savvy normie

How Nuking My PKM Paralyzed My Productivity for 6 Months

A bit of backstory

Ok, I gotta come clean. This is a clickbait title. It is misleading if understood literally. But it is actually true if interpreted right. Let me explain.

I stopped trying to write anything for this blog in the fall of 2025 after posting just two posts. There were some objective reasons for that. Reasons that made me almost incapable of doing anything at least marginally challenging. So, most of my "projects" fueled by enthusiasm just stalled as I have fallen to the safety net of doing only what is strictly required (but what I low-key hate, but that's a different story).

My third post for the blog that I was working on then was about how I finally "nuked my PKM". By the time I dropped it, I had already spent numerous hours trying to come up with a perfect "complete" story of how I ended up nuking my PKM, while simultaneously trying to weave a social theoretical analysis of the phenomenon into the narrative. The draft at this point was already 4359 words.

I kid you not. 24555 characters. This is like half of an academic article. A post. For a blog. That was supposed to be a side gig. And even 'side gig' would have been a gross overstatement. More like a chance to vent publicly. The "Express" part of Tiago Forte's CODE. (I'm actually almost kidding because what follows goes against Second Brains, in a way. But if anything, the E in that acronym is the most important part, whereas so many of us only pay attention to the C, it seems).

So I stalled. And then I abandoned the whole thing completely. Because if something can't be done in a very elaborate perfect fashion, it shouldn't be done at all. Right?

So, this is what the title of this post actually means. I did nuke my PKM and I did go into an unproductive spiral (a "hiatus", if you will). But there's no causal link between the two, of course. And there's a story in between.

This is how one draft of that post started: "If complexity were a Disneyland, a PKM vault would be the main attraction. A few weeks ago, I nuked mine—about 1,500 notes. It looked clever. It felt heavy. And ... it kept me stuck." Care to elaborate? Yes. But not in 45 hundred words.

A few disclaimers, though

First, I definitely do not recommend that everyone just follow suit and rm rf their PKMs or Second Brains (or whatever the cool kids call them these days).

Second, I do think that most of us need some form of note-taking. What most of us arguably don't need is collecting crap promiscuously and linking the shit out it then. Just because our brain can be hypothesized to store knowledge that way.

Third, it might just be that I "used" my PKM "wrong".

Finally, and most importantly, it could be that I outgrew the contents of my "vault" because of other pivots in life, which is a very specific reason, indeed.

Or maybe it’s AI that made knowledge hoarding management less relevant … Just saying.

A post that almost killed me (a very succinct version)

Now here's the shortest possible summary of my story and a bit social theorizing of PKMs sprinkled on top.

I didn’t start with ā€œa note taking system.ā€ I started with paper and a few Word files. Because saving information for future reference was a necessity (it really was -- just trust me on this for now). But it wasn't note taking. And there definitely was no management involved. An no cult.

Then the "ritual" arrived enabled by the software that made the social practice possible and the YouTube bros that theorized it like any other practice associated with late modernity. Evernote, OneNote, Notion, Roam, Obsidian. The whole parade of ā€œthis time you’ll think better.ā€ Each hop felt like progress. Covers and emojis. Bi-directional links (This was huge!). Graphs that sparkle. (And can be used to flex but not much else.) Each time the story was the same. A promised salvation through new "features". Euphoria. Disillusionment later. Each new piece of software was a "game changer". Until it wasn't. Many know the story a bit to well.

Eventually, I had to face the truth. It wasn't really the software. Although, I am going to admit that some are indeed better suited for taking usable notes than other. It was me. And what note taking was supposed to be for me and how it really should work for me (and, arguably, many others). Let's start with only two things that I realized before this post spirals out of control like its predecessor.

No more "just in case"

Indiscriminate collection of information into a Second Brain hardly makes any sense. I mean it. Especially in 2026. Admittedly, I never was the type who just 'clipped' all sorts of random stuff from the internet thinking I might use it some day. But I definitely was the type who stored numerous research notes hoping to use them in an article or book one day and/or in an attempt to learn as much as possible about the stuff that may be at least marginally related to a possible research project. At least. Marginally. Possible. One day.

A careful reader will notice a trap behind the ideas in the previous few sentences: there's so much that can qualify, that collecting may never end. Which it never did for me, of course. Unsurprisingly really. I just kept reading and researching and collecting (just to make sure I know everything about everything) while producing (subjectively) infrequent and (also subjectively) quite mediocre academic output at best. And no other output outside academic at all.

But here's the worst paradox. When it actually comes to a specific project like writing an article or putting together a lecture on a subject, it turns out that what you already have and/or know is never enough or not detailed enough. And you are still forced to run back to the original sources to actually understand the thing you’d ā€œcapturedā€ in your PKM. Shocking, I know. But it worked like that every time in my case.

So, here's a rule I am trying to enforce from now on: no knowledge collection unless it's for a specific ongoing or realistically upcoming project, not just a fantasy project someday maybe. Or it's a reference that I actually use at least semi-regularly (which I actually have quite a few). No more "just in case" for me. Only "just in time". I don't collect. I save what I actually use. And prepare to ship.

My knowledge graph is not my freaking brain

The second pillar of PKMs is linking. As soon as there was software that allowed linking (and YouTube bros who made it into a cult), I became a serial linker. I linked as though my life depended on it. Everything to everything else. Because this replicates the organization of knowledge in our brain, we were being told. And this is how you discover new connections and produce new knowledge. Right?

So let’s link a clipped note on milk and one on the color white and one on cows. Because these ideas are related in some way. And this is more than enough for a productivity nerd who's serious about their knowledge vault. (And because we now could. Thank you Roam.)

And if a certain idea Y comes to mind when working on an idea X, don’t forget to create and empty placeholder page for idea Y linked from the page of idea X. You will flesh this idea out later. When you get round to it. Because idea Y is totally worth exploring. And you will totally get round to fleshing it out. When? When you're doing Progressive Summarization of your existing notes, I guess.

This is sarcasm. And exaggeration. Obviously. But in all seriousness, little did I realize back then on the peak of my Roam-->Obsidian (and some Notion in between) euphoria that linking anything in a software does not equal to my brain being able to hold and understand all these relations at once. Let alone being able to aptly "juggle" them all in it’s mental RAM, so I can reproduce the whole intricately linked entity or any of its parts in any of the possible ways in speaking or writing. I know I have put this painfully clumsy, but I have already given myself permission to suck, so suck I will.

I came to believe that linking notes, mindlessly "clipped" or meticulously "crafted" (did anyone else notice how LLMs love this word?), empty, fleshed out, or "progressively summarized" in a software is not going to help much with understanding and learning. At least it did not help me in a perceivable way. YM, of course, MV. And recreating a local Wikipedia looks pointless if not ludicrous in this day and age. (You can still impress Redditors and Obsidian bros with you knowledge graph!)

Coda

So, here I was with a PKM that was not only (mostly) useless but was actually getting in the way. And after deliberating this and replaying this scenario in my head (admittedly for way longer than I should have), I moved about 30 notes (all research related) out of the old vault and deleted the rest. No archives, no cold storage. Just the good old rm -rf. And you know what? It felt good.

Originally posted: 2026-05-10