A philosphical look at the Mac
Author: Mike Furniss
Date: July August, 1989
Keywords: Hyperbit column hypercard learning
Text: When describing Hyper-information, it's common to talk of this way of organizing information in terms of how human beings think. The association by subject in a web of volitional relationships, as opposed to the linear presentation of ideas in printed media, is what really distinguishes Hyper-information and gives it so much promise. It is similar or identical, in some respects, to how we think. Indeed, the tendency to get lost or feel lost in HyperCard stacks is not an unfamiliar feeling to me in my own thinking. ''What was I thinking, again?'' Or, ''How did I get onto this line of thought?'' When you get lost in a stack, think about how that same thing happens in thought. I enjoy thinking about the similarities between computers and the human mind. It's clear that making computers work more like we think is the best way to increase the usefulness of computers and make them better mind-levers, or ''wings for the mind,'' as John Sculley describes his vision of computers. When the Mac does a search or sort or a nifty reformat in seconds, we might get a feeling of inferiority imagining how long it would take us to do any such thing. Yes, it's all ones and zeros under the hood, but it sure can do some fancy stuff through the hierarchy of software. But the question arises -- How do modern computers compare to the human brain/mind in shear power? Turns out, they don't compare very well. Not yet anyhow. Let's look at human vision for example. The construction of real time, full color, stereo images that our eye-brain does, turns out to be a task that is way beyond even the biggest and fastest CRAY supercomputer. The information rate and the processing complexity of the eye-brain ''wetware'' outclasses anything people have been able to come up with in a computer. So as you sit and look at your screen, the computing that's going on just processing the image on the Mac screen into an image in your mind is far beyond anything the Mac happens to be doing. That's not to mention the interpretation of the image and the thinking that goes into what you happen to be doing. The really interesting thing is that neurophysiologists tell us that nerves work really slowly, something like a hundred miles and hour for nerve impulses to travel over neural networks. (Note: Have you seen the ''neural network'' HyperCard stack?) But the electronic impulses in computer hardware travel at near the speed of light. Yet our brains are able to process many types of information as a much higher rate than supercomputers. Why? How? Theorists suggest it is a combination of fundamentally different hardware architecture and far superior software, both designed by the Great Developer over millions of years. Our nerve networks and brain organization appear to be parallel, working together all at once on problems, while computers are mostly serial, going at it one step at a time, albeit quickly. It appears that parallel processing is far superior to serial processing, but a lot harder to engineer. While that proba- bly only a part of the difference, it's one that's widely recognized and is the subject of much advanced computer and artificial intelligence design work. We'll probably see a more attempts to approximate neural networks in computer design, and more software that, like HyperCard, works more like we think. It looks pretty certain though that the ranking of our brain/mind system as No. 1 is safe for a good while to come, and our external computing creations have a long way to go to match these few pounds of wet grey stuff we carry around with us.
Copyright © july august, 1989 by Mike Furniss