March 24th, 2013
We view technological progress as the tail wagged by the dog of science. Science progresses, creating opportunities for new and better technology. Technological progress is the shadow of scientific progress.
Of course, technology also advances science by providing new and better instruments and devices with which scientists can further their research. Think how much science has been accomplished with the electron microscope or the computer. So we’re willing to grant technology a somewhat more active role than the passive recipient of the benefits of science.
But there’s another area of technology-driven change that deserves our consideration: the way that technology changes our thinking. I have elsewhere claimed that the greatest impact of the computer on civilization will be the way that it teaches us subjunctive thinking, in the manner that writing taught us sequential thinking. In this essay, I propose to discuss how technological progress has led us to a perverse way of thinking.
Consider a comparison of the early games machines with a modern machine:
Consider a comparison of the early games machines with a modern machine: The upper two rows, shaded in yellow, represent the data storage capacity, while the lower two rows present the processing capacity. Recall my concept of process intensity: the degree to which a program relies on processing rather than data to achieve its objectives. A data-intensive program shovels data from the hard drive to the screen; a process-intensive program calculates the visual scene from basic concepts.
Observe that the data-storage capacities of modern games machines have increased by a factor of millions, while the processing capacities have increased by only thousands. We have made thousands of times more progress with data than with process. That, in turn, has affected the way we think about computing.
Thought experiment: an alternative universe
The reason we have made so much more progress in data than in process is twofold. First, demand for data storage has been more insistent than demand for processing power, because people grasp the utility of data storage more readily than the utility of processing power. But the more important controlling factor, I think, is physics. It’s easier to shrink memory elements, be they in silicon, magnetic media, or optical media, than to build more powerful processors. The reasons have to do with the nature of light, electrons, and semiconductor structure.
But let’s imagine an alternate universe in which the physical factors were reversed. Let’s suppose that making processors faster was easier to do than making memory cells smaller. In this alternate universe, the comparison table might look like this:
How would entertainment software be different today in such a world? My guess is that there’d be a lot less software overall, because designing good stuff in such a world would be more difficult. Without the convenience of vast amounts of storage space, designers would have to rely more heavily upon algorithm design, which requires a different way of thinking: process-oriented thinking, which relies on algorithms and, ultimately, mathematics. Since there are fewer such people, we’d see less games software. On the other hand, the rewards for making such software would be greater, and these greater incentives would encourage people, I think, to learn to think in more algorithmic ways. This would in turn push society in the direction of this kind of thinking. Ultimately, it would be for the benefit of civilization for this accelerated emphasis on algorithms.
Alas, the laws of physics didn’t turn out that way. They leaned in a direction that supported our prejudices, and so we still have not yet learned to tap the true power of the computer.
Back in the early 1980s, when we were stuck with the Atari 2600, programmers had no choice but to think in terms of algorithms. Looking back on those days, it seems to me that the programmers thought about game design at a deeper level. It’s true that, with 30 years of experience and the contributions of ideas from thousands of people have greatly advanced the state of understanding of game design. Still, if you play some of those old Atari 2600 games, you can readily see that the ratio of design sophistication to hardware capacity is greater with those old games than with modern games.