1. 2

Editor’s note:This forum presents models relevant to interaction design and service design. It describes the models, how they might be used, and why they matter. It also describes the models’ origins and contrasts related models. In its first year, forum articles described models of - Innovation process- Design process (Analysis-Synthesis Bridge)- Customer experience cycle - Learning in design and product development- The trend from a mechanical to a biological frame in design (era analysis)- Design research types (map)- Interaction types (taxonomy)However, none of the articles presented a model of models. We correct that oversight here. Models are ideas about the world—how it might be organized and how it might work. Models describe relationships: parts that make up wholes; structures that bind them; and how parts behave in relation to one another.For example, the sun rises in the east, moves across the sky, and sets in the west. Or the earth orbits the sun. Models support communication and learning. Models help bridge the gap between observing and making, between research communities and design communities [1]. Models are especially important in interaction and service design.



  2. 1


    In the course of trying to put together material to compose a new web site which I hope to “announce” in January, I was reviewing my (growing) email file of Model Reports filings, most of which originated by you. I was looking for a different post made somewhat recently, again, I think one of yours. It was related to research on an expanded version of the Forrester/Meadows World model, but updated to be more environmentally aware by incorporation of source and trigger variables that drive the Meadows variables in her model.

    In my saved email search (we should really be able to be more efficient and effective with this given our supposed competence with IT these days), anyway, I found THIS post of yours

    Reading the summary I have a feeling of what I’m going to find, but I thought I’d ask the question anyway:

    • Does this more or less portray the “state of the art” these days? A compendium of models, possibly identified and grouped by pattern of model and problem set and areas of potential utilization?
    • Or is there something more significant (operating at a higher level) that is currently announced research or announced operating capability?

    The reason that I ask, is because that has been the focus of my activities for the last 10 years or so since I officially “retired” from employment in the business world. A version of MDA, Model Driven Architecture. Rather than drilling down, trying to get models to write code, I’ve been looking outward using a version of systems thinking as a guide. Models defining components shared between systems, of systems, of systems. Then using the models, not so much to write code, but to get the models to document themselves and their behaviors and to be able to do analysis on themselves, and thus, indirectly on the environments and systems they attempt to specify.

    Thus, models of models … of models of models. My strategy for doing this is to use techniques devised by folks like the Object Management Group relative to computer systems analysis, and give them more capabilities and use them for analysis of systems of (all?) types. The scaling range of the models becomes infinite,as you are aware. But that is also true in terms of the OMG’s modeling schemes. In software, the problem is solved by what the OMG identifies as a 4 layer architecture for modeling.

    • an operational level
    • a model of the operational level
    • a model of the model of the operational (a meta model)
    • a model of the meta model, called the MOF.

    From the summary the “model of models” as described in this paper seems out of alignment with what I just described. It sounds more lateral (orthogonal) at the “model of the operational level” level. Am I understanding that correctly?

    I’m trying to create material to better, and fuller, BEGIN TO describe what I am doing circa the beginning of the year. But I’m also getting a bit cautious about whether or not I’m trying to describe something that someone has already attacked, and or accomplished. In which case I need to put in some caveats and do a comparative analysis.

    If not, the research described here is still valuable to me, and I want to thank you for having brought it to our (my) attention.

    In my “system” if it is more or less the compendium variety, it would get applied this way:

    • The operating environment is what it is. We only hope we understood it better, and I’ve got some ideas for that, but that’s not the question.
    • The model of the operating environment doesn’t directly change, but in the compendium answer, this is, I assume, one of the selections in the compendium, based on modeling technology,viewpoint, etc. except I want to target running a combined, integrated model that simultaneously incorporated multiple mpodeling techniques and viewpoints, as the model runs. All appropriate (based on problem type) models are integrated as a unit process analysis vehicle.

    Now comes the question about “state of the state” of modeling

    • In my “concept” the next layer is a model of models, a metamodel, which defines, organizes and executes the model set as an automated version of the operational environment, using the model components as though they were operational components driven by the model. The meta-model isn’t just descriptive, its operational.
    • And, the same is true at the MOF level. The MOF runs the meta-model, which runs the model, which emulates the operational environment.

    This all sounds like a lot of complication to run a model. However, using this “schema” as defined by the OMG (which is a codified version of an ability to specify recursive design schemes for modeling to allow any arbitrary depth of model specification to be accomplished) dynamic modeling solutions can be used to model and analyze systems to arbitrary depths using patterns as specifications.

    In my scheme, the compendium becomes a book of patterns, that become applied, as appropriate, to the problem being studied. Similar to a Six Sigma tool box, but specificational of dynamic entities with behaviors, component patterns, information requirements and products, etc. Which then get overlaid to become complex patterns defining how operational elements do or should behave.

    I’m kind of assuming at this point that is NOT what these people are doing. Is it? Or, do you know of anyone else who is, or who is interested in what I just described.

    Your opinions and input would be much appreciated. I much enjoy following your posts, both here and in the other ST forums.

    Best regards,