Nobody Reads a Spec Sheet
Your future self is a cold reader. Stop leaving him facts and start leaving him a way back in.
S.T.E.A.M.
Diablo 4 gives you five kinds of finishing work on a piece of gear — sockets, tempering, enchanting, aspects, masterworking — and one late-stage action that locks the item for good. Do the locking step first and you keep whatever the piece was at that moment, forever.
So there’s a mnemonic. STEAM. Socket, Temper, Enchant, Aspect, Masterwork. Finish those before you do the thing you can’t undo.
I picked it up somewhere in the community a few seasons ago. I couldn’t tell you who wrote it, and I’ve never gone looking. Since then the game has taken on an expansion, a new class, an entire new crafting system, and enough balance patches to make most of what I knew obsolete. Players argue about whether the exact order is still optimal. It doesn’t matter. I’ve never looked STEAM up again, and it has outlived several versions of the thing it was written about.
Now: the same stretch of weeks I picked up that acronym, I sat down and worked out how critical strike damage actually functions in that game. Which stats stack together and which ones multiply, and why the number the game displays isn’t the number the game uses. That was real work. An hour, hour and a half, guide open, calculator out.
Here’s what I wrote down.
CSD — bottom value not top. Additive bucket. Diminishing returns.
Nine words. Every one of them accurate.
I have no idea what it means.
Why one of them survived
It isn’t difficulty. The damage material is simpler — it collapses to a single rule, and I’ll show you the rule in a minute. It isn’t attention, either. I paid far more attention to the math than I ever paid to a crafting checklist.
Look at what the mnemonic actually gave me. STEAM is a handle — five letters, no reasoning stored, just a compressed way back to the reasoning. But the handle alone isn’t what saved it. The rule around it did: finish STEAM before the irreversible step. That’s the part that says when the handle matters. A note that doesn’t tell you when it applies never applies.
My note has neither. My note names a topic. That’s all.
And that’s the actual failure, which took me an embarrassingly long time to see. Most notes record what the information was. Useful notes preserve why you’d ever want to go get it. Mine preserved a conclusion and threw away every reason I’d want the conclusion.
Because the person opening that file isn’t me. It’s me in eight months, without the hour and a half, without the guide open, without the calculator, without the state of mind that made nine words sufficient. He isn’t a worse reader than I was. He’s a cold one. The shorthand was never wrong — it just stopped being legible the moment I stopped being saturated in the material.
That’s the problem this article is about. Not clarity for other people. Re-entry for yourself.
You already do this
At work I’d never ship a note like that.
Goal at the top, then symptoms, error messages, possible causes, step-by-step fixes, changelog. If it’s a call: why they called, what I did for them, what the resolution was, next steps if any.
I do it because I know I’ll come back cold, and because someone else might come back instead of me, and because a document nobody can enter is a document that doesn’t exist.
And here’s the part I’d rather not have noticed: I do it at home too. Sometimes. When I’m working through a video guide, I take notes, and every so often one turns into a handle — a mnemonic, something with a shape.
Sometimes.
I make them when the game tells me the mistake is permanent. That lock warning appears on the screen. So I don’t need a new practice; I need to stop letting the material decide.
Because most of what you learn doesn’t come with a warning. Most learning doesn’t announce its own stakes.
The Re-entry Layer
The model I eventually stole wasn’t from note-taking. It was from advertising.
Not because notes need hype. Because advertising is the one craft that assumes, as its starting condition, a reader with almost no context, very little attention, and no obligation whatsoever to care. That’s a fair description of you in eight months.
Four things go on top of the note you already wrote. Together they’re the re-entry layer.
Headline. What the material is for, not what it’s about.
Payoffs. One line per insight, and each line has to carry the insight and why it matters. Bullets are just how they sit on the page; the requirement is the claim.
Open loop. The question you never answered, written as a question rather than a task.
Next action. The single physical thing to do when you pick this up again.
Here’s my electrolyte note. I mix my own version of the LMNT powder: 2,500 mg sodium chloride, 385 mg potassium chloride, 390 mg magnesium malate, in sixteen to thirty-two ounces of water.
That’s what I wrote down. Numbers, correct, done.
Six months later I opened it and the first thing I noticed was that the sodium is five times the potassium, which looked so lopsided I assumed I’d mistyped something. I had no way to check. The only route back was to redo the research — which means the original research bought me nothing.
The version with a re-entry layer:
The mix I made for night cramps — and why the sodium number looks wrong.
— The sodium was intentional. I started this while eating very low carb, when increased sodium loss was part of what I was trying to account for.
— The 5:1 sodium-to-potassium ratio is not a typo. The formula really is built that sodium-heavy, and that’s the thing that will look wrong to me every time I open this.
— Magnesium is the smallest number and the one I understand least.
Still open: the cramps stopped, but I changed two things at once, so I don’t know what did it.
Next: use the kitchen scale. These are weights, not spoonfuls, which is why batch two tasted like the ocean.
Same information. One is a spec sheet. The other is a door.
Now watch what the second line required. To write this looks wrong and isn’t, I had to notice it looked wrong, work out why, and commit to a claim about it. I couldn’t have produced that line by tidying. Reformatting a note moves it around. Writing down what something is for forces you to attempt an interpretation — and attempting one is how you find out whether you actually have it. Sometimes the attempt fails and you discover you were confused. That’s the point.
Which brings me back to the nine words.
Use the bottom number. The headline figure isn’t the one the game uses.
— The tooltip’s top value combines things the damage formula keeps separate. On my character it displayed around 665% when the real gain was closer to 106%.
— The other conditional damage stats are inflated the same way, for the same reason.
Still open: haven’t rechecked any of this since the patch.
Next: hover crit damage, write down the bottom number, redo the math on the gloves.
I could compress that to a handle. Bottom, not top. Very STEAM.
But notice which parts of that note are load-bearing. The percentages will move; they always do. The reason the display disagrees with the math is a different kind of fact — not permanent, nothing in that game is, but it tends to outlive any single number. Put the reasoning above the values. Values expire on a schedule somebody else controls.
This works the same on a book note, a fix you found at 1 a.m., a recipe you adjusted, a prompt that finally behaved. Anything where the version of you who wrote it knew more than the version who has to use it.
They give away the recipe
LMNT publishes that formula on their own site. They’ll tell you exactly how to skip them.
And they still don’t lead with it. What’s out front is “More Salt, Not Less,” and their claim that optimal outcomes sit at sodium levels well above the government number. Not the formula. A reason to care about the formula.
The numbers are free. The reason isn’t the thing they hide — it’s the thing they put first.
I’ve had it backwards the whole time.
The Closed-Note Test
Pick a note you believed was worth keeping. Not the oldest one — an ancient irrelevant note fails for boring reasons. Pick one you’ve carried across two apps and a phone.
Don’t open it.
Write the headline from memory. One line: what is this actually for. Then open it, see how close you got, and fix the line.
That’s the whole method, and it’s also the diagnostic.
Close the source. State the use. Reopen. Repair.
I couldn’t do it on the nine words, and I’m the one who wrote them. That isn’t a filing problem, and it doesn’t mean I never learned the material. It means my note failed to preserve a usable way back in. The information was in there. The road wasn’t.
Don’t go renovate the archive. That’s a weekend that eats itself. Add the layer when you’re making a note that matters, and when an old note earns its way back into use by actually coming up.
The good news is that this takes a few minutes and works on the notes you already have.
The bad news is that software can suggest the hook. Deciding whether the hook is true is still yours.
You know how to do this. You already did it for a video game, using five letters some stranger left on the internet.
Nobody is coming to do that for your notes.
