Good Products Are Hard to Vary

Summary

Naval Ravikant explores David Deutsch’s principle that “good explanations are hard to vary” and applies it broadly to product design, engineering, and the nature of learning. He argues that when a product or design reaches its optimal form, it becomes nearly impossible to change without degrading it — the iPhone’s form factor, the airplane wing, and the streamlined car are all examples of convergence toward a “Platonic ideal” where each component constrains the others so tightly that variation would only make things worse.

The conversation then shifts to the fractal nature of knowledge and high-density writing. Naval explains that good writers produce work so interconnected that readers will absorb different layers of meaning depending on their level of readiness. Learning is an iterative process — you read something, grasp 20%, return later and get 25%, supplement with podcasts and AI tools, and gradually deepen your understanding. All knowledge is ultimately a communication between author and reader, and both must be at a certain level for the transfer to occur.

Highlights

”Good Products Are Hard to Vary”

“Good products are hard to vary. Go look at the iPhone—the smooth, perfect, beautiful jewel. The form factor hasn’t really changed that much since the original one. It’s all around this single screen, the multi-touch, embedding the battery, making it fit into your pocket, making it smooth and sliding in your hand, essentially creating the Platonic ideal of the truly personal pocketable computer.” — Naval Ravikant, 0:33

yt-dlp --download-sections "*0:33-1:35" "https://www.youtube.com/watch?v=7PN9nlKNRFI" --force-keyframes-at-cuts --merge-output-format mp4 -o "good-products-hard-to-vary.mp4"

”The Airplane Wing Is Perfect”

“There’s the famous saying, I think from Antoine de Saint-Exupery, where he says the airplane wing is perfect not because there’s nothing left to add, but because there’s nothing left to take away. That airplane wing is hard to vary.” — Naval Ravikant, 1:35

yt-dlp --download-sections "*1:35-2:42" "https://www.youtube.com/watch?v=7PN9nlKNRFI" --force-keyframes-at-cuts --merge-output-format mp4 -o "airplane-wing-hard-to-vary.mp4"

”Learning Is Fractal”

“Good writers write with such high density and interconnectedness that their works are fractal in nature. You will meet the knowledge at the level at which you are ready to receive it.” — Naval Ravikant, 2:42

yt-dlp --download-sections "*2:42-3:17" "https://www.youtube.com/watch?v=7PN9nlKNRFI" --force-keyframes-at-cuts --merge-output-format mp4 -o "learning-is-fractal.mp4"

Key Points

• David Deutsch’s interconnected philosophy (0:00) - Reading Deutsch across disciplines is useful because his ideas on memes, epistemology, wealth, and knowledge all interconnect and reinforce each other
• Good explanations are hard to vary (0:33) - When you look back on a good explanation, all the parts fit together and constrain each other so tightly that it could not have been otherwise
• iPhone as Platonic ideal (0:33) - The iPhone form factor has barely changed across 16 generations because Apple designed the right thing — the Platonic ideal of a personal pocketable computer
• Antoine de Saint-Exupery on perfection (1:35) - Perfection is achieved not when there is nothing left to add, but when there is nothing left to take away
• Wind tunnel convergence (1:35) - Modern cars all look similar because the most efficient design emerges from wind tunnel optimization and is hard to vary without losing efficiency
• Electric car as next hard-to-vary design (1:35) - The internal combustion engine was hard to vary until batteries improved, and now the electric car design is similarly hard to vary
• Fractal nature of learning (2:42) - High-density writing is fractal: you absorb 20%, then 25%, then supplement with podcasts and AI to gradually deepen understanding
• Knowledge as communication (3:08) - All knowledge is a communication between author and reader, and both must be at a certain level to absorb it

Mentions

Companies

• Apple (0:33) - iPhone cited as the prime example of a product that is hard to vary

Products & Technologies

• iPhone (0:33) - The Platonic ideal of the personal pocketable computer, unchanged across 16 generations
• Electric cars (1:35) - Next hard-to-vary design after the internal combustion engine
• Grok / ChatGPT (3:00) - AI tools mentioned as aids in the iterative learning process

People

• David Deutsch (0:00) - Philosopher and physicist whose “hard to vary” principle anchors the entire discussion
• Antoine de Saint-Exupery (1:35) - Quoted on the perfection of the airplane wing
• Brett Hall (2:42) - Podcaster who covers David Deutsch’s work

Surprising Quotes

“Both Apple and its competitors have tried to vary it across 16 generations of iPhone and they haven’t been able to materially vary it.” — Naval Ravikant, 0:33

“There’s a complaint now among some designers that in modern society products and objects are starting to look all the same. Is that because of Instagram? Why is that? Well, at least in the car case, they all look like they’ve been through a wind tunnel design because that is the most efficient design.” — Naval Ravikant, 1:35

“All knowledge is a communication between the author and the observer or the reader. And you both have to be at a certain level to absorb.” — Naval Ravikant, 3:08

Transcript

Naval Ravikant: 0:00 I think reading Deutsch across all the different disciplines is very useful. Even when he talks about memes and meme theory, that comes from evolution but crosses over straight into epistemology and conjecture and criticism. And it reaches far beyond his definition of wealth: the set of physical transformations that you can effect. That takes into account both capital and knowledge, and it clearly shows that knowledge is the bigger component. And then that can be brought into business and applied into your everyday life. It can apply to the wealth of nations and it can apply to the wealth of individuals. So there are a lot of parts that interconnect together.

Naval Ravikant: 0:33 He says that good explanations are hard to vary. So when you look back on a good explanation you say, well, how could it have been otherwise? This is the only way this thing could have worked. All these different parts fit together and constrain each other in such a way that there’s now some emergent property or some complexity or some outcome that you didn’t expect, some explanation that neatly explains everything. That doesn’t just apply to good explanations, it applies to product development. Good products are hard to vary. Go look at the iPhone—the smooth, perfect, beautiful jewel. The form factor hasn’t really changed that much since the original one. It’s all around this single screen, the multi-touch, embedding the battery, making it fit into your pocket, making it smooth and sliding in your hand, essentially creating the Platonic ideal of the truly personal pocketable computer. So that product is hard to vary. Both Apple and its competitors have tried to vary it across 16 generations of iPhone and they haven’t been able to materially vary it. They’ve been able to improve the components and improve some of the underlying capabilities, but materially that form factor is hard to vary. They designed the right thing.

Naval Ravikant: 1:35 There’s the famous saying, I think from Antoine de Saint-Exupery, where he says the airplane wing is perfect not because there’s nothing left to add, but because there’s nothing left to take away. That airplane wing is hard to vary. When we figure out the proper design of the spacecraft to get to Mars, I will bet you that both at a high level and in the details for quite a long time that thing will be hard to vary until there’s some breakthrough technology. The basic internal combustion engine design was hard to vary until we got batteries good enough and then we created the electric car, and now the electric car is hard to vary. In fact, there’s a complaint now among some designers that in modern society products and objects are starting to look all the same. Is that because of Instagram? Why is that? Well, at least in the car case, they all look like they’ve been through a wind tunnel design because that is the most efficient design. The reason they all look swoopy and streamlined is because they’re all going through a wind tunnel and they’re trying to find the thing that cuts through the air with minimal resistance, and so they do all end up looking the same because that design is hard to vary without losing efficiency.

Naval Ravikant: 2:42 Good writers write with such high density and interconnectedness that their works are fractal in nature. You will meet the knowledge at the level at which you are ready to receive it. And you don’t have to understand it all. This is the nature of learning: you read it, you got 20% of it, then you go back through it, you got 25% of it, you listen to one of Brett Hall’s podcasts alongside…

Naval Ravikant: 3:00 Now you got 28% of it. Now you go to Grok or ChatGPT, you ask it some questions, you dig in on some part, now you got 31% of it.

Naval Ravikant: 3:08 All knowledge is a communication between the author and the observer or the reader. And you both have to be at a certain level to absorb.

Naval Ravikant: 3:17 When you’re ready to receive different pieces, you will receive different pieces. But you will always get something out of it, no matter what level you’re at, as long as you can even just communicate and read the language.