Mistakes of the Future

On Jamais Cascio's blog, he asks an interesting question (which he borrows from Paul Kedrosky, who attributes the idea to Richard Feynman). The question is: "If all knowledge about [your area of expertise] were to expire, what one sentence would you tell the future?"

Jamais listed me as someone he would like to see answer the question. Feeling unoriginal yesterday, I borrowed this peerless advice from Joe Miller: "Those who ignore the mistakes of the future are bound to make them."

But as I thought about that quote, I wondered, what mistakes have been made in the past about understanding the mistakes of the future?

Consider this:

• In 1900, there was no such company as General Motors. Fifty years later, in 1950, GM was the biggest corporation in the world.

• In 1950, there was no such company as Microsoft. (In fact, Bill Gates was not even born yet!) Forty-five years later, in 1995, Microsoft was the most highly valued corporation in the world.

• In 1995, there was no such company as Google. Just ten years later, in 2005, Google was one of the most important companies in the world.

• In January 2005, there was no such company as YouTube. Just 21 months later, YouTube was sold to Google for $1.65 billion.

See a pattern here? It's hard to predict the future. It's nearly impossible to accurately forecast successes, so what chance do we really have at knowing the mistakes to avoid?

It should also be obvious from the data above that change is happening faster than it used to. Most of us know that, of course, but this reality only complicates the challenge. Change is happening faster, so mistakes are coming sooner. Is there any hope?

I've been asked before about the premise of CRN's mission. Very bright people have suggested to me that trying to shape the evolution of a technology along "responsible" lines is, essentially, a fool's errand. We don't know enough, they say, and even if we did, no one really has the power to guide the evolving, emerging future.

Perhaps they are right. It certainly is easy to look at the past and conclude that we, all of us, are very nearly blind when it comes to looking ahead.

But somehow I'm not satisfied with that answer. I'm not willing to throw up my hands and let the future overtake us without at least trying to have an influence.

We do have some advantages that did not exist in 1900, or 1950, or even 1995. We have more history to learn from, for one thing. We have greater access to knowledge than anyone has ever had, thanks in part to Google. We have closer and more immediate connections to people, all around the world, who can join together to get things right.

Dennis Gabor has said, "The future cannot be predicted, but it can be invented." So let's begin inventing the future we want, right now, today.

One thing we know for certain is that if we do not try, then we can not succeed.

Mike Treder

Tags: nanotechnology nanotech nano science technology ethics weblog blog

Comments

[Nato Welch]

The problem with "it's impossible, give up", is that it too closely resembles "it's impossible, //so let me//". That's a recipe for being taken advantage of. Resignation about the future eventually becomes resignation about the present. If we throw up our hands about predicting the future as citizens and leave it to others, then those others will be the ones calling the shots when the future turns into the present. It is therefore critical that we be concerned about both.

I don't see CRN's task so much as being magically able to foretell the future as to be prepared to quickly analyze the future when it comes. Collaborative scenario forecasting takes time; It wouldn't be good if policy makers and voters had to deal with disruptive technologies without the benefit of such forecasts already having been done.

[Brian Wang]

I think it is possible to be far better at future prediction than most attempts have been. It requires more work and more information about the present and past situation. Many "futurists" just declare themselves to be one and start making stuff up. They also claim no accuracy for their predictions. These people are no better than magic 8 balls or astrologists.

Future predictions are quite accurate for some narrow domains. Some things that are fairly accurate population predictions, some economic forecasting, some market forecasts, some environmental forecasts, some political forecasts etc...

Some statistical descriptions of the current and past situations are quite accurate. Such employment data from some countries, housing reports and economic data from some countries etc...

Some business plans are accurate predictors of the future. Some business, process advantages and behaviors are enduring/durable.

By properly modeling and incorporating what we know and are more certain about then we can stick closer to what is likely, probable and possible. It is important to try to understand relative power and durability of trends and how and why things change.

Also, many predictors are overly willing to make predictions which would require mass violations of solidly established laws and regulations. Some of those things could happen but making a prediction such as everyone will be cured of cancer within 2 years that ignores FDA approval testing and times is a nontrivial change. Somewhat like predicting the baseball world series will be played every 3 months.

Getting more detailed predictions of technological winners can be tougher. But again some predictions are more likely. In sports, a prediction of Yankees will the world series in 6 games in 2 out of the next 10 world series has some statistical likelihood behind it and some sports economics.

Predictions that some currently unknown companies will overturn the status quo in some aspect of computer technology has some precedent. But it is useful to know where the status quo is weak and where is it strong. How much of a change is the shift ? How much of the shift is technological and how much marketing ?

Predictions can be more like scientific hypothesis. Assumptions can be tested. Tracking of progress towards validation or invalidation should be possible. Invalid predictions should be analyzed for faults. Predictors should learn from mistakes and look for opportunities to get constructive feedback.