An important aspect of future predictions is estimating the timeframes for the realisation of particular predictions.
It is often mentioned (quotes?) that we tend to overestimate short-term progress and underestimate the long-term progress. In the short-term people are often overoptimistic in regards to when particular products can be brought to market and do not foresee all possible setbacks. With larger time scale the dominant problem is that people simply do not realise all the possibilities and all synergetic effects.
This leads to widely divergent opinions about future technologies, where predictions from supposedly qualified experts range from "in 25 years" to "in 5 centuries" to "never". While it's certainly hard to argue for the validity of your own "good" forecasts, you can rest assured that the majority of the "experts" are uninformed and their methodology is nothing more than making uninspired random guesses.
states that something is possible,
he is almost certainly right.
When he states that something is impossible,
he is very probably wrong."
Arthur C. Clarke
This can be taken as evidence that widespread opinions are not a valid predictor of technology's feasibility and importance. It can be even argued that even scientific consensus can and should be ignored. Of course, this doesn't mean the scientific arguments themselves should be ignored, but that at the early stage most scientists are simply not competent and too persistent in their preconceived beliefs to rationally evaluate new ideas.
Some empirical research in 1970-1980s indicated that most specialists could only comfortably predict developments in their specialisation areas only about 5-7 years into the future.
then they fight you, and then you win."
This pattern is commonly repeated with new technologies. The Internet, nanotechnology, Tablet PCs all passed these phases and have been widely regarded as mere curiosities, to say nothing of famous scornful remarks about the lack of merits of telephone, car, airplane and TV in the past.
While there is no obvious way to reliably predict the timeframes of future developments, there must be some good shortcuts.
- A possible way is to find the key technology that is required for certain development and can be described quantitatively, e.g. carbon nanotubes for space elevator and their length. This parameter can then be plotted on the time graph and some reasonable forecasts can be made.
- Another way is to guess the number of necessary technological iterations (i.e. generations of the technology) and use the length of R&D cycles as the basic time unit. For example, this can be done to predict when a certain type of a plane/car/CPU can be developed.
- It may be so that the attitude of the society (including the masses, the scientific establishment, mainstream media, etc.) follows a similar path with many different technologies. In that case, when a formerly disputed prediction is finally voiced by a few reputable publications, we can be reasonably sure that the predition is most likely true, but the majority simply hasn't yet catched on and should be ignored.
- We can note similarities with and the models behind technology adoption that have come before: The uptake of email addresses, use of Wikipedia, participation in social networking technologies.
- Perhaps we can look at activities surrounding devloping technologies and draw some parallels with how older technologies have developed. Comparing and contrasting different industries in different times can reveal common patterns, which can be used in predicting. For example, one may ask: In what ways does the development of robotics resemble the development of computers? Are spending and investment patterns similar in large corporations and governments? What kinds of activity are we seeing in amateur communities? How are standardization efforts progressing, are there standard component interfaces? Are there new markets emerging (e.g. a parallel between hobbyist markets for early personal computers and toy robots)? Is experimentation with new products and technologies and their adoption happening in the same way as it was with an older technology before?