Sustainable Success

I'm currently going through The Wizard And The Prophet by Charles C. Mann - a modern take on the challenges of population growth and environmentalism. In it, two philosophies on addressing population growth are presented, attributed to Wizards and Prophets:

• Wizards believe technology will allow humanity to reach higher populations than we could ever hope for by growing the pot so to speak.
• Prophets believe that a larger pot will only feed our appetite and that we must learn to reduce, slow down, and increase efficiency.

As a child, I was firmly in the wizard camp. Even as recently as 5-10 years ago, I was pretty much convinced that the AI singularity would surely happen in my lifetime, and all of the world's problems were guaranteed to be solved. Over time though, I have been introduced to some of the deep fundamental complexity that comes with math and physics, and had to come to grips with the fact that they are perhaps significantly more complex than I had originally thought. This has been enough to give me doubt over the timeline of progress, and the possibility of theoretical limits to technological progress. I have also learned more about the extent of side effects from our existing progress: Species being hunted to extinction, soil being drained of nutrients, the removal of old growth forests, the melting of permafrost, the introduction of invasive species. There is no fast 'undo' button for these effects. And the world is permanently worse because of them. Maybe there is something to this Prophet business after all. I mean what is the point in rushing for a quick victory when doing so might take away from what that eventual victory will include. Wouldn't it be better to just play it safe and slow, still making progress, but doing so in a deliberate, low-cost manner? Surely there must be a path to success which does not permanently deplete existing resources.

That is about as far as I have made it in the book(not far). So far it focuses mainly on a sustainable path for humanity, but the principle of sustainability extends far beyond just environmentalism. To give a more precise definition of sustainability, I will pass the definition off to the mathematical definition for system stability - the ability of a system to course-correct to maintain finite states over time. In other words, nothing should trend off endlessly in any direction. Many systems are not stable unless controlled in a particular way with additional calculated inputs to the system. My first experience in this space was building an 'inverted pendulum' in university - an attempt to sustain a system in a state only accessible by means of a controller.

In a lot of ways, this balancing act mirrors what we want out of our systems. These systems do not naturally tend towards a state that we desire. We want to keep them in that state, and so we add external inputs to the system. If we add external inputs in the right way, we have a good controller. If we add external inputs in the wrong way, we have a bad controller, and we end up with a system that trends forwards towards an extreme outcome. So what other sorts of systems might we want to curtail towards desirable states, and then sustain them in that state?

• You could have a sustainable relationship
• You could have a sustainable business
• You could have a sustainable diet
• You could have a sustainable path to happiness
• You could have sustainable personal finances

One analogy here would be for an individual considering taking financial risk by seeking huge wealth quickly by purchasing a volatile stock or cryptocurrency. Another might be someone risking their career by quitting their stable job to start a new one as an artist. Risks to one's health are another possibility as one could get injured playing concussion prone sport, develop health problems from diet, or develop an addiction to a drug. Surely an individual should be able to enjoy their life without risking permanent misery.

Ultimately, we want all of the valuable things that come from taking the risks that Wizards do. But we should be interested in long term success here, and we should make certain that we sacrifice as little as possible from our existing position when we take risks.

• Wizards produce more value than anyone else.
• Prophets consume less resources than anyone else.

Sustainable success requires both of these philosophies.

My habits for sustainable success:

Lots of little risks

Taking big risks is not a sustainable strategy for success. Consider the following game:

• You choose any amount of money to bet.
• You pick a single card from a deck of cards.
• The deck is shuffled and a random card is flipped.
• If it is not the card you guessed, you double your money.
• If it is the card you guessed, you lose your bet.

According to the logic of value creation, you should play this game repeatedly until the end of time because you have a good chance to make a huge win, meaning it has a large expected value.

Now consider the same game with one more rule:

• You must always bet everything.

The expected value has not changed, but it is no longer a sustainable strategy to play until the end of time because at some point you will lose everything and have no way to recover.

Make it a common practice to take risks, but make sure that your risks do not grow in size. If you find what was supposed to be a small risk is ballooning in size far beyond what was expected, it might be wiser to just cut losses and go back to smaller risks. This brings us to the next best practice:

Fail like you would Practice

"Fail fast" is a common mantra in tech. I like to take this one step further. A good failure plan is a good practice plan. Fail Fail often, fail openly, and fail deliberately. Stability controllers require feedback, even if it as simple as "something is wrong". Failure is not a bad thing! It is opportunity! It is the lifeblood of stability and without failing and sharing then you are not learning what not to do.

• Fail Deliberately: If you win every blind auction bid that you ever make, you are almost certainly bidding too high.
• Fail Openly: If you have never seen someone make a mistake even after a long history of work, it is more likely that they are hiding them than that they aren't there.
• Fail Often: If you put off failure for too long, when you eventually do it may be too large.

Following these failure principles will lead to a faster convergence on a good control system. Failing deliberately means that you will keep yourself grounded - giving you a better idea about how close you are to failure, and what that failure might look like or feel like when it is happening. Failing openly means not hiding your failures, and sharing them with others. This builds trust between yourself and others, and is an important step in being able to learn from others. Failing often means that you don't go too far or without exploring the area a bit more. This is important because sometimes failure in one space is very different than failure in another space. Treat failing like a skill that you must continually practice to improve at.

Always learn, teach, do

Always learning maintains neuroplasticisity, helping you learn faster in the future. I find it also keeps you grounded because you will be able to empathize with people who know less. It is important to start off poor at a task so that you can be rewarded for improving. Always doing something means that you get to feel a sense of accomplishment, and can directly create value for others. Always teaching helps spread knowledge and builds your communication skills. Teaching in the workplace is especially great - you can't be promoted unless you can be replaced!

80/20 Rule

What would a principle list be without an 80/20 rule? This 80/20 rule is fairly simple: Spend 80% of your resources on the best known option, and then spend 20% on experiments aimed at continuous improvement or other options that will push your boundaries. I find that this split allows me to maintain overall focus on the best path, but still gives a dedicated amount for experimentation.

Minimize Stuck Time

Sometimes things get stuck in an undesirable state. Relationships, careers, businesses, emotions... You might have a productive friend who gets stuck binge watching 8 seasons of some particular tv show. You might have a relationship that is stuck in a routine that isn't making anyone happy. You might have a business where most employees are just coming in to collect a pay cheque. You might be in an emotional slump. The controller keeps putting in the same input, and getting back the same undesirable output.

If a system is stuck in a particular state, it is important not to blame the the state or the system or the controller. It is often the combination of the three that leads to 'stuckness', rather than one thing in particular. You might have a fine controller, but in one particular system it has trouble in one particular state. I advise the following:

1. First, learn to quickly recognize when something is stuck. Look for a repeated pattern of action and little incentive for actors in the system to change.
2. Next, the best way to get unstuck is to force a state change. Do something to change the environment. Attempt this several times, trying a wide array of different directions or approaches before progressing to the next step.
3. Force a controller change. Change the inputs. Attempt this several times, trying a wide array of different directions or approaches before progressing to the next step, and after each attempt, try mixing in some state changes as well.
4. Force a system change. Change the rules of the game, or, ultimately decide not to play. If you can change the rules, expect to have to make controller and state changes as well.

Moving quickly between one attempt and the next is important, but in the interest of stability is often best to add a recovery period between each attempt.