What are Dirty Bits?

January 01, 2015     #notes

The term “dirty bit” is a reference to the fact that even virtual goods can be tainted by their association with human and environmental degradation. But “dirty bit” is also a term of art in computer science. A dirty bit is a bit that is associated with a block of computer memory and indicates whether or not the corresponding block of memory has been modified. Dirty bits are used to “mark segments of data that need to be processed or have yet to be processed.” This double meaning seems to me to be significant, and I am working on ways to draw out the analogy. Suggestions welcome.

Towards an Environmental History of Computing

October 20, 2014     #info

This project grew out of a course I taught several years ago on “information and computer ethics” for undergraduate majors in our Informatics & Computing program. Most of the topics that I covered in class were already at least vaguely familiar to my students: privacy, intellectual property, cyber-crime, professional ethics, and the ethics of design.

There was one set of topics, however, that proved almost universally novel and disturbing, even to this reasonably well-informed (and generally very jaded) audience. This was a series of lectures in which I explored the environmental consequences of electronic digital computing. As we traced the global life-cycle of a typical laptop computer or cellphone from its material origins in rare earth element mines in Africa and South America to its manufacture and assembly in the factory cities of China through its transportation and distribution to retail stores and households across America and finally to its eventual disposal in places like the slums of Agbogbloshie, Ghana, the students discovered that the computer industry is built on more than just abstractions, algorithms, and information. Whether it was studying the toxic byproducts of semiconductor manufacture (there are twenty-nine EPA Superfund sites in Silicon Valley alone, the largest concentration in the nation), or the enormous amounts of energy and water consumed daily by massive Google and Facebook server-farms, or the use of child labor in the “computer graveyards” of the developing world, they were forced to confront the fact that computer power comes at a cost, and that the physical infrastructure that enables their virtual interactions are resource-intensive, pollution-producing, and potentially damaging to the environment. For many of these aspiring computer professionals, this was a sobering reality.

The point of these lectures was not to be alarmist, but to make an important point about the material underpinnings of the information revolution. The public conversation about the about the social implications of computer technology must consider its costs as well as its benefits. But it occurred to me as I was delivering my grim message that there were also valuable lessons to be learned for historians of computing. By focusing on the materiality of the electronic digital computer as constructed technology, I was forced to think about tangible objects, specific social contexts, and particular times and places in ways that I had not done previously.

And so this paper is an attempt to develop what I am calling the “an environmental history of computing.” As most of my readers will be aware, environmental history is about more than just the study of the environmental “impacts” of human activity. Environmental historians are interested both how humans shape their environment and are shaped by it. Environmental history emphasizes the active role that nature plays in influencing human affairs. And finally, it is concerned with the changing ways that humans perceive and understand the natural world.

For historians of technology, environmental history has proven an enormously productive tool for thinking with, a way to expand the scope of our field to include new actors, places, and questions. I am particularly interested in ways in which focusing the material underpinnings of the digital economy allows me to explore a global history of computing that encompasses surprising and heretofore neglected participants in the digital economy.

My goal is to develop a history of computing that moves beyond inventors, algorithms, and information. The phrase “computer power” is more than just a metaphor; when we look beyond the consumption of the digital, and focus on the physical infrastructure that makes our online interactions possible, we realize that, just as with more traditional forms of technological and industrial development, computer power comes as a cost. From Bitcoin “mines” to server “farms” to data “warehouses”, virtual commodities can have surprisingly material dimensions. They can be resource-intensive, pollution-producing, and potentially damaging to the environment.


© 2015 Nathan Ensmenger