Problem Solving & What to do with your Life
Work on hard and important problems, but which and how?
Sometimes just as I get down to sleep, in the time between setting my bed and closing my eyes - I often think about Life. Capital L Life. Sometimes philosophically, in the sense ‘Why do we exist’ and ‘What is the meaning of our existence’, but not a lot.1 What I almost always end up thinking about, which ironically in turn makes me sleepless is ‘Which problem am I working on?’.
If you enjoy reading pieces like this, do consider subscribing to this newsletter to receive interesting posts like this straight into your inbox. I write about a range of issues from ideas, mental models, book reviews to policy analysis, law and technology etc. It’d be great to have you over!
Problem Solving
Problems abound the world. But what kind of problems am I talking about? The ones that appear in your mathematics textbooks? Or the ones in your day to day life? Well, as much as they do seem hard enough, neither. I’m talking about important problems that if solved would have a positive impact on a large number of people.
There are three qualifiers here. First, that it has to be important. Second, it has to have a positive impact. Third it has to impact a large number of people.
Important problems are often tied up in orders of magnitude and impact. Which is to say that most problems are important either because they affect a large number of population, or for however number of people it affects, has really harmful outcomes.
So, we have ourselves a problem matrix.
The Y axis is magnitude i.e. how many people does that problem affect. The X axis is Impact i.e. for however many people it affects, how harmful is that problem. As we can see we have four types of problem.
Arranging problems in order of difficulty and importance would lead us to the following classification:
Large magnitude, extremely harmful problems: As mentioned in the matrix, these are the world’s most important and urgent problems. Most of them are what we call as existential problems2 i.e. problems that if not solved, are likely to lead to the end of our civilization or cause substantial harm at scale. Even though it is a loose category, problems like climate change, pollution, water conservation form its core. Not to be confused with existential risks.3 Other very important and hard to solve problems are of bringing electricity to more people,4 taking internet to more people, improving world healthcare, educating more people, and taking more people out of poverty.
Small magnitude, extremely harmful problems: These are problems that affect a small number of population (of the whole), but are extremely harmful. Most problems here are problems of disease. Most of us are not likely to suffer from these diseases, but because they are so extremely harmful for the people that they do end up affecting, they are a priority problem to solve.
Actually the third and fourth quadrants are a very interesting space. A space I call the ideas space.5 Let’s first see what is the characteristics of this space are according to the matrix.
The upper right quadrant is small magnitude mildly harmful problems. A core part of problems that reside here is that they are almost always perceived as non problems or seem to be invisible as a problem. What I mean by that is since these problems do not have a extremely harmful impact, we do not seem to think of them as an issue in the first place. Take Uber for example. The problem it solved was the inability to hire taxis on demand. Now in a pre Uber world, for most of the people that was an inconvenience which existed because ‘that’s how it was’. Most people did not either think of it as a problem, or even if they thought of it as a problem, not a solvable problem at least. But it still did have mildly harmful effects on people such as the increased time cost, and other causal problems that arose from being unable to book a cab on demand. It was just that for the people facing this issue, the world they lived in was static with all its problems and issues.
Before I move on to the y axis i.e. scale of the problem, there is another reason why I classify these problems (or startup ideas to problem solving) as mildly harmful. Even though Uber has a market cap of 93 B dollars, a world without Uber would not be catastrophic in any way. Which is why a lot of the problems that a lot of start ups solve are mildly harmful problems.6 However, the more interesting aspect about scale in these statrupesque problems in the idea space is that the number of people for whom the problem is solved initially are few in number, but it tends to scale up later. Take desktop computers as an example. The initial people who used it were rich people who could afford to pay for that kind of stuff. But as technology developed, the price became cheaper and desktop computers came to be widely used. Over time, it started with solving only a few people’s mildly harmful problem (computation, storage, drafting etc.) and then went on to solve this for a large number of people. Therefore most of the problems that start in the upper right quadrant actually are fourth quadrant type of problems which becomes apparent once the start up trying to solve it achieves scale. Between the quadrants, there are no defined lines, and often things move from one section to another.7
Applying yourself to Problems
The non exhaustive problems contained within the previous matrix are some of the particularly hard problems which attract the world’s best minds. Such kinds of problems are important because they matter on a sufficiently large scale or effect that makes them a priority to solve.
Even though some of these problems are economic in nature i.e. related to how can we better spend money to generate the maximum beneficial output; most of the world’s most pertinent and important problems are problems of engineering. Given that most of these problems are engineering and innovation, it is not a surprise that the world’s smartest people are in hard sciences. The reason why innovation is so tied up with technology and not policy is because a good policy might have a large positive impact, but it is technology that solves a sufficiently hard problem for the most amount of people over a long enough time period. Which leads us to a quite unpopular truth that we all subconsciously know, but do not like to accept. The engine to move the world forward is hard sciences. Only hard sciences. So, if you want to contribute to solving the world’s most important problems, you need to study science. Hard science. Not political science, arts, law, literature or commerce.8 It is a hard truth, and even though many people will draw long parallels or try to prove how sociology, history or literature advance the world, they do not. Hard sciences like Mathematics, physics, chemistry, biology etc. do. Other fields have their place in the catalogue of episteme, but it is an illusion to say that they are hard science’s equal in importance or impact.
Going back to the interval between when I set my bed and before I go to sleep, I almost always think about this. I ask myself if I am working on a real problem in any way and my answer is always no. The only problems I have been working on for 22 years of my life have been my own. And that’s somewhat shameful. Born into the internet age and with an above average IQ, I was, and still am placed with almost all the things I need to apply my skills and talent to world’s hard problems. And yet I have never solved any real problem. Far from solving, I have never even tried to solve any problem. Worse, I have not even been able to assess something as a problem in the first place. And as unfortunate as it might sound, it is probably your story too. A tale of wasting immense human potential.
My learning after thinking hard and long on this subject was to be clear on the idea that if you actually want to be resourceful to the world, the first thing you need to do is to find out which hard problem you want to work on. There are numerous reasons why you should apply yourself to hard problems, but I will list two.
First, if you do not apply yourself to these important and hard problems, you are likely to under utilize your intellectual capacity. The kind of problems you are likely to encounter while working on these problems are going to be of order of several magnitudes tougher than you next most difficult problem (which might be personal finances, or how to finance your child’s education, or how to secure a good job etc.). The intellectual rigor of such a task is unmatched. There is a reason why the CEOs of the most impactful companies are very smart people. There is a feedback loop between their own curiosity that leads them to certain problems and what solving hard problems requires them to read/know. Most knowledge is Yak Shave. One of my favorite CEOs is Stripe CEO Patrick Collison, and I’d recommend going to his website to find out why.
The second reason is a bit more elaborate. If you are a problem solver, you have two options. Either you can join the people or organizations who are already working on a problem that captures you intellectual interest, or you could do it yourself i.e. open a start up. Now, the first option carries with it benefits of a stable salary, a somewhat set career path, promotion guarantees etc. The second option is far riskier. But as Vinod Khosla, ex CEO of Sun microsystems points out in this talk, innovation never comes from established companies, it always comes from the start ups.9 I have a special corner in my heart for entrepreneurs and start ups, because they not only solve exciting new problems, but also because they generate wealth.10
Here’s what Paul Graham says about founder’s motivation and why it is almost always curiosity driven problem solving and not monetary benefits:
Here’s what he says about why startups play non zero games11:
These excerpts are taken from his essay, Mean People Fail.
Orders of thinkers
I have been thinking about problem solvers and how they think about problems. That has lead me to classifying them into five categories according to the scale and importance of the problems they aim to solve. Here is my classification:
0 order thinker: People who only solve day to day usual problems of their own and of their family. 99.99% of population comes under this.
1st order thinker: People who show some sign of ‘community based’ thinking. However the problems they aim to solve are more in nature of hinderances than actual problem. For example, a RWA secretary writing to the local MLA to fill a pit that has been causing parking issues.
2nd order thinker: People who solve the same kind of problems/defects as 1st order thinkers, just on a larger scale. Eg: A slightly influential merchant trying to contact political persons to try to solve the issue of intermittent power supply in their ward.
3rd order thinker: People who aim to solve actual hard problems such as water availability, child nutrition, school drop out rates, drug usage, better electricity supply, scaling up internet etc. There are degrees within third order thinkers according to the scale at which they’re trying to solve the problem at. For example, a headmaster of a local village school who has worked for many years on girl child education is also a third order thinker, as is someone like Kailash Satyarthi who has won the Nobel prize for his work on countering child trafficking and abuse globally. The only difference between these two is the scale at which they are operating. The nature of problems is same.
4th order thinkers: These are people trying to solve really important existential problems. These are the type of problems that if not solved will lead to the end of humanity or will lead to civilization scale harms. They are the most notorious problems in the bottom left quadrant of our matrix - ‘large magnitude, extremely harmful’. Someone like Elon Musk would be considered 4th order thinkers. If we have to save our planet and civilization, we need more.
Most of us are born, live our lives and die being zero order thinkers. And perhaps there is some relief in not having to deal with such hard problems in addition to one’s own. But it is only higher order thinkers who do anything of use for the world. Otherwise, as mere mortals we have zero impact on the world and leave the world as we found it. With zero cosmic footprint. Insignificant.
Peter Thiel in his book Zero to One makes the point that ‘development and new technology has never been an automatic feature of history’. The underlying argument is that development is not a given in any manner. The reason why we have come from computers that occupied a large room and could barely do high order calculation to cheap smartphones which can do operations at speed and scale thousands of time better, is because very smart people spent very long time thinking and attempting to solve these very hard problems. Most of the prosperity around us is due to a handful of smart people applying themselves to hard problems over a ling time. Elon musk makes this argument here. Imagine what would have happened if they lived where the rest 99.99% of us live.
What to work on?
The natural question to ask after reading this, or thinking about hard problems is - ‘Which hard problem do I want to work on?’. For me, it is an ongoing assessment and I haven’t yet picked the problems I want to work on. Of course, due to the restrictions that my domain puts on me, it is likely to be at the intersection of law, technology, finance and policy. Which in turn makes me ineligible to be working on the most important problems of the world which lie in hard sciences. Alas. Sometimes I wonder if I should have made the switch.12
For starters, here is one list of ‘World’s hardest problems’. You can take your pick amongst them. As you might have guessed a lot of these are hard science issues, and there’s only so much that non science people can contribute to the core. But there is plenty of room available at the periphery. And it is needless to say that for the core to work well, the periphery has to work even better. So take you pick, and work on edges.13
A common advice is that you don’t always have to think in terms of scale when you’re thinking of problem solving. Even if you make something better for a limited number of people, that is still a net positive. Many argue that actually the way to solve problems at scale is to first solve your own problems, and then other problems at limited scale. Paul Graham’s advise is to ‘Do things that don’t scale’.14
A formula that Elon musk works by to decide what to work on is: utility delta * Number of people it will effect. Delta is a mathematical operator (operation) that signifies the rate of change of a quantity with respect to some other quantity. In this case utility delta means how much increase in any kind of utility does the product or idea bring. The reason why this formula works is that if you do something that is really useful but only for few people, that’s great. If you do something that is moderately useful but for a lot of people, that’s great too. If we go back to the hazy area between third and fourth quadrant, we can see that most of the ideas that first solve a problem for a few number of people are likely to solve that problem for a larger number of people later. Scaling is organic.
If you want more perspective on what you should work on, I’d recommend reading Alexay Guzey’s piece titled ‘What should you do with your Life: Directions and Advice’.
In the end, all we can hope for is that we find something that we are truly passionate and excited about. Something that makes us want to think and work, not happy. Finding happiness in work is almost always a lost cause. For hard problems will frustrate you and trouble you. But the real aim is to simply not be bored. To be lively and excited. I’ll end this essay with a few words from Paul Graham that he wrote in his essay ‘A project of one’s own’.
And Sam Altman:
Pick your hard problem, decide how you are going to attempt to solve it and spend time solving it. Who knows, you might bring the next change that revolutionizes this planet. Not applying yourself to hard problems is a waste of your intellectual capacity and a disrespect to your capabilities. Respect them and put them to use.
Here is a great thread on how start up founders come up with ideas.
If you liked this piece, you can also share this post with your friends, family and loved ones by clicking here:
I would love to hear from you! Leave your comments down below.
This reminds me of the time when my law of crimes professor once asked us point blank, ‘What is justice?’. As you must have imagined, she was met with silence, until the class rescuers (students who chime in when nobody wants to say anything when asked to do so) started with their version of it. If you have any idea of what justice is, I’d recommend contacting Michael Sandals. Poor chap wrote a whole book trying to figure it out.
Oxford University runs a whole institute centered around the idea of existential risks, their study and how to mitigate them. It is called Future Of Humanity institute, and it aims to ‘bring the tools of mathematics, philosophy and social sciences to bear on big-picture questions about humanity and its prospects.’
Existential risks are those events or processes that only need to go wrong once to end all of humanity. There is a fantastic podcast on precisely this topic that ‘explores the future of humanity and finds dangers we have never encountered before lurking just ahead.’ It is called The End of the World with Josh Clark.
Which would in turn require innovation in generation, transmission and storage of power. This is an issue I personally face daily. There is an acute power shortage issue where I live. Sometimes I only get 10-12 hours of electricity in a day. My sister, who is a Masters Candidate in Sustainable and renewable energy has been working on and around this issue, and we hope to write a piece about it soon.
Patrick Collison, the CEO of Stripe once said in response to a question about the culture of decision making at Stripe if they optimize for sameness in thinking for more uniform decision making said, ‘in the multidimensional space of ideas and philosophies, there is a stripe blob and I may be closer to the center of mass because I have been with it longer, but it (the blob) is distinct and it is shifting and changing, and it ought to change’. So yes, idea space.
Which is not to say that there aren’t start up which are not solving extremely harmful problems. See here.
I do not think it happens for ALL of the startups. Take CRED. It is a service which helps people pay credit bills on time by reminding its users and storing all bills in one place. But is is only available to a small number of people called ‘high trust individuals’ who have a credit score of more than 750 points. CRED is unlikely to move from third to fourth quadrant simply because credit worthiness cannot be increased artificially. It is not a market that can grow by mere adoption since credit worthiness is an externality CRED cannot increase.
These fields have their own ‘hard problems’. But most of these problems are problems of philosophy, with non visible and probably non existent implications for how we advance the world and make it more prosperous. Law and political science has long struggled to find an answer to questions like ‘what is justice?’. But anyone can tell you that these are not hard problems, these are unsolvable problems with no definite answer. Pontification does not advance the world. See here.
Vinod gives an interesting explanation. He says that because most of established business runs on a model that has already proven successful, they are not incentivized to change anything about their business model. However future is all about changes, and how to do things better - it is only start ups how have the vision of change.
On a comparative, if you are a salaried employee, you cannot have much impact on wealth generation or even injecting liquidity into the market through spending. There is only as much consumerism you can indulge in. Start ups on the other hand, create jobs and transactions that did not exist earlier. Start ups are always non zero sum game.
Non zero sum games essentially means an arrangement in which one someone does not have to lose in order for the other to win. In other words, it is a win win situation. Zero sum games on the other hand are arrangements in which someone will have to lose in order for someone to win.
For context, I hold a physics honors degree from Sri Venkateswara College, University of Delhi. Before that I studied Physics, Chemistry, Mathematics and Biology in my +2. After graduation I decided to shift streams and took up Law.
Most of the smart people I know say that the real innovation and knowledge creation happens at the boundaries and intersections of field. That is where the excitement lies. Find your periphery.
As you might have noticed, Paul’s influence on this piece has been immense. That is because I have been listening to a lot of his talks and reading a lot of his work. The man in simply brilliant. I am in shock that I came across his work so late. I thank my friend Sandesh Atyam for suggesting his works to me.