3. But what are First Principles?
WORDS TO LIVE BY
Elon Musk, CEO of SpaceX and Tesla is notoriously popular for re-instating the age-old Aristotle approach to solving problems in his endeavours. That approach is thinking from first principles. I will explain more in detail, but first, try to grasp the concept Elon is trying to get at in the following discussion he was having with Kevin Rose in an interview in 2012:
Elon: I think it’s also important to reason from first principles rather than by analogy. So, the normal way we conduct our lives is we reason by analogy. We’re doing this because it’s like something else that was done or it’s like what other people are doing.
Kevin: “Me too” type ideas?
Elon: Yeah, yeah…slight iterations on a theme… and it’s kind of mentally easier to reason by analogy rather than from first principles. First principles are kind of a physics way of looking at the world and what that really means is, you kind of boil things down to the most fundamental truths and say: “Ok, what are we sure is true? Or, sure as possible is true?” and then, reason up from there. That takes a lot more mental energy.
Kevin: Give me an example like what’s one thing that you’ve done that on, that you feel works for you?
Elon: Sure. So, somebody could say and in fact people do, that “Battery packs are really expensive and that’s just the way they’ll always be, cause that’s the way they’ve been in the past.” Well, no… that’s pretty dumb, you know? Because if you apply that reasoning to anything new, then you wouldn’t be able to ever get to that new thing. So, it’s like you can’t say “Nobody wants a car because horses are great and we’re used to them and they can eat grass, there’s lots of grass all over the place, and you know there’s no gasoline people can buy, so people are never going to get cars”. People did say that. And for batteries…they would say “Oh, it is going to cost you know, historically it has cost $600/kWh and so it’s not going to be much better than that in the future”. You say “No, ok…what are the batteries made of?” So, first principles would say: “What are the material constituents of the batteries? What is the spot market value of the material constituents? It’s got cobalt, nickel, aluminum, carbon and some polymers for separation and steel can.” So, break that down on a material basis and say: “Ok, what if we bought that in London Metal Exchange… what would each of those things cost? Like oh jeez, its $80/kWh!” So, clearly you just need to think of clever ways to take those materials and combine them in the shape of a battery cell and you can have batteries that are much, much cheaper than anyone realizes.
HOW TO THINK
Aristotle, over 2000 years ago in his book Metaphysics, defined a first principle as “the first basis from which a thing is known.” It is the most fundamental assumption that cannot be deduced from any other proposition/assumption. Rather, it is the surest thing we know of, from which we can deduce some other truths.
When practically applying this idea in terms of thinking approach, we need to break a design/process down into the core elements and then afterwards, put them all back together in a way such that effectivity is maximized. It is an iterative process of deconstructing something and then reconstructing it back, but only finding you arrived at a better design/process than before. This is exactly what Elon Musk was talking about above.
Let’s try to decipher how thinking in terms of first principles may have been like for Elon Musk and his team for creating the lithium batteries for Tesla. The questions listed may have been asked like a chain and the possible next steps are purely my take on what may have been done. The important point is the takeaway of the idea behind how you should have the first principles thinking mindset.
The cyclical flowchart below depicts how the battery problems of having high cost, low energy density and low battery lifespan in status quo might have been tackled via first principles thinking in Elon’s team.
I designed the chart as a sort of pathway you go through by asking the nuanced, appropriate questions and solving those separately (or even simultaneously if possible) until you reach an optimal balance of desired outputs that have the capability of changing lifestyle. By focusing on actual internal components of the battery and if they can be optimized for lower price point while keeping high energy density in mind, Elon’s team tries to solve the high cost and low energy density issue. By thinking of charge states and efficient cooling systems, it helps in resolving the low battery lifespan problem.
Combining them all to form the most effective design for performance by at least an order of magnitude (10X) means a significant improvement, enough to be framed as an “innovation” within the lithium-ion battery technology (more on this later in the chapter). It would be an innovation if this reduction in battery price while meeting functionality goals, leads to the final price of the cars to be such that it enables users to enjoy a higher relative affordability, maximize their future cost benefits and travel faster sustainably compared to existing gasoline powered cars.
However, note that if we can ever find an entirely different method other than lithium-ion batteries to power an EV that truly works, and if that turns out to be more effective, then that would be a whole new level of innovation. For instance, there had been a lot of discussion on hydrogen fuel cell powered cars (cars that produce its’ own electricity which powers the electric motors through reaction in fuel cell between hydrogen and oxygen to produce only water and heat as by-product). Currently, due to several cost and storage issues of hydrogen, we do not know yet if this method can be pursued as well as compared to current lithium-ion technology at a mass scale. Thus, we divert our focus to making the best possible batteries.
If Elon were to go by history (or by reason of analogy according to him), then he would have contacted battery manufacturers and purchased current ones in huge quantities only to have a bulk volume discount. He would then think of attaching them in his designed cars and sell at a price maybe lower than other companies but would not be so great for value to users that they would replace it with their own internal combustion engine cars.
However, Musk knew that by breaking down a battery in its constituents he can understand the actual chemistry of it and maybe discover unnecessary or suboptimal contents in the batteries which could be tweaked or removed entirely. Additionally, he may find contents that are missing which could make battery operate way better. All these design changes while ensuring cost is never increased but rather aimed to be as low as possible, changed the whole game. He found a new way to construct the lithium-ion batteries which could be manufactured at such low costs never imagined possible while achieving high energy density and high battery lifespan.
When it got out in 2012, the Tesla Model S (base model) had a max range of 208 miles and a useful battery life of 8 years/125,000 miles with a minimum battery capacity retention of 70% by end of design life; it is not dead yet but only lacks range. To put this into perspective, when Nissan Leaf debuted in 2010, which was also a battery powered EV, it had a range of only 73 miles and within 2 years time, its’ battery capacity retention dropped by 27.5%. As you can see, there was a huge difference in what Elon Musk could potentially achieve. Today, the Tesla Model S has a range of 405 miles. All because of thinking from first principles.
GREAT IMPACTS
You may think how much of an impact an innovation can have. Well, let’s find out. If you look at the volume-weighted average prices of lithium-ion batteries over the last decade (basically $/kWh), you will notice how much of an impact an innovation like this has on the overall market competition and demand, leading to more supply and production of such batteries, allowing for industry to enjoy economies of scale as the high price is severely reduced by ~10X:
Now, this is not to say that electric vehicles are currently the best mode of transport since regardless of how cheap/great the battery is, the fear that most people may have once their battery is at 5% charge is: “Where is the nearest electric charge station?” The lack of infrastructure of available electric charge stations as compared to traditionally present gas stations poses a big challenge to EV owners.
To counteract this from his side, Elon brought about his Powerwall invention in 2015 that would allow an EV charging station to be affixed at your home, powered by solar panels that are integrated in your house—aiming for a sustainable and worry-free situation for users. From regulatory and industry growth perspective, the Infrastructure Investment and Jobs Act of 2021 will provide support for the electric vehicle industry—$550 billion in funding will be allocated to invest in a network of EV charging stations across the U.S.
These are the steps which needs to be taken to allow for the EV industry to scale, expand and bring the ultimate benefits to users. These may take time, but the fact that someone’s trigger led to a technology breakthrough followed by several companies taking similar steps, enables a sort of lubricant in the supply chain process whereby industry average prices may boil down…a concept which no one thought was possible for the stubborn lithium-ion batteries…until someone knew that thinking from first principles can entirely solve the problem for good.
INSIGHTS AND CORRELATION
A great example of first principles thinking by deconstructing the problem down into core elements and then reasoning up from there to innovate can be explained through a short example developed by John Boyd, a fighter pilot and military strategist. The following has been taken from an article by the author Taylor Pearson who expands upon Boyd’s strategy:
Imagine three separate scenes—a motorboat towing a skier behind it, a tank rolling across the desert, and bicycle cruising down the street. If you break them down into the constituent parts, you have:
motorboat: a hull, outboard motor, and a set of skis being towed behind it
tank: treads, a gun, and armor
bicycle: wheels, handlebars, and gears
You can use these constituent parts to make many different incoherent wholes, but a coherent and useful whole would be a snowmobile: you take the treads from the tank, an outboard motor and skis from the boat, and handlebars from the bike and combine them to make a snowmobile.
It requires taking all the data you’ve gathered in the observation phase, breaking it down deductively into its constituent parts and then recombining those parts through creative synthesis to form a new model of reality that lets you make better decisions and actions.
The people that had all the information on the mortgage market in 2006 and still lost money failed to take apart their existing beliefs about how the mortgage market worked and see the data through a different framework.
One group saw unending market growth and the other group saw a bubble waiting to pop.
The point of this example was to allow for us to understand that innovation in a certain sphere lies within a deeper analysis of the fundamental constituents that make it up and how and why they interact the way they do, without having to accept that things can’t go better than the current status quo. The more we understand how things operate on a micro level, it is easier to re-group those and formulate a macro level operation that is more effective and sustainable. The methodology used when building the snowmobile out of a motorboat, tank, and bicycle are the same when creating the next generation lithium-ion batteries.
Consider that people throughout history had been riding in carriages with wheels and knew how to make their baggage and much more from leather, but who thought of combining the “wheels” from the carriages and the “baggage” they carried to create a sort of rolling suitcase? It was not until 1970 that Bernard Sadow patented the first luggage with wheels. You did not have to carry such large weight by hand or drag it anymore! Why it took that much time says a lot about how deluded we are into reason by analogy than thinking from first principles.
Elon Musk on the other hand was doing the same thing as Bernard, but instead he was trying to combine the optimal set of materials and amount (Bernard’s case: 2 relatively small wheels under 1 big baggage) from established periodic table, research, and market analysis (Bernard’s case: past time inspiration of carriage/car components, leather made bag products and mechanical linkages) that would exhibit desired energy density, range, and price (Bernard’s case: Least human effort and fast travel).
MAKING SENSE
As you can understand, this way of thinking has paved path for innovation. When I say innovation, I always want you to think about how impactful it is to users/how much lifestyle has changed since it is for them that we are making the products and not only us. This brings me to clarify further on the concept of improvement and innovation.
Throughout the last article, I have instilled the idea that innovation should be the aim once improvements are no longer worthwhile. An important note must be made regarding improvements. In the last article, most of the places where I have been criticizing improvements, I was referring to “Continuous Improvements” which do not really have a great impact on users (e.g., making the 30:1 gearbox bit stronger and compact). However, interestingly as you may have noticed in this article, I have mentioned places of order of magnitude improvement, i.e., “10X Improvements”, which have the potential to change lifestyle (e.g., Elon’s lithium-ion batteries) and when it does extravagantly, we will refer to this as innovation.
Now, let’s go a little deeper down the rabbit hole. Innovation itself are of two kinds:
(a) Innovation within the current level of technology
(b) Innovation to a new level of technology
Generally, when we say 10X improvement has occurred, it does not mean an actual multiple of 10; it could have been 4-5X as well—the point is that users need to have their lifestyle dramatically improved regardless of what that multiple was. Generally aiming for an order of magnitude is better, since you are being considerate of the margin of error. This huge improvement generally occurs in an existing field of technology. This is what innovation within the current level means. For being able to innovate within the current level, you need to answer the question: How can I improve the solution?
Elon did not make the first lithium-ion batteries, but he fine-tuned the crucial aspects of it such that cost and performance to run EV are not hindered but are rather exemplified to users in multiple ways. Now, he sets an industry example for other companies to follow, enjoying the benefits of economies of scale with 10 times cheaper batteries, which is the most important part of the whole car.
When I made a photoelectrochemical filtration mechanism to destroy volatile organic compounds from marijuana and convert it to harmless gas instead of capturing those via carbon filters only to be thrown away later, I essentially innovated to a new level of technology. You see, for being able to innovate to a new level of technology, you need to answer the question: How can I replace the solution?
There are several ways to solve the problem, but it needs to be categorized. For instance, current technologies depended upon capturing techniques (carbon air filters, HEPA filters, etc.) which are to be judged by A, B and C metrics but my technology solved the same problem but with a destroying technique (PECO technology) which not only scores significantly high in the A, B and C metrics but opens the door to new type of technology…a technology that solves the same problem in a different, better way. I have replaced the current solution with a new one that is much more effective in solving the root problem.
Note that in both the cases for Elon’s lithium-ion batteries and my PECO technology-based filtration system, innovation has happened, which means that user’s life has been greatly affected for the good. But the type of innovation that happened were different since one was within the existing technology level through 10X improvement and the other was a shift to a new level of technology. Both these are different, but both influence the life of users positively in great measures, meaning that these were cases of innovation.
Only when you are making continuous improvements on saturated designs is when it shall be more appropriate to give the advice on not reinventing the wheel (e.g., if Elon’s team now suggests adding a material X so that energy density of battery improves by 4%) although as mentioned in previous article, the maintenance team dealing with manufacturing machines may be confined to only making continuous improvements so that machine sustains production over time. Continuous improvements are also done by asking the same question as 10X improvements: How can I improve the solution? But the answers to them often yield little to no life changing benefit to end users, hence they are pursued where required and not when deemed such. Again, continuous improvements are fine and necessary unless the design, process or technology has reached high saturation. If Elon’s team made the same suggestion way back in 2010, it may well have been considered, but it now seems that the 4% on top of what they have made a decade later adds very little value—perhaps it is getting closer to saturation?
Taking all these discussion in our mind, let’s boil it down to 3 refined conclusions:
1. Continuous improvement on current technology is fine and necessary unless the design has reached high levels of saturation. From then on, if you make improvements, it is as if you are re-inventing the wheel.
2. Innovation within the current level of technology is when 10X improvements are made on the existing design that changes lifestyle of users for the better. To do this, ask: How can I improve the solution?
3. Innovation to new level of technology is when a new design solves the problem in a different, better way for users. To do this, ask: How can I replace the solution?
Remember, first principles thinking allows for innovations to occur, but it does not mean that while doing the small continuous improvements, you won’t apply the same mindset. In fact, when faced with a design issue you do not know from the get-go if you will be mildly improving some specifics in current design or making huge improvements in that sphere or if you will be replacing it (of course it matters on your scope, role, time, and situation as mentioned in the previous article). Say you may start pursuing continuous improvement, soon you may find yourself steering towards a 10X improvement (innovation within current level) and soon you may arrive at a point where you realize the need of creating a design that solves the problem through a different framework much better (innovation to new level). All the stages always contain thinking from first principles. Never stop this way of thinking!
We will dive into the next article which is filled with graphs and plots that explain the concepts of continuous improvement, 10X improvement, and shift to new technology with respect to the level and saturation of the technologies. We will also link concepts to who the relative stakeholders/users are, how many of them are on each level and how that impacts our design decisions. Finally, we will also see a time dimension to the different technology levels and understand the pace at which technology is running and what you need to do to catch up.
