One way to put a banner on the course of human civilization is to call it an escape from uncertainty ( most of which would have led us to death). Why do we live in societies? To save ourselves from the uncertainty of getting attacked... Why do we build houses? To mitigate the uncertainty of getting our food washed away in the rain... Why do we rear children? So that we don't fast ourselves to death in old age. So in all these cases, the sheer uncertainty in the future has made us make expensive investments in the present. And those who didn't give in to these fears took a risk.
The pay-off of that risk is, however, independent of the past. In some cases, the risk turned out to be a golden gamble. Just like a flood washing away houses of people...the ones who hadn't built their houses were a little less sad. Risk can give an unexpected reward, but the higher the risk, the greater the return.
But at times, the risk is otherwise. Instead of choice for the individual, it becomes a calamity enforced upon him. Say a flood, so how do all homeowners save themselves from this risk? In other words, how do institutions mitigate the high losses of the risk? The short answer is insurance. They keep collecting a premium from all members of the society and then channel them to the one who faces a dis-fortune. But, here, we need a convincing algorithm to convince people to keep pooling their money and then give it away to someone else. The answer is in statistics.
Unlike other forms of mathematics that provide a language for existing things in nature or human logic, statistics take a step into the future. Statistics does not, but our interpretation of it does. When figuring out gravity, physicists devised equations from the first principles ( sometimes intuition ). Then solved those to arrive at concrete solutions, and then a plethora of experiments were fired at those theories validating their genuineness. However, after some time, this exercise became tedious; there were far too many things to consider, emergent behaviour made the first principle approach void, and intuition needed to be more reliable. The other way around is to come up with empirical solutions from the large number of experiments done.
This, in a way, is looking at the past, finding a pattern, and then taking a shot at the future. This is the discourse of the engineering world today. The three decisive factors here are
- The past ( the data that we have)
- the prediction ( how we analyse that data to find a pattern)
- the future ( how does our prediction fit the real world)
So, the products of the modern era have tried to excel at these fronts, to get to know their consumer's data better, and then employ the fastest computers to predict and reduce the error.
The second and more remarkable change is in the product we sell. There is a rising demand for products that meet aspirations rather than needs. For example, more people would like to flaunt a self-driving car, which has a greater probability of crashing when given at a lower price. This shows that we are selling a fixed feature and a box that captures the feature and the fuzziness around it. The lesser the fuzziness, the higher the price.
Also, the consumer today is less likely to be loyal. Since the risk of choosing a new brand is calculated in an economic sense. Thus, we have a quirky customer looking for features ( even with errors ). So what does it say about the future?
We are in for a battle for crazy ideas and realising them. We can take risks and experiment with an awareness that there is a market for that. And that we, as a customer ourselves, need to be mitigating our risk through the instruments at our disposal.