Header 1

Our future, our universe, and other weighty topics


Wednesday, June 19, 2013

The Crystal Ball Formula: Forecasting the Future

The Crystal Ball Formula: Forecasting the Future


John William Waterhouse, The Crystal Ball

Whenever we hear about some Cool New Thing that we might enjoy in the future as a result of technological innovation, our first question is usually: when will we be able to get that? Let us make this kind of speculation a little more systematic. Let us try to figure out whether there is some general formula or equation we can use to calculate when some future innovation will be available so the average man or woman can benefit from it. We can call this formula the Crystal Ball Formula.

The average unsophisticated technological prognosticator will use a very simple formula to calculate when a future invention will be available. The seer simply estimates how long it would take to produce the innovation, given generous funding. Then that length of time is added to the current year to produce an estimated arrival date for the innovation. For example, a writer might estimate that we could have robotic lawn mowers in ten years, so the writer might then say (in the year 2013) that you will not have to worry about mowing your lawn once it gets to be 2023.

This naïve formula can be expressed as follows:

Ya = Yc+ N

where Ya is the year when the future innovation will arrive,
Yc is the current year,
N is the number of years of development needed to produce the innovation.

However, this formula is too simple. For one thing, it works only in a case when well-funded scientists or inventors are working full blast on some future invention, and will keep working on it until we get it. Just as common (or more common) is a case when development of some future invention is delayed because of a lack of market demand or a lack of public interest.

For example, consider the possibility of lunar hotels. There is little market demand for lunar hotels. If companies started working full blast on lunar hotels, they might be able to develop them in fifteen years. But if we are to estimate when lunar hotels are available, we should really factor in a decade or more in which there isn't much work being done on lunar hotels. So we might assume that perhaps in the year 2025 development will speed up on lunar hotels, and that they might then be available around 2040 after fifteen years of work.

So let us make a revised formula that takes into effect cases in which technological development is slowed because of a lack of full-time current funding. The revised formula goes like this:

Ya = Yc+ N1 + N2

where Ya is the year when the future innovation will arrive,

Yc is the current year,

N1 is the number of future years before Ya when relatively little work will be done to achieve the invention (or when the development work is not well-funded),

N2 is the number of years of well-funded, full-time work needed to achieve the invention.

This formula leads to more realistic estimates. For example, if we want to estimate when we will have floating villages, we might imagine that N1 is 20, because there is currently no demand for floating villages. We might also estimate that Nin this case is 15, if we think that floating villages could be built after 15 well-funded years of development. So we would then estimate that floating villages will arrive in about 35 years.

But our formula still needs some additions, because there are two other things we need to consider. The first is the fact that technological innovations are often delayed because of regulatory roadblocks or social roadblocks. For example, newly developed drugs in the US often have to wait years before they are approved by the FDA. A recent paper by three professors complained that US “drug war” regulations are slowing to a crawl research on psychoactive drugs that could produce breakthroughs in treatments for conditions such as post-traumatic stress syndrome. Another example is restrictions on stem-cell research during the Bush administration. It may well be that many exotic future inventions will be delayed because of similar roadblocks.

Another thing we need to consider is that even after a technological breakthrough occurs, it may be a long time before it gets out of the laboratory and becomes available at a price that is affordable to the average person. This factor is typically ignored by technological prognosticators. We had video phones in the Bell labs around 1972, but it was only about the year 2000 or later that the average person could afford to have a phone conversation in which he would see the person he was talking to (through systems such as Skype). We had electricity working well in the laboratory about thirty years before the average person had an electrified home. Today we see numerous medical breakthroughs that are still ridiculously expensive long after they were developed. Some cancer drugs cost more than $100,000 per year.

To take these two factors into account, let us modify our formula again to make it look like this:

Ya = Yc+ N1 + N2 + D1 + D2

where Ya is the year when the future innovation will be available and affordable for the average person,

Yc is the current year,

N1 is the number of future years before Ya when relatively little work will be done to achieve the invention (or when the development work is not well funded),

N2 is the number of years of well-funded, full-time work needed to achieve the invention,

D1 is the number of years that the invention is delayed by regulatory delays or social restraints or cultural inhibitions,

D2 is the number of years between the time the invention is created, approved and accepted and the time the invention is affordable by the average person.

So now we have our finished Crystal Ball Formula. The good news is that we have created a more realistic way of assessing when future inventions will be available to the average person. The bad news is that this more realistic formula leads us to believe that lots of those cool futuristic things we are hoping to have one day may not be available to us until much later than we had hoped for.

For example, consider a youth pill, one that makes old people young. In this case we might imagine that N2 is 25, meaning that it would require 25 years of full-blast, well-funded work to create a youth pill. We might optimistically assume that N1 is 0 (as might be true if the pharmaceutical companies are working full blast on such an innovation, and keep doing so until they succeed). But to be realistic we might estimate D1 as 10, meaning that if someone invents a youth pill it is likely to be delayed for at least a decade by regulatory constraints and resistance from conservative opponents. We might also estimate D2 as 10, meaning that if someone invents a youth pill and gets it approved, it will probably be at least another ten years before it is affordable. So alas, you aren't likely to get a youth pill in your medicine cabinet until another 45 years or more, unless you're a millionaire.

Depressing as these more realistic estimates may be, they help us plan our lives better. If it's really going to be 45 years before you get that youth pill, you had better eat your vegetables now and cut down on your red meat now and stop smoking now if you hope to live long enough to use that youth pill.