Nathan Wailes - Blog - GitHub - LinkedIn - Patreon - Reddit - Stack Overflow - Twitter - YouTube

Predicting the Future

Owned by Nathan Wailes
Last updated: May 29, 2023
8 min read

Table of contents


How to Predict the Future

1. Identify a trend



2. Identify potential barriers to that trend


3. Ask yourself, "What would the cost of this new invention be relative to cheaper alternatives?"

  • The goal is to try to use some deductive reasoning to figure out what constraints in the world would need to be overcome.
  • Example: Underwater cities are something predicted, but their cost relative to living on the surface is so great that no one has ever done it.
  • Example: Flying cars are expensive relative to 


Learning resources


Books

Misc articles

People

  • William Gibson
    • The future is already here – it's just not evenly distributed.
      • The question I'm immediately led to ask is, "How can we determine which innovations / variations are going to spread to the rest of the world and which will fizzle out?"
  • Daniel Kahneman
  • Mark Cuban on predicting the future
    • http://www.youtube.com/watch?v=gnLCEELm ... ed#t=6m11s
      - Learn the technology that's out there; learn what it can do now and what it may be able to do in the near future
      - Put yourself in the shoes of various businesses; how would you do things differently if you were in their shoes?
      - Use the past as a source of analogies; for example, when evaluating whether businesses could use high-definition screens, Cuban acknowledged that they were currently very expensive ($10,000) but also saw that PCs had once been very expensive but had rapidly become cheap enough to become widely used.

Superforecasting tips

  • For his initial forecasts, Tim takes less time than some other top forecasters. “I typically spend five to fifteen minutes, which means maybe an hour or so total when a new batch of six or seven questions come out,” he said. But the next day, he’ll come back, take another look, and form a second opinion. He also trawls for contrary evidence on the Internet. And he does this five days a week. All that exploration makes him change his mind a lot. “I update constantly,” he said. “That’s just the way my mind works, although more typically it applies to real work as opposed to [tournament] questions.” 15 By the time a question closes, Tim has usually made a dozen forecasts. Sometimes, the total is closer to forty or fifty. On one question— whether the United States and Afghanistan would reach an agreement on the continued presence of American troops— he made seventy-seven forecasts. It may look as though Captain Minto is sailing straight for the Charybdis of overreaction. But I haven’t yet mentioned the magnitude of his constant course corrections. In almost every case they are small. And that makes a big difference.

  • Edge.org - Edge Master Class 2015: A Short Course in Superforecasting, Class V

How Predictions Go Wrong


Failing to Anticipate a Different Type of Solution to a Problem

Examples:
- Henry Ford's famous quote: "If you ask people what they want, they'll say 'a faster horse'." In other words, people often don't imagine a radically new way of doing things (eg using cars instead of horses).
- Peter Thiel's famous quote: "We wanted flying cars, instead we got 140 characters." And Andreesen (I think?) pointed out that the REASON people wanted flying cars was to get places faster, but in fact we were able to come up with a much better solution, which eliminates the need to travel in the first place. Our stuff is delivered to our door instead of us needing to go out to buy it. We can visit relatives via video calls instead of needing to physically go visit them.



Failing to Anticipate a Barrier

Failing to anticipate a barrier from a person/group that has an interest in that prediction not coming true

Examples:
- People anticipating a crisis after the '08 crash failed to anticipate that the US government would be able to prevent it.
- People backing Barack Obama in '08 expected a change in how government worked but failed to anticipate Republicans in Congress blocking every initiative.


Failing to anticipate a barrier in the laws of physics

Examples:
- Francis Bacon failed to anticipate that a law of physics would make it impossible to have machines of perpetual motion.
- Space movies / books (Star Trek, Star Wars, alien invasion movies) may have failed to anticipate that the laws of physics may make it impossible to travel across the universe very quickly.
- Back to the Future II failed to anticipate that the laws of physics may make antigravity impossible (hoverboards, hovercars)


Failing to anticipate a cost-benefit barrier (where the cost of developing / using an invention is much higher than the benefit)

Examples:
- Woz: "We thought computers were going to be used for all these weird things--strange geeky things like controlling the lights in your house--and that turned out not to be the case." Nathan: While this is starting to happen now (eg movement-detection lights at 1776), it isn't widespread because flipping a lightswitch often isn't such a high-cost activity that it is worth paying for computer-controlled lights.


My predictions of the future

  • self-driving cars will become very common
  • education will be made available to 99% of children in the world; they'll have a laptop and will be able to learn english and useful skills over the internet w/ movies and fun games
  • a bunch of my predictions are things that are already happening, and I'm just predicting what the end result of these trends will look like.
  • apparently transportation will become extremely expensive, possibly making the world of the future much more city-oriented / densely populated (ie people will not be spread out all over the place as much as nowadays). [source: oil 101]
  • if transportation becomes extremely expensive, there may be a new move toward local manufacturing, or an ease of wage laws so that cheap labor can be brought in from abroad. idk, i'll need to think about this.
  • education will become fun, extremely cheap, and of much more similar & higher quality for everyone. people will be learning via computer games.
  • english will become even more prevalent in other countries (eg developing countries). this will result in more competition for jobs between americans and foreigners.
  • people are going to be fully entering the workforce earlier, like age 12 or before; it's too wasteful to have people doing nothing until they're 21/22. competition from other countries will force people to educate themselves more efficiently.
  • general prediction: things will become faster, easier, and less stressful
  • it seems cameras and microphones are getting smaller, with better quality, and cheaper. it seems that in the future cameras will be placed in many more places and people may be more concerned about their privacy than they are now. i need to find that quote of someone famous saying that the next big wave was the reduction in price of detectors of all kinds (cameras, audio, etc.). I think it may have been that HLS professor...
  • Crowdsourcing is going to be a big deal
    • Investing / stock market trading
      • Numerai

        • Numerai is a global artificial intelligence tournament to predict the stock market. You download our data, build a model, and upload your predictions. Because we encrypt our datasets with structure-preserving encryption techniques, we're able to give away all of our data for free. This turns the stock market into a machine learning problem — so you don’t need to know anything about finance to participate.


Things to meditate on / think about

  • http://donottouch.org/ - What kinds of things could be done with that kind of real-time collaboration of hundreds of people (or more)? I found this video to be a pretty powerful experience, it reminded me of being in the band and going to a baptist service. I find there to be something emotionally powerful about being part of a large group of people all acting together as one giant organism. But aside from the emotional part, there could be practical applications.
    • I think the commenting system of Reddit / Facebook already kind of allows for the large-scale collaboration in that everyone can post a comment, but the collaboration is very crude. Comments are just listed one after the other, and people can upvote.
    • The thing is, the video shows a semi-simulated very-rapid collaboration (it's simulated because not everyone is acting at the same time). So the question is, what could be done with such rapid collaboration that cannot be done with the kind of collaboration we already see?

Others' predictions of the future

Sci-fi authors' predictions of the future

  • Charles Stross
    • rec'd by Naveed
  • Robert A. Heinlein
    • rec'd by Naveed


Prediction websites

A list of prediction markets



When–if ever–is it possible to calibrate your subjective probabilities?

  • StackExchange - Why is it possible to calibrate your subjective probabilities?
    • Calibration isn't something fixed from birth; you can train yourself to be better-calibrated by playing a game where you assign a probabilities to a bunch of things and then you're scored by how well-calibrated your subjective probabilities are. But my question is, why is it possible to become well-calibrated using such a game? It needs to already be true that you're right more often about statements that you're more confident in than statements you're less confident in. But why should this be the case? Are there any decision-theoretic assumptions or rationality axioms that imply this is the case? Has it been empirically observed that humans are more often right about what they're more confident in?

Misc ideas




Weather forecasting


http://en.wikipedia.org/wiki/Weather_forecasting
"The chaotic nature of the atmosphere, the massive computational power required to solve the equations that describe the atmosphere, error involved in measuring the initial conditions, and an incomplete understanding of atmospheric processes mean that forecasts become less accurate as the difference in current time and the time for which the forecast is being made (the range of the forecast) increases."
- how analogous is that to investing? if it's analogous, how should that affect investors?

Books about Weather / Meteorology

Books I've found on Amazon:
Meteorology Today by C. Donald Ahrens
The Weather Revolution
Atmosphere, Climate, and Change
Meteorology for Scientists and Engineers
Introduction to Theoretical Meteorology
A First Course in Atmospheric Thermodynamics
An Introduction to Dynamic Meteorology, Fourth Edition
A Short Course in Cloud Physics
Atmospheric Science, Second Edition: An Introductory Survey
The Atmosphere: An Introduction to Meteorology



Invisible in the Storm: The Role of Mathematics in Understanding Weather
http://www.amazon.com/Invisible-Storm-M ... 578&sr=1-1
Review: http://www.theactuary.com/arts/book-review-september/
Review: http://www.harvard.com/book/invisible_i ... g_weather/
Review: http://www.thenational.ae/arts-culture/ ... -endeavour

"forecasters would have to live with the fact that very small uncertainties in our data of what the weather is like now can lead to very large errors in the calculations of the forecast a few days ahead" [...] There are "many detailed interactions, and there are degrees of unpredictability. But there are also many stabilising mechanisms..."






Follow me on Twitter