. What happens is that information -- in the form of title to the asset -- gets transferred from the dealer to the consumer). I haven't thought about this deeply, and how it pertains to VR technology, but on the other hand, Ian and I have published a number of works on "mediating spaces" in the proceedings of earlier ISSS conferences.
I look forward to discussing these deep topics with you, but suggest that you might have to a little remedial education for other people in the audience about what OO modeling is, and what it isn't. (I'm remember being the 1999 Villanova conference honouring Russ Ackoff, where many of the S3 community thought that ERP packages such as SAP would help companies become more systemic. It was interesting watching the enthusiam drain from their bodies, as they sat through an SAP presentation, and started to realize how stultifying the technology can be).