The Semi-Finished Product Filed Under “Knowledge”
In most people’s mental model, notes only have three tiers: see information → turn it into knowledge → use it to create. The act of “recording” in between gets quietly filed under “knowledge.”
But as I said in the overview, records deserve to stand alone as their own layer. Because it’s an independent intermediate form: it’s relevant to you, but not necessarily useful forever; it might just be something you’ll need someday, or something you’re using right now to clarify your own thinking. That kind of thing doesn’t yet qualify as knowledge — only what’s structured for repeated future reuse counts as knowledge.
A record is an index — it’s semi-finished knowledge.
Recognizing this cures a very common ailment: mistaking “recorded a lot” for “learned a lot.” Recording doesn’t equal mastery, just as prepping ingredients doesn’t equal cooking a finished dish. But note — calling records semi-finished is not, in any way, diminishing them. Quite the opposite: records are one of the highest-conversion methods from information to knowledge. They’re the first site where you personally process information with your own hands, the stage in this entire pipeline that should least be skipped — and yet the one most often is.
This essay is about how to make good use of that “first site.”
The Act of Writing Is Itself a Retrospective
Let’s start with the most underrated action: writing something every day.
Many people misunderstand “writing” as something you can only do once you’ve figured things out. Actually, it’s the reverse: the act of writing is itself spontaneous reflection, itself a retrospective. You don’t write because you’ve figured it out — you figure it out by writing. Words are the developer fluid for thought: the ideas in your head that you assume are clear reveal their vagueness, their gaps, their shaky logic the instant they have to become a complete sentence.
I’ve seen too many examples of this in the community. Someone who spent over a decade in sales and was restarting mid-career was asked to publish a long essay every week. At first, they had no idea what to write or how. Forcing themselves to record and reflect daily, three months later, their own words were: give me a topic and I can now easily produce a thousand or two thousand words, and sit quietly at my computer writing for two hours, finding that state of flow. They summed up the change in one plain but heavy sentence: reflection + pain = progress.
Writing is the most efficient form of recording precisely because it brings “pain” — that friction of forcing vague ideas into clear text. As I said in the previous essay, the most dangerous use of AI is letting it eliminate friction for you. And recording is the place where you actively manufacture friction. The discomfort here isn’t something to avoid — it is growth itself.
There’s a more practical reason too: only what’s written down can possibly be structured, can possibly enter the knowledge layer, can possibly one day be recombined into creation. A thought that isn’t written down evaporates by the next day. Recording is the only way to pin fleeting thought onto time.
Records Need to Be Low-Friction: Complete First, Perfect Later
Since recording matters this much, it needs to happen as easily as possible. The first principle of recording is low friction.
The biggest friction comes from perfectionism — always waiting until you’ve “thought it through,” “have a solid block of time,” “feel ready” before writing, and the result is you never write. The fix is simple: lower the bar to the floor. Someone in the community wrote a line in their retrospective that I love, roughly: to break my own perfectionism, I first write down whatever I can still remember right now. Completion always comes before perfection.
In terms of tools, this means you need a place where you can “just throw it in” — a card-style, zero-pressure quick-capture entry point (a lot of people use tools like Flomo, or just any window that’s ready to write the moment you open it). Its job is exactly one thing: catch the thought, at the lowest possible cost, in the three seconds it appears. No formatting, no categorizing, no agonizing over where it goes. Catch it first, sort it out later.
Separating the quick-capture entry point from careful refinement is a key design choice at the record layer: capturing should be fast, messy, and careless; refining should be slow, clean, and deliberate. You shouldn’t worry about folder structure the instant a spark of inspiration hits — that’s for later. At this stage, the only enemy is “not writing it down.”
Give Records a Structure: The Five-Step Retrospective
Low friction solves “whether to record,” but if records stay forever at the level of a running diary, it’s hard for them to move toward the knowledge layer. So besides quick-capture, you need a fixed structure to close out each day’s records.
I strongly recommend a five-step retrospective template — a lot of people in the community use it, and it’s simple enough to run every day:
- What did I do today? — State it objectively, without judgment.
- What problem did I run into? — Make the sticking point specific, the more specific the better.
- How did I solve it? — Write down the action and the reasoning, even if it was a lucky guess at the time.
- What knowledge did I actually put to use today? — This one matters most. It forces you to confirm: did what you learned actually get “run”?
- How will I adjust tomorrow? — Loop the retrospective back into action.
These five steps work because they turn “recording” from passive bookkeeping into active processing. Step four especially maps onto a phrase that keeps coming up in the community: runnable knowledge. Knowledge isn’t something you know — it’s something you’ve used. Learning without using is as good as not learning; asking yourself every day “what did I actually put to use today” installs a check valve on whether your records are converting into real capability.
With this structure in place, your records stop being a pile of fragments and become a series of semi-finished products carrying problems and actions, ready to be structured upward. They’re already standing at the door of the knowledge layer.
Next-Day Polish: Using a Cooling-Off Period to Push Records Toward Knowledge
There’s one last step to records — and I think it’s the most elegant one: next-day polish.
The method is dead simple: spend five minutes every day editing — not what you wrote today, but what you wrote yesterday. Cut redundant words, replace vague terms with precise ones, add transitions that connect ideas, break up sentences that run too long.
Why “yesterday”? Because there’s a mechanism hidden here called the cooling-off effect. Looking at your own words a day later, your brain no longer auto-fills the gaps — you see what you actually wrote, not what you think you wrote. At that point you’re more of a reader than a writer. Where it’s wordy, where it’s hollow, where the logic jumps — you can see it instantly. It’s like running the text under cold water; the problems surface on their own.
The significance of this step goes far beyond “making it read more smoothly.” It’s the key step that pushes a semi-finished record toward reusable knowledge:
- Five minutes of editing every day is compound interest from a micro-habit — a flaw you fix today, you naturally avoid tomorrow when writing, and over time it becomes muscle memory, evolving from “write then fix” to “not making the mistake while writing” in the first place;
- The process of editing forces you to re-judge: what did this record actually distill? Is it worth promoting to a formal knowledge card? Anything you can’t fix or can’t clarify gets tagged “needs restructuring” — a sign it isn’t ready yet;
- One sentence to sum up its payoff: trade time for perspective, trade small adjustments for compound returns, go from “can finish it” to “can do it well.”
This rhythm of “quick-capture → five-step retrospective → next-day polish” is essentially the middle segment of the I.P.O. (Input–Process–Output) flow that a lot of skilled practitioners run. Someone in the community turned it into a very concrete pipeline: use Cubox / RSS to roughly capture inspiration and material (the information layer), use Flomo to polish the next day, subtracting and adding (the record layer), then output the finished product in Notion (the creation layer). Records, you can see, sit exactly in the middle — they receive from the information layer upstream, and feed the knowledge and creation layers downstream.
The Test: Have You Changed
One final judgment call, to help you tell whether recording is actually working.
Someone in the community only realized this after three retrospective essays: the test for whether a record is good isn’t “is it well-written” — it’s “have you changed.”
That sentence deserves to sit on every note-taker’s desk. Recording isn’t about producing beautiful documents — it’s about making you actually change. If you’ve been recording for a long time, your text keeps getting neater, but you as a person haven’t moved — it means your records stopped at formatting, and never made it to the knowledge and action layer. Conversely, even if the handwriting is ugly and the sentences are rough, as long as it made you make a different decision the next day, that record succeeded.
Get recording right, and you have a machine that continuously processes information into semi-finished products. But a semi-finished product is still not a finished one — it needs to be structured, called on repeatedly, before it truly becomes your capability.
That’s the job of the next stage: the knowledge layer. We’re going to turn the knowledge base, from a bookmark folder that keeps growing bigger, into a genuine “capability sediment zone” that actually does work. See you in the next essay.
This is the third essay in the “From Information to Creation” column. Previous: Layer One · Information . Next: the third layer — Knowledge.
Responses