Thank you for your well-argued letter! Though it may be hard, I’m going to attempt to convince you that Definite vs. Indefinite Optimism is a useful dichotomy and that continued progress depends on more optimistic determinism.
Quick aside: After going back through Startups - I realized that Thiel (or at least Masters) uses ‘definite’ and ‘deterministic’ interchangeably and I am going to do the same. This interchangeable use has gas-lit me into believing I just couldn’t get the term straight for years.
I agree that Many good things can come from indefinite optimism. People have embraced indefinite optimism because it does produce results. That’s why it is insidious. Pasteur is a great example. So is Alexander Fleming. Indefinite worldviews and definite worldviews enable exploration of different parts of idea space - there are things that someone with an indefinite worldview can discover that would be inaccessible to someone with a definite worldview because Indefinite worldviews leave you open to serendipity in ways that definite ones do not.
Here we begin to diverge. Do we actually lack big visions? Google’s vision is “to provide access to the world’s information in one click” - that’s pretty big. Many people we know have the vision to cure aging, eliminate poverty, or change how we do science. However, Definiteness doesn’t have anything to do with vision.
Shamans and psychotics have visions. Visions say “we should vaguely go towards those mountains.” Visions not falsifiable. I’ve never heard of a situation where someone has failed to get to their vision, sometimes even in death. You also can’t argue with a vision beyond “I don’t like that vision.”
Startups does not mention the word vision once in the whole discussion of Determinism and Optimism.
A definite plan looks like “I have a hypotheses for how I’m going to get from here to there via these steps.” It’s also key that the hypotheses is falsifiable. Falsifiable hypotheses are important and Good plans have several steps.
Plans are worthless in an indefinite world, leading to the attitude that Everything can incrementally improve from small things with tight feedback loops. To say that everybody incrementally improving doesn’t have a plan is both disingenuous and false. The difference is the timescale. Indefinite approaches plan on the scale of quarters and features. Definite plans exist on the timescale of multiple years or decades. Tesla legitimately did put its secret master plan online in 2006 and have been consistently executing against it for 14 years now.
Instead of being an example of the success of an incremental mindset, Apollo is an example of the value of those long term plans, especially in contrast to the Russian Space Program at the same time.
Beginning with Kennedy’s speech in 1961, NASA started executing on James Webb and Von Braun’s plan to go to the moon. Good plans have several steps so they first tested what space even did to humans in the Mercury program and then how multiple people could survive in space for many days in the Gemini program, all the while learning to build increasingly bigger rockets. But they always knew they were going to the moon and why they were taking each intermediate step. Note also that Kennedy said “we will go to the moon within the decade.” This is a falsifiable hypotheses. The US could have failed at it and almost did.
By contrast, the Russian space program did take an indefinite, incremental approach and it ultimately cost them. At the beginning of the space race the USSR consistently beat the US to one milestone after another (first satellite, first living thing in orbit, first human in orbit, first space walk, first woman in space, etc.) but ultimately failed to get a human to the moon. Why? Khrushchev only ever allowed Sergei Korolev (the head of the Russian space program) to run programs directly focused on the next milestone.^1 This highlights a key reason why we need more optimistic definite thinking - It is more expensive in the short term to do things with long term payoffs. In an indefinite worldview, that up-front cost is worthless.
SpaceX and Blue Origin illustrate this point. Long before they talked publicly about landing a rocket, SpaceX designed its Raptor engines so they could reignite after turning off. This feature both makes them more expensive and is completely worthless … unless you want to use them to land.
Blue Origin’s rockets are all designed to run on hydrogen. As a fuel hydrogen is annoying - it has to be kept very cold and has a nasty habit of leaking through solids. Every other rocket builder has chosen to use hydrocarbons like hydrazine or hydrogen peroxide. Avoiding hydrogen is a great idea if you just want to take the next step of getting to space. However, of all these fuels, H2 is the only one that can easily be manufactured from materials found on the moon. Because Blue Origin’s goal isn’t just aiming to get to space, or even just to the moon. It has a roadmap to a world where the Moon is a stopping off point to the Asteroid belt and then the rest of the Galaxy.
You assert that Definite vs Indefinite is a false dichotomy because there will always be unknowns. A definite worldview allows for unknowns. A definite worldview means you believe the universe is deterministic. It does not mean that you believe that you’re omnipotent, which is the only way there could be no unknowns.
The difference between definite optimism and indefinite optimism lies in how you respond to the unknowns. A definite optimist creates long term plans despite the fact that those plans could be wrong. An indefinite optimist responds to the existence of unknowns by not creating a long term plan at all.
Definiteness means that you have a long-term plan but that doesn’t mean that plan doesn’t change. It does mean that when that plan changes, it maintains a long timescale.
This stands in contrast to an indefinite worldview where there is no long term plan in the first place.
In hindsight, both the indefinite and definite approach end up looking like a walk through The Idea Maze . That is why the contrast between definite and indefinite worldviews has to do with their approach to the future.
The hindsight problem makes it hard to know whether PayPal pivots were due to a definite worldview that ran into unknowns or a secretly indefinite worldview. Peter did assert that one of the reasons they sold PayPal was because they ran out of good ideas and would have gone into indefinite optimistic mode.
I’m going to assert that most people trying to innovate in march 2020 have an indefinite optimistic worldview. If you disagree, the argument for that will need to wait for another letter.
The dominance of this indefinite worldview leads to aversion to the upfront investments associated with long-term definite plans, cutting off large swaths of idea space.
I hope I’ve convinced you both that there is a true dichotomy between definite and indefinite optimism and that for the sake of progress, we need to shift the worldview of innovators towards one of definite optimism.
^1: Check out the Moonrise podcast for an amazing narration of this story