What It Is

Ideas You Borrowed

What if most of your ideas come from copying others without you noticing?

Much everyday reasoning works by analogy. We do something because others do it, because it worked before, or because it is the accepted way.

In engineering, that can sound like "everyone uses this" or "our last project did it this way."

Analogy is fast and often useful. But it carries hidden assumptions along with it.

Thinking From First Principles

First principles thinking takes a different route.

You break a problem down into its most basic parts: the facts you can check and confirm are true.

Then you question every assumption that rests on top of those facts. Many turn out to be conventions or borrowed beliefs.

Finally, you rebuild a solution from those basic facts, rather than starting from existing solutions.

Analogy vs First Principles

Both approaches solve problems. They start in different places.

Reasoning by analogy starts from existing solutions. You copy what others do because it seems to work for them.

First principles thinking starts from basic facts. You build up a solution that fits your actual problem.

Analogy is quicker. First principles takes more effort, but it exposes the assumptions analogy hides.

Imagine Priya's Slow API

Imagine a software engineer named Priya. Her API endpoint takes a hypothetical 800 ms to respond.

Her team wants to add a Redis caching layer. The reason? "That is what everyone does."

This is reasoning by analogy. The fix is copied from others instead of derived from the problem.

And it has costs: new infrastructure to run, plus the risk of serving stale data.

Measure What Is Actually True

First, Priya sets aside the team's assumption and asks what is actually true.

She profiles the request. It shows about 700 ms in one database query, 50 ms in network time, and 50 ms in serialization.

Now the picture is clear. Almost all the delay sits in a single query.

Measuring where time is really spent comes before picking a tool.

Build Up From the Facts

Next, Priya looks at that slow query. It scans a table with no index.

That is a basic fact she can check. Now she builds up from it.

Adding one index cuts the query to about 20 ms. The total drops to around 120 ms.

No cache. No new infrastructure. The fix came from the problem itself instead of from what others do.

Questions to Carry With You

Use three key questions whenever you face a problem.

What do I know for certain? Which of my beliefs are just conventions or borrowed assumptions? What would I build if I started only from what is true?

For engineers, this means measuring before picking a tool, and asking what a system truly needs before copying a popular architecture.

It also means learning a new technology by understanding the core problem it solves, instead of memorizing its features.

A team adopts microservices only because large tech companies use them. This is an example of reasoning by _____.

• analogy
• first principles

Answer: analogy

Choosing something because it seems to work for others is reasoning by analogy. First principles would start by asking what this system truly needs to do.

First principles thinking _____ a problem into facts you can confirm are true. It then _____ a solution from those facts. Reasoning by _____ instead copies what works for others.

Answer: First principles thinking breaks down a problem into facts you can confirm are true. It then builds up a solution from those facts. Reasoning by analogy instead copies what works for others.

First principles has two moves: break the problem down to verified facts, then build a solution up from them. Its opposite, reasoning by analogy, starts from existing solutions others use.

Reasoning by analogy is fast and often useful. What is its main risk?

• It carries hidden assumptions that may not fit your situation
• It is always slower than reasoning from basic facts
• It never produces a solution that works in practice
• It requires measuring every part of a system first

Answer: It carries hidden assumptions that may not fit your situation

Analogy brings along hidden assumptions from the original situation. If those assumptions don't hold for your problem, the copied solution may be wrong or wasteful. Analogy isn't always wrong or slow; it just goes unchecked.