Did you know Japan's government was still requiring floppy disks for official paperwork as recently as 2022? Fax machines are still everywhere in Japanese offices today.

One of the coolest, most advanced countries in the world is running its software like it's the 90s. And I think we do the exact same thing.

Japan builds the most reliable trains on Earth. The Tokaido Shinkansen runs about 130,000 trains a year between Tokyo and Osaka. The average delay across all of them is roughly 24 seconds. In 2017, a train left a station 25 seconds early and the railway company issued a public apology. Since the line opened in 1964, it has carried around 7 billion passengers with zero passenger fatalities from accidents on board.

That's not an accident. That's a culture.

Japan's obsession with precision, craftsmanship, and getting things exactly right is real, and it produces some of the best-engineered physical objects in the world. Trains, cameras, cars, tools. Things built to work exactly as intended for decades.

But the same culture that produces 24-second train delays also produces software that looks like it's stuck in 1999.

Japan's Ministry of Economy, Trade and Industry has actually put a number on it. Their 2018 "DX Report" warned that if companies didn't modernize their legacy IT systems, Japan could face economic losses of up to 12 trillion yen a year starting in 2025. That's roughly 2% of GDP, every year, from old software alone.

The same report found that about 20% of Japan's mission-critical business systems were already over 20 years old, and projected that number would hit 60% by 2025. On top of that, Japan expected a shortfall of 430,000 IT specialists by 2025, up from a shortfall of 170,000 in 2015, largely because the engineers who understand these old systems are retiring and nobody young wants to inherit them.

METI gave this scenario a name: the "2025 Digital Cliff."

Here's why the same mindset that builds perfect trains struggles to build good software. A train is a finished object. You design it, test it, manufacture it, and move on to the next version. Software isn't like that. You ship it, real people use it in ways you never predicted, bugs surface, and you patch it. Then you patch the patch. Software is never actually done, and a culture built around finishing things perfectly doesn't have a natural home for something that's designed to stay unfinished forever.

Japan also leaned hard into outsourcing its software work, layering subcontractor on top of subcontractor. Over time, that created a strange side effect: companies became dependent on systems that the company's own employees no longer understood. When a system is old, tangled, and nobody wants to be the one who breaks it, you don't replace it. You patch it and hope.

Japan can build good software. Companies like Toyota and Sony clearly can. The interesting part is that a culture engineered around getting things exactly right can make it genuinely harder to start something messy, experiment in public, and be willing to break things on the way to better.

I’ve done the exact same thing with a notebook.

I buy the notebook and imagine the perfect system: every morning at 6:00, three pages, date at the top, gratitude section, goals, reflection. Then I miss a Tuesday. The system suddenly feels broken, so I decide to start fresh next Monday. Next Monday quietly becomes never.

The problem was never discipline. The problem is that somewhere along the way, journaling stopped being a habit and became a product that had to launch perfectly or not at all.

There's an older, simpler way to keep a notebook, and it predates the daily-journal industry by a few hundred years. It's called a commonplace book.

The idea is almost embarrassingly simple: collect things you find interesting. A quote from a book. A question you can't answer. A sketch. A business idea. Something you overheard on the train. A paragraph worth remembering. A weird fact. Something you disagree with. No daily requirement, no required structure, no streak to protect.

Leonardo da Vinci's notebooks are the most famous surviving example, and they're a mess in the best sense. Across roughly 13,000 surviving pages, his notes jump from anatomy to hydraulics to flight to grocery lists to half-finished sketches of weapons, sometimes on the same page.

That randomness wasn't sloppiness. It was the point. Ideas from completely different parts of his life sat next to each other on the same page, which is exactly how unrelated things start connecting in ways they never would if everything were filed neatly into its own separate notebook.

Your commonplace book doesn't need to look good. It doesn't need a system. It doesn't need daily entries. It just needs to be somewhere to put the things that make you stop and think for a second.

Don't have a perfection problem, just start if you want to.

How many interesting things have you already forgotten?

A commonplace book is a place to collect anything worth remembering:

The challenge: Grab a notebook and fill one page this week.

a mix between a journal, notes app, scrapbook, and idea bank—but instead of recording what you did today, you're recording what you discovered.

What's something in your life you haven't started because you're waiting for the perfect system?