Learn with Phoebe · Productivity & Knowledge Systems

Learn PKM with Phoebe

Count what is in the place you keep things. Then count the last five times you needed something you knew you had saved, and how many you actually got back. Those two numbers are much further apart than anyone expects, and closing the gap is not a tidiness problem. Six sessions on the loop that makes a collection pay: what to capture and what to call it, where a note lives, what makes it reachable eight months later, and a review short enough to survive a bad week. Tool-agnostic, and honest about what the evidence does and does not support.

6sessions
48notes, one collection
0found by looking, in one folder
14of 16, once linked
45min per session

The six sessions 🎯 · one collection, six passes

The running situation is a real-shaped collection at Daybreak, the coffee-subscription company from the sibling courses: eighteen months of one person's notes, forty-eight of them, two live projects, a reference shelf and six notes called things like "Interesting". Every session does something different to that same collection, and session 4 asks it sixteen questions and counts how many it can still answer.

📥 Session 1 · easy

The collection problem

Saving is one click and finding is a guess, so collections grow and stop paying. What the research supports, what it does not, and the controlled study of PARA that does not exist.

▶ Start here ✂️ Session 2 · easy

Capture that survives

A keep-or-drop rule you can apply in four seconds without arguing with yourself, one inbox, and the title that decides whether a note is ever findable again.

four seconds 🗂️ Session 3 · medium

PARA, and one home per note

Projects, areas, resources, archive, in the author's own words. Organize by actionability rather than by subject, one home plus a pointer, and what filing genuinely cannot fix.

one home 📏 Session 4 · bench night

The retrieval bench

One collection, five filings, sixteen questions asked eight months later. Found by looking, found by searching, or gone. Tidying moves one number and links move the other.

★ the bench 🔁 Session 5 · medium

Review that actually happens

Say what is in a note before you open it, because recalling beats rereading and rereading feels better. The gap that grows with the horizon, and a fifteen-minute weekly pass.

fifteen minutes 🧭 Session 6 · hands-on

A system you keep

The smallest version of the loop that still works when the week goes badly, what to do after a month away, and a final scorecard on the collection you actually have.

🔒 final scorecard
easy - everyone medium hands-on bench night

Your passport 🎫

Clear all three quiz questions in a session and it gets stamped. Six stamps means the collection gives things back. Stamps live in this browser only.

What this is not, so you land in the right place 🧭

If your question isGo here instead
How do I link notes by idea, with atomic notes and permanent addresses, the Zettelkasten way?the Obsidian course, when it lands
How do I read a book properly and turn it into notes?the Reading course, when it lands
How do I build this as databases and dashboards in one specific tool?the Notion and Obsidian courses
How do I capture a live talk on paper while somebody is still speaking?Sketchnoting ↗
How do I give a topic a radial shape with one centre and keyword branches?Mind Mapping ↗
What do I keep, what do I call it, where does it live, and will I ever get it back?you are in the right place

This is the loop, not the tool. Everything here works in whatever you already use, and the author of the best-known method agrees: he calls "a second brain is one app you install" the number one myth about his own work.

Choose your path 🗺️

Three doors into the same conviction: a collection is not what you have, it is what you can get back.

😖 I save things and never find them 1→ 2→ 4 the gap, the title that closes it, and the measurement
🗂️ My system is tidy and useless 3→ 4→ 5 why filing is not finding, proved, then the loop that fixes it
🎯 The whole thing 1→ 2→ 3→ 4→ 5→ 6 from a pile that only grows to a loop you actually keep

What is honest here 🧾

There is no controlled study showing that PARA, a second brain, or any packaged note method improves what you retrieve or produce. That absence is the starting point of session 1, not something worked around. What is evidenced is the mechanisms these methods borrow, and those are quoted with both halves of every finding.

ClaimStatus
Found by looking, found by searching, lost; reachability, orphans, links, tag spreadReally computed from the collection's own graph: a real traversal, a real folder walk, a real keyword match
The collection itselfConstructed. Forty-eight notes written to be realistic. No published figures on real collection size, orphan rate or link density were found, so the numbers describe this collection and nothing else
The verdict line, and the thresholdsJudgement, badged as a heuristic and printed on the widget so they can be argued with. The ordering of the five filings does not depend on them
"PARA is proven to work"Not claimed. Searched for directly across several angles; what exists is a separate, survey-based, correlational literature that does not test these methods
Retrieval practice and spacingQuoted with real numbers from the papers, including the ones the internet gets wrong: no "spacing doubles recall" figure exists in the paper it is attributed to
"Most people never reread their notes"Not stated. No study was found; the figure in circulation traces to one blogger testing their own workflow
TaggingMeasured as cost and return in this collection only. No claim that tagging harms retrieval; the literature reports no clean winner between folders and tags

The knowledge map 🧠

The whole course at a glance - hover a session to spotlight its concepts, click any node to jump in.