Markdown for Note-Taking: Cornell, Zettelkasten, PARA

Most people's notes are a graveyard. A folder of half-finished thoughts from three years ago, a meeting outline with no context, a shopping list mixed in with a project plan. The notes exist but they don't compound — you can't find them when you need them, and you can't build on them later.

Good note-taking systems solve this. Markdown is the substrate that makes them durable. This guide covers three proven systems — Cornell, Zettelkasten, and PARA — translated into practical Markdown workflows.

Why Markdown for notes

Three reasons notes belong in Markdown:

1. Longevity. Plain text is readable in 30 years. Evernote, Notion, or Apple Notes may not exist in 30 years.
2. Speed of capture. Typing - idea is faster than clicking a bullet button. Markdown syntax stays out of your way.
3. Compounding. Interlinked notes ([[other note]]) let you build a second brain. The longer you use the system, the more valuable it becomes — unlike a linear list of notes, which gets harder to navigate over time.

The ecosystem is mature: Obsidian, Logseq, Foam (VS Code), iA Writer, Typora, and dozens of other tools all work with the same underlying .md files. Switch tools without converting data.

System 1: Cornell notes

Cornell is a structured note-taking format developed at Cornell University in the 1950s. Each note page has three regions: a main notes column, a narrower "cues" column for questions and keywords, and a summary at the bottom.

Why it works

• Forces active processing (you write cues after the fact, which reviews the material)
• Creates built-in study material (cover the notes, quiz yourself from cues)
• Scales well for meetings, lectures, and reading notes

Markdown template

---
title: "Topic"
date: 2026-03-04
type: cornell
source: "Where these notes came from"
---

# Topic

## Cues

- Key question 1?
- Key term: definition
- Question 2?

## Notes

Main body of notes, written during the lecture / meeting / reading.
Uses normal Markdown — bullet points, sub-bullets, code blocks,
whatever fits the content.

- Point 1
  - Supporting detail
  - Supporting detail
- Point 2

## Summary

One-paragraph synthesis of what this page covers. Written at the end,
not during. This is the part you'll reread in six months.

Save as cornell/YYYY-MM-DD-topic.md. Use the Cornell Note-Taking template as a starting point.

When to use Cornell

• Lectures, talks, podcasts, YouTube explainers you're actively learning from
• Meetings where you want to retain detail
• Technical books you're studying, not just reading

When not to

• Quick capture of fleeting ideas (overkill)
• Project tracking (use PARA)
• Building long-term knowledge networks (use Zettelkasten)

System 2: Zettelkasten

Zettelkasten ("slip box" in German) is a note-taking system popularized by the sociologist Niklas Luhmann, who produced 70+ books and 400+ papers using it. The core idea: small, atomic, interlinked notes that compound over years.

The three note types

1. Fleeting notes. Quick capture — an idea, a quote, a question. Lives in an inbox.
2. Literature notes. Your own summary of something you read. One per source.
3. Permanent notes. A single atomic idea, written in your own words, linked to related notes. These are the core of the system.

Fleeting notes get processed (converted to permanent notes or deleted) regularly. Permanent notes accumulate and interlink.

Markdown structure

zettelkasten/
├── inbox/              # Fleeting notes — unprocessed
│   └── 2026-03-04.md
├── literature/         # Summaries of sources
│   ├── deep-work-newport.md
│   └── thinking-fast-slow-kahneman.md
└── permanent/          # Atomic ideas, interlinked
    ├── attention-as-capital.md
    ├── context-switching-cost.md
    └── deliberate-practice-requires-feedback.md

Permanent note template

---
title: "One clear idea as the title"
created: 2026-03-04
tags: [attention, productivity]
---

# One clear idea as the title

State the idea in 1–3 paragraphs, in your own words. Not a quote. Not
a summary of someone else's book. Your own understanding.

If you can't state it clearly, you don't understand it yet.

## Related

- [[context-switching-cost]] — explains why attention as capital matters in practice
- [[deliberate-practice-requires-feedback]] — connects to what you do with attention
- [[deep-work-newport]] — the source this originated from

Key principles:

• One idea per note. If you have two ideas, make two notes.
• Atomic. The note should make sense without needing to open any of its links.
• Written in your own words. Paraphrasing forces understanding. Quotes go in literature notes.
• Link generously. Every link is a future path back to this idea from another context.

Why it works

The value comes from links, not notes. After a year of disciplined use, you start seeing unexpected connections — note A and note C, which you wrote six months apart about different topics, turn out to be variants of the same idea. That's where new writing comes from.

When to use Zettelkasten

• Long-term intellectual work (research, writing, teaching)
• Topics you'll think about for years, not weeks
• Any domain where compounding understanding matters

When not to

• Short-lived project work (use PARA)
• Lecture or meeting capture (use Cornell)
• Todo lists (use a task manager)

System 3: PARA

PARA (by Tiago Forte) organizes all your digital life into four categories:

• Projects — short-term efforts with a defined outcome (ship feature X, plan the offsite, write the essay)
• Areas — ongoing responsibilities without a completion date (health, finances, direct reports)
• Resources — topics of interest for future reference (machine learning, woodworking, travel ideas)
• Archive — completed projects, inactive areas, outdated resources

Markdown structure

notes/
├── 1-projects/
│   ├── q1-launch/
│   │   ├── index.md
│   │   ├── timeline.md
│   │   └── meeting-2026-03-04.md
│   └── website-redesign/
├── 2-areas/
│   ├── health/
│   │   ├── index.md
│   │   └── running-log.md
│   ├── finances/
│   └── team-management/
├── 3-resources/
│   ├── machine-learning/
│   ├── public-speaking/
│   └── woodworking/
└── 4-archive/
    └── 2025/

The numeric prefixes keep the folders sorted in a useful order (projects on top, archive on bottom).

Why it works

• Action-oriented. Organizing by what you're doing (not what topic something is about) makes retrieval context-aware. Looking for meeting notes? They're in the project.
• Lifecycle built in. Projects complete, move to archive. Areas shift as life shifts. The system reflects how responsibilities actually evolve.
• Flat enough. Four top-level categories is manageable. Two clicks to any note.

PARA project template

---
title: "Q1 Launch"
status: active
started: 2026-01-15
target: 2026-03-31
---

# Q1 Launch

## Outcome

A single sentence describing what "done" looks like.

## Context

Why this project exists. What it's blocked on. Who owns it.

## Plan

- [ ] Milestone 1
- [ ] Milestone 2
- [ ] Milestone 3

## Log

- **2026-03-04:** meeting notes → [[meeting-2026-03-04]]
- **2026-02-20:** kicked off design review

When to use PARA

• Work-adjacent notes tied to specific projects and responsibilities
• Anyone who has to remember "what's the status of that thing from last month"
• Teams coordinating across multiple active efforts

When not to

• Pure knowledge capture (use Zettelkasten)
• Structured learning (use Cornell)

Mixing systems

You don't have to pick one. A common combination:

notes/
├── daily/              # Daily journal and quick capture
├── cornell/            # Cornell notes from meetings and learning
├── zettel/             # Atomic permanent notes
├── 1-projects/         # PARA projects
├── 2-areas/            # PARA areas
├── 3-resources/        # PARA resources — often links to zettel/
└── 4-archive/          # Everything completed or inactive

The daily folder is an underrated addition. One file per day named YYYY-MM-DD.md catches anything that doesn't fit elsewhere — stray ideas, things you did, people you talked to. Review weekly and promote interesting items to their permanent home.

Tools

The guide is tool-agnostic, but for starting points:

• Obsidian — the most feature-rich Markdown note app; graph view, backlinks, plugins, templates. Free for personal use.
• Logseq — outliner-first; great for daily notes and structured queries.
• VS Code with Foam — if you're already in VS Code for code, Foam adds Zettelkasten features.
• iA Writer — if you mostly write long-form and just want simple folders.

See Best Markdown Editors in 2026 for detailed comparisons.

Starter workflow

If you've never used a system before:

Week 1

• Create a notes/ folder.
• Add two subfolders: daily/ and permanent/.
• Every day, open today's daily file (daily/2026-03-04.md). Write anything that crosses your mind.

Week 2

• Look back at last week's daily notes. Any idea worth keeping → new file in permanent/ with a descriptive name.
• Link related permanent notes with [[note name]].

Month 2

• You'll notice patterns. Add more folders as natural categories emerge (not by theory). Maybe cornell/ for meetings, maybe projects/ for active work.
• Resist the urge to build an elaborate system upfront. Let the folders evolve from your actual notes.

Month 6

• You have 200+ notes. Search is your friend. Links show you connections you didn't plan.
• Pick one of the three systems above if you feel the current structure hitting limits.

What to avoid

• Over-tagging. Tags feel productive but add cognitive overhead when you have to decide which tags to use. Prefer links ([[other note]]) and search over tag taxonomies.
• Daily formatting changes. Your notes system is infrastructure, not a hobby. Set it up, use it, resist redecorating.
• Copying without processing. A note that's just a quote or a clipped article is low-value. Always add your own synthesis — even one sentence — or delete it.
• Hoarding. If you can't find a note or don't remember why you wrote it, delete it. A smaller, used knowledge base is better than a larger, ignored one.

Closing

Cornell, Zettelkasten, and PARA all work. What doesn't work is switching systems every three weeks because the current one isn't perfect. Pick one, run it for six months, then evaluate. Markdown gives you the portability to change your mind without losing anything.

The format is plain text. The discipline is yours.