The Oil Drum

I purchased the World3 model CD, but was rather disappointed. The main item is a program that lets you choose one of the 10 canned scenarios and run it. This lets you look at the scenarios in more detail (you can get various charts or all the raw numbers) and compare them.

But you can't alter the model at all. Given the complexity of the model, this is perhaps a reasonable choice. I doubt it is easy to alter just a few initial values and parameters in such a complex model in a way that makes sense, unless you understand the entire model and have had some training in this type of modeling.

You can look at the model, sort of. But all you get is a large picture of tons of circles and arrows, with indecipherable variable names like "del ind out pc 40". No legend for what the various pieces are, no explanation of what the boxes, circles, and arrows mean (beyond the variable names).

It does seem to be open source. If you have a spare $1899 you can get a copy of Stella, the simulation software the model runs on. (You can also get a free save-disabled trial version that expires in 30 days).

What would I like to see? More detailed explanations of: the model elements and how they fit together. Probably the LTG folks have papers in journals that go over all this stuff -- I admit I haven't searched for it yet.

I consider these modeling efforts to be critically important for high-level decision makers to "get it" that everything is interrelated and you can't just try to force one problem (energy for example) to go away and ignore the others.

But for these models to have the requisite credibility they MUST be open for scrutiny. I expect that experts in different fields will scrutinize how the model treats their field, and render an opinion on how realistic the model is in that sub-area (e.g. population, or agriculture, or energy).

And generalists (like many of us at TheOilDrum) who have some appreciation of modeling should be able to get into the details as well.

It sounds like Mr. Lawrence has these goals in mind:

One goal of the project is to develop a simple game-like user interface that makes the model accessible to those without experience in modeling complex systems. Others with more expertise will be able to go into the model, understand how it works (“transparency” is another goal), and develop their own scenarios. Model users from around the world will be able to communicate with each other using a web site dedicated to model discussion, modification, and operation.

I'm interested in the "more expertise" version. However, the game version could be important for spreading the word. Have you considered contacting Will Wright, the creator of The Sims videogame? He did a game called SimEarth in 1990.

Finally, here is a disclaimer paragraph from the LTG CDROM:

The ten scenarios published in the book and presented on the CD are World3-03 scenarios. If you import the equations into STELLA and make changes in the model to conduct experiments with other assumptions, you are not any more working with World3-03, and you should not represent your results as coming from our model.

You can see their point. They don't want people altering their model and then claiming the results are "from the LTG World3 model". Openness has its consequences.

Don't try to predict the future. Get ready for it.