Take a photo of a barcode or cover
alexpghayes 's review for:
Thinking in Systems: A Primer
by Donella H. Meadows
i had high hopes based on vicki boykis' recommendation on twitter but was quite disappointed. the book begins with a solid verbal introduction to "systems thinking," which to the best of my understanding here means using differential equations or agent based models. there is then a nice but informal discussion of these systems and properties they can exhibit, illustrated on some very simple toy examples.
at this point, everything falls apart. the examples become primarily sociological: meadows describes systems based approaches to the economy, to addiction, and myriad other social problems. the issue here is that she never demonstrates that systems thinking actually applies to sociological problems, instead claiming rather breathlessly that they do, much in the tradition of other brigading but failed academic projects by physic, econ, cs, etc. in practice, i know of very few differential equation based models applied outside of natural sciences contexts in modern settings, which suggests to me that the claims to broad social applicability in this book did not stand the test of time and may amount to little more than (engaging!) thought experiments.
the book concludes with some generally good advice on how to intervene on real world systems, but i was so irritated by "intervene based on a model which is sure to be correct despite the fact that you never validated it" reasoning that i mostly skimmed it in a huff
at this point, everything falls apart. the examples become primarily sociological: meadows describes systems based approaches to the economy, to addiction, and myriad other social problems. the issue here is that she never demonstrates that systems thinking actually applies to sociological problems, instead claiming rather breathlessly that they do, much in the tradition of other brigading but failed academic projects by physic, econ, cs, etc. in practice, i know of very few differential equation based models applied outside of natural sciences contexts in modern settings, which suggests to me that the claims to broad social applicability in this book did not stand the test of time and may amount to little more than (engaging!) thought experiments.
the book concludes with some generally good advice on how to intervene on real world systems, but i was so irritated by "intervene based on a model which is sure to be correct despite the fact that you never validated it" reasoning that i mostly skimmed it in a huff