«

»

A time capsule reveals the limitations of tech predictions

NCAA-Blount-topper

 By Simon Davies

While rummaging through an old data archive today I discovered a document that I thought had been lost to history. It was a profoundly humiliating read – even more so than those woeful newspaper articles I churned out as a teenager in the late 1970’s warning that new technologies would totally devastate human employment.

If the past teaches us just one lesson, it is never to trust fortune tellers.

The file was a set of notes that I’d compiled almost exactly twenty years ago (June 1993) in which I tried to imagine the world in 2005. Honestly, be wary of making predictions; they rarely match the reality of the future. I will provide the gory supporting evidence in a minute.

Yes, I know I posted a blog on here a while back urging everyone to be futurists, but I take it back. Now I’ve seen the hard evidence of the futility of future gazing I wouldn’t recommend the practice to anyone.

Humans have always been hopeless at predicting the future. If the past teaches us just one lesson, it is never to trust fortune tellers. Long range forecasts are overwhelmingly wrong. The few that endure do so invariably through good luck or the simple law of averages.  Or, as Arthur C Clarke put it “The future is not what it used to be”.

In 1948, the founder of IBM demonstrated his visionary prowess by predicting that as many as twelve companies may someday have their own computers.  And, according to the May 1967 edition of Popular Science, 75 per cent of all US college graduates would go, within a decade, into jobs in the computer industry. Little wonder that we no longer fear the future: the prophets have let us down so often that we are cynical – and largely optimistic.

In 1948, the founder of IBM demonstrated his visionary prowess by predicting that as many as twelve companies may someday have their own computers. 

Nevertheless, some researchers are happy to dip their toe in the waters of the future – even if the stench from a millennium of dead prophecies has made everyone nervous.  Most people now generally agree that the margin of viability in prophecy appears to be ten years.

John Naisbitt, for example, wrote a magnificent best-seller in 1982 called “Megatrends”, in which he used Big Data-style research methods to identify forces that would shape our destiny. The book was widely viewed as a masterly study on the future, and for some years it seemed to be right on target. Twelve years later his predictions went off the rails.

Naisbitt went for the Big Picture. He argued, for example,  that short term thinking would become long term thinking, yet a world recession reversed this prediction. He predicted that centralisation would yield to de-centralisation – yet throughout the world, predatory corporate practices have killed this hope. As soon as a trend emerges it instantly induces a spectrum of contrasting forces. Nonetheless, I decided to work within Naisbitt’s twelve year margin.

Returning  to my pathetic document, let’s get my most embarrassing speculations out of the way. Take this mad prediction:

As soon as a trend emerges it instantly throws open a range of contrasting forces.

“It’s likely that you will be operating your computer through voice recognition. Keyboards will be largely redundant.  People will find more satisfying uses for their hands. A new generation of software will transcribe the human voice at the rate of 100 words a minute. The “tuning” process will be fine enough to produce an accuracy just slightly worse than the average professional typist.  Instant language translation using artificial intelligence will be in general use. The barrier of language will be removed.”

OK, OK, we were all really excited in the early 90’s about voice and speech recognition. How were we to know the exponential research outcomes would start to plateau? Even text translation was still relatively hopeless in 2005.

The predictions become even more absurd:

“Powerful Virtual Reality (VR) systems will be commonplace, though they will then be cheap enough to suit the pocket of most households, and spectacular enough to actually deserve the name “virtual reality”. Sensual and erotic VR will be extremely popular.  Used in combination with designer drugs it will provide a unprecedented form of recreation that makes computer games and television insipid and obsolete.”

How were we to know the exponential research outcomes would start to plateau?

Designer drugs and VR? What was I smoking!

I couldn’t even accurately predict mundane statistics: “The speed and capacity of personal computers will be such that 100 gigabyte hard drives the size of a paperback novel and capable of storing the equivalent of a small library will cost less that a thousand dollars.”

At the time my friends thought I was insanely optimistic with this prediction, but I was out by a factor of ten. Such space was available in 2005 for little more than $100. These days if you shop around you can snap up three terabyte storage devices for less than $80.

I should have seen that one coming – as I should have second-guessed the heady growth of Web connectivity. Back then – with in the low millions – I predicted 200 million people would be connected to the Web by 2005. The actual figure turned out to be a billion.

Some predictions were a mixed bag. I was half right in predicting that “computer technology will have reached the point of total compatibility – virtually all machines will be able to talk with each other” but wildly off the mark with the view that the Web “will be so complex that one and two year university courses will be offered to help people find their way around the system”.

At the time my friends thought I was insanely optimistic with this prediction, but I was out by a factor of ten.

Whoever could have imagined the idea of search engines back then? At the time – unknown to me – JumpStation was in development, but it would be five years before Ask Jeeves was launched.

“Schools will finally have caught up with the information revolution through the Internet, mainly because it will offer a range of low cost, high impact education packages,” I naively gushed. “As a result, face to face teaching will be directed more to low-achievers, disadvantaged students, and to lessons in languages and crafts.”

Yeah, right. You can really see where all those savings went to the most needy students.

Oh… here’s one that seems reasonably accurate: “A network of satellites will make it possible for you to be contacted wherever you are on the globe. By 2005, many people will have been issued with cradle to grave “universal personal numbers”, which will follow them wherever they are, and allow them to be contacted in any circumstance. Mobile telephones will be as cheap as ordinary fixed wire phones.”

It turned out that the one aspect on which I was generally correct was the extent of surveillance that would be created:

Whoever could have imagined the idea of search engines back then?

“In the course of our day to day contact with organisations we will generate mountains of personal information which will become known to more and more computer systems. These computers will become so much a part of our life and fortunes that we will scarcely notice their existence.”

“The relationship between public and private sector will grow in importance over the next two decades to the point where all personal information may ultimately be subjected to government scrutiny and control.”

“To minimise the possibility of terrorist attacks or sabotage of information networks, government may well demand more and more access to our personal computers and our communications.”

Well that was one that hit the mark with disturbing accuracy.