@Kevin Liu January 7, 2026

<aside> 📌

I was never a fan of the idea of being a “blogger”, or paying for a Medium account just to have a valid opinion on something. Writing has been a fundamental aspect of being human since we’ve been scratching rocks on cave walls, just like how children have a natural inclination to draw random shit on the wall with crayons, so a blog seems like a silly (and possibly expensive) prerequisite for my ideas to simply exist for other people to read.

By no means am I a professional writer, but I do enjoy it, and I’ve been paid to string words together as a content writer in the past, and work as a content design intern at ServiceNow, so maybe I’m at least okay.

I think of writing as nothing more than a canvas I use to pin my ideas down so I can actually see them. I think I have a pretty chaotic mind, so drawing doodling and writing really helps me empty it out. Usually, I get my thoughts out on a yellow legal pad, but I’m often too impatient to write nicely, I cross things out, write in the margins, and outside the lines I’m given, so reading them later gets difficult. I’ve also found that some ideas need to typed out to feel "real". When did the digital world start becoming the real world?

Anyways, instead of a “blog”, what follows is a raw draft. It’s an evolving space where my original notes meet some help from AI. I do think that AI-generated writing can easily be detected as a “derivative” of real human writing, which has its downsides, but for all intents and purposes of this draft, the AI-ness is secondary to the exploration of the ideas itself. Read away, or not.

</aside>

Why I am spending the next twenty weeks building a digital brain that outlasts the tools we use to hold it.

We have reached a breaking point in our relationship with technology. We own more devices than ever, we subscribe to more apps than ever, and yet, our digital lives have never felt more fragmented. We are living in an era where "finding" a piece of information is often harder than creating it in the first place.

My project stems from a very personal friction: the mental weight of digital amnesia. I know I saved that reference image. I know I wrote that project goal down. But was it in a Notion block? A Slack thread? A random Figma scratchpad? I spend my days acting as a manual bridge between my own apps, copying and pasting data from one silo to another just to keep my head above water.

Beyond the mental load, there is the "Subscription Tax." I pay for Google One, Dropbox, and iCloud. I am paying three different companies to hold three different versions of my life, yet none of them talk to each other. I am paying for storage that I don't truly control.

My project is about fixing this. It is about making it so your notes, files, and ideas live with you on your own physical "key" instead of on a company’s server. To do this, I am researching a shift from a world where apps own data to a world where data is its own master.

Why the "File" metaphor is failing

For forty years, computers have found information using Location-Addressing. When you look for a file, you are telling the computer where it is: Documents/Client_Work/Project_v2.docx.

This worked when we had one computer and ten files. But today, "location" is a lie. That file is on your laptop, mirrored in a cloud, cached in a browser, and shared in a chat. When you change the location (move the folder) or change the name, the link breaks. The computer "forgets."

The Solution (?)

To build a system that is unbreakable, we have to move toward Content-Addressing. Instead of identifying a file by where it is, we identify it by what it is. This is where the math comes in. In my system, every piece of data—a sentence, an image, a PDF—is passed through a mathematical function called a Cryptographic Hash (specifically $SHA-256$). This function takes the content and spits out a unique, fixed-length string of characters: a "Fingerprint."