I've mentioned that working at Adobe is different from working at, say, Electronic Arts (or anyplace I've worked in the games business). It's not just a matter of size or money (I think EA is about half as many people, and about half as much revenue), it's a matter of culture. I believe that Adobe is the way it is because it was founded by engineers (John Warnock and Chuck Geschke) and is still run by the same engineers. It's a little mind-boggling to hear that one of your top executives just got another patent (can you imagine Trip Hawkins or Larry Probst being granted a patent on computer game technology? - not that Trip and Larry aren't smart, but they are marketing people, not engineers). The understanding of engineers runs deep into the company, from the fact that the new buildings have offices instead of cubes to the understanding that people need appropriate tools to do their work to the idea that people (engineers in particular) are not interchangeable and not easily replaceable and should be - for lack of a better word - nurtured. There are an alarming number of people at Adobe who have been there five years or more, in an industry where the average job stay is often quoted as two years. I'm still in the new job honeymoon period, but I think I like it.
During my - ahem - enforced idleness, I spent a number of weekends in the woods. A favorite place was (actually three favorite places were) north of San Francisco near Ukiah. South of Ukiah on 101 is a little town called Hopland (three guesses what they grow there!). Hopland is home to the worldwide headquarters of a company called Real Goods. Real Goods calls itself "the center for independent living", which is Mendonesian for "we sell stuff to get you off the grid and other goodies" like solar panels (including a package to run/charge your laptop), hemp clothing, hippie chic, energy-efficiency products like compact florescents, insulation, and so forth. They also sell books - books on building houses out of straw, building houses out of rammed earth (begins to sound like the Three Little Pigs, doesn't it?), building passive solar houses, building active solar houses, buying land, identifying constellations, the perils of EMF, and so on. In amongst them, I found the book I'm going to review this time around.
Beyond The Limits
Donella H. Meadows, Dennis L. Meadows, J┐rgen Randers
Chelsea Green Publishing Company, PB, 300 pp.
Beyond The Limits is not the usual book I review - it's not a book about software engineering, it's not a book about games, it's not a book about computer languages. So why review it? For the answer we must descend into the misty depths of time, all the way back to ... 1972.
In 1972 a book was published called Limits To Growth. It rocked the world. Published by The Club of Rome, Limits To Growth was the result of a two-year study at MIT which included the creation of a computer model called World3 to study trends in pollution, resource usage, population, and so on (sort of an early version of Balance of the Planet or SimEarth). Limits To Growth generated headlines like A COMPUTER LOOKS AHEAD AND SHUDDERS, and SCIENTISTS WARN OF GLOBAL CATASTROPHE - normally the stuff of bad SF movies (remember The Milpitas Monster or Attack of the Mushroom People?). Oil companies built advertising campaigns against it, parliaments discussed it, and in a medium-sized town in Oregon, a junior high science class read a summary of it. I don't remember offhand the name of the teacher (Judy something - maybe Davis) or the class, but I do remember the flash of understanding that the world was in trouble, that a computer (!) had said we were going to destroy the world. Powerful stuff for an 11-year old raised on too much Star Trek and SF & F.
Flashing back to the present, you ask "so what does Limits To Growth have to do with Beyond The Limits, and what do either of them have to do with computer games?" Well, you see, Beyond The Limits is the sequel (20 years later) to Limits To Growth, and it discusses in substantial detail how you build a computer model. Computer models are the centerpiece of best-selling computer games like, oh, say, could it be ... SimCity? Could it be ... Civilization? And I've not found much written on computer models worth reading - the best thing I have is still my notes from a class on computer modeling at UC Santa Cruz more than a decade ago (in which I learned how NOT to simulate a nuclear blast, but that's another story). Beyond The Limits is a nice, non-technical discussion of building a computer model, documenting it, and using it, and even includes pointers to where you can get copies of World3 if you want to play with it yourself.
So what goes into a computer model? The obvious answer is "biases", and the authors are well aware of this, since they are attempting to model the Real World accurately enough to predict the future. We have the advantage of wanting to model the Real World the same way a movie does - with flash and dash and verisimilitude instead of accuracy. There's nothing in Beyond The Limits that works against us (the fabulists, if you will) and for them.
Since Beyond The Limits is fundamentally a book on ecology, you will have to put up with a lot of ecology (and politics) in order to get the model stuff. The first half consists of chapters on Overshoot, Exponential Growth, Sources and Sinks, and The Dynamics of Growth in a Finite World. Much of the material is eco-centric and may or may not agree with your biases, but that's really irrelevant. What the authors do in this preliminary material is establish the basics of the model they intend to present, with discussions of positive and negative feedback, steady states, and possible outcomes. Illustrations are used effectively to demonstrate elements of modeling and outcomes. I have no doubt that you could use a drawing program such as Visio to create similar diagrams to help you define a model in your own work, and if you were ambitious, a Visio plug-in to dump the data in a useful form to a modeling engine.
Sometimes we have to go a little farther afield to find information which helps us work. Beyond The Limits is a good example of a book you're unlikely to find in your local computer store (or the computer section of your local bookstore), but it's well worth digging a copy up. If you need to, you can always call Real Goods in Hopland and ask them for one. But, as I've said before about books I like, you can't have mine.
Evan Robinson is a freshly-hired Engineering Manager in the Graphics Products Division at Adobe Systems, Inc., where he manages a central technology and resources team (for which he is currently hiring). Previously he was Director of Games Engineering at Rocket Science Games, and before becoming a manager he was a consultant, programmer, technical director, and game developer. You can email him to talk (well, OK, type) about books, software, development, or the meaning of life.
Copyright © Evan Robinson. All Rights Reserved.