Peter Cochrane's Hard Drive 1996
Homepage > Publication & Opinion > Hard Drive

---

Dead 100%
For millennia all our knowledge could easily be contained in one human brain. Father would teach son, mother would teach daughter and each generation would add a little more. It then became necessary to record information on cave walls, clay tablets, skins and parchment. We had reached a critical epoch - all human knowledge could no longer be contained in one brain. Specialism and co-operation were vital - no individual could do everything, compete and survive.

Today, our problem is acute. It is no longer possible to be an expert in anything other than a few virginal topics. For example, just 500 years ago it was possible to be an expert artist, engineer and scientist at the same time. Even 30 years ago it was possible to have a detailed understanding of telecommunication networks, cars, domestic appliances and military systems. Who now understands all of it? No one! Fortunately we do not have to understand the chemistry of a safety match to use one - but someone somewhere does.

Despite specialisation and an exponential growth in knowledge, we still see people of outstanding ability able understand and contribute more than the average. Unfortunately, they die and their expertise is lost for all time. The question is, can we capture their expertise and presence for future generations? Do they have to die 100%?

Multimedia holds out the prospect of being able to capture such wisdom, with the spoken and printed word supported by animated pictures. Suppose as some great teacher gave lectures and interacted one-on-one with students, it was recorded. Over a period of years it is feasible to capture over 95% of all likely questions and debate. How difficult then would it be to construct an artificial persona of this great mind? Might we sit and listen and watch a lecture long after the death and still be able to interact in a meaningful way?

This vision is partially available with today?s crude CDs of a mere 630 MBytes capacity - soon to be replaced by the next generation at 7 GBytes. We can already visit the Natural History Museum and receive a description of Barionix or the derivation of the word dinosaur. Granted, the presentation and style are limited by the medium, but it is a first indication of what might be done. Our rudimentary artificial intelligence systems can now filter and assemble the right slices of dialogue. Animation and pictographic representations on demand can also be triggered by a well-phrased question. What is required is a little more intelligence to filter what we require from ill posed questions or propositions. If only the medium could respond to: ?Describe a dinosaur named Barionix?, or ?Was Barionix a dinosaur, and if so, what was it like??

Just imagine the solution to Fermat?s last problem, not scribbled on paper and lost for ever, but recorded for all time. Alternatively, the many works and ideas of Newton that never saw the light of day. Perhaps, more powerfully we should contemplate artificial intelligence systems able to access such works, the coalescing of myriad concepts and results that currently escape us due to our limited human ability and memory. Perhaps in future, none of us will die in the strict sense, but our essence, an echo of our passing will live on. Perhaps then we will become a hybrid being, a network of total experience, learning and understanding continuously incremented by the inputs and interactions of a peripheral work force of individual human minds acting in unison with us. At 49 years old, and another decade or two to realise this world, I might just see you there.

Peter Cochrane holds the Collier Chair for the Public Understanding of Science & Technology at the University of Bristol. His home page is:
http://cochrane.org.uk