Daily Balajisms - The Idea Maze
A grit to face computational irreducibility
A good founder has a bird’s eye view of the idea maze. The history of an industry, the competitors, all the pitfalls and emerging technologies.
“Anyone can point out the entrance to the maze, but few can think through all the branches.”
Ideas come and go, but what counts are the years someone explored a specific idea in detail and practice. And all attempts implementing that idea using available technology. And all the dead ends people encountered, while going through the idea maze.
Sometimes a new door opens in the idea maze, as technologies mature and become available. Balaji mentions an example of a specific tech stack (LAMP stack) that enabled the web2 companies, like social networks. New technology stacks will scale web3 companies.
Marc Andreesen explains in a podcast that at a16z they use the idea maze concept of Balaji Srinivasan to assess early-stage founders. Often, they find that even a 20-year-old kid has been exploring a certain business idea and actively tinkering around it, since they were 13 . Palmer Luckey, a founder of Oculus (and Anduril) was very young, when he started to work on his VR kit, which he later sold to Facebook.
A related concept is a tech tree, that according to Balaji also resolves the tension between “technological determinist” and “a great man” theories of history. I think Wardley mapping could be an interesting tool to build such tech trees, as it provides a two-dimensional view of innovation supply chain and commoditization stage of individual parts in that innovation supply chain. A generalized concept is a vector-theory of change, mapped through micronarratives, and introduced by Dave Snowden – “more positive stories like these, and less negative stories like those”.
Prime number maze is a related concept at the core of Balaji Srinivasan’s belief system. He said on a podcast with Lex Friedman, that even longevity is not the ultimate goal for him. He wants to live long just to try solving the prime number maze – a metaphor for a system that has a simple rule that might escape our attention or current capacity to comprehend it. Balaji mentions how rats cannot be trained to solve prime number maze. It’s just too abstract for them. Computational irreducibility by Stephen Wolfram is a similar concept.
Founders need a grand strategy, fast OODA loops and lots of grit to survive many winters and dead ends in the idea maze.