Coffer

Bases turn notes into views

David V. Kimball — Fri, 09 Oct 2026 00:00:00 GMT

Obsidian's Bases turn a folder of notes into a database: you describe which notes to include, which properties to show, and how to sort them, and Obsidian draws a table or cards. Coffer reads the same Bases syntax and draws the same views on the site, so a base you build in Obsidian publishes as it is.

Write a base in a code block, as below, or embed a saved one with ![[Name.base]]. Every view here comes from this one block:

filters:
  and:
    - file.inFolder("notes")
formulas:
  days: ((now() - published) / 86400000).round()
  topic: file.tags[0]
properties:
  formula.days:
    displayName: Days old
  formula.topic:
    displayName: Topic
  note.published:
    displayName: Published
views:
  - type: table
    name: Newest
    order:
      - file.name
      - published
      - formula.days
      - formula.topic
    sort:
      - property: published
        direction: DESC
    limit: 6
  - type: cards
    name: Cards
    order:
      - file.name
      - published
    sort:
      - property: published
        direction: DESC
  - type: list
    name: By topic
    order:
      - file.name
      - published
    groupBy:
      property: formula.topic
      direction: ASC

The table's "Days old" column is a formula: today's date minus the note's published date, in days. "Topic" is the note's first tag, which the list view also groups by.

What works

Filters with and, or and not; formulas that use properties, other formulas, and functions such as if(), now() and date(); text, number, date and list methods; file.hasTag(), file.inFolder() and file.hasLink(); sorting, grouping, limits and column summaries. Table, cards and list views draw as themselves, and other view types fall back to a table.

A base is drawn when the site is built, so it shows the notes as they were at that moment. It doesn't change while someone reads the page.

Writing in Obsidian with Vault CMS

David V. Kimball — Tue, 06 Oct 2026 00:00:00 GMT

This notebook is an Obsidian vault. Every note is a Markdown file in src/content/notes/, with its images in the same folder. Write and link in Obsidian the way you normally would, and Vault CMS handles the rest: new notes get the right properties, renamed notes keep their old addresses working, and publishing is a commit.

A new note

Create a note from the Notes view on the vault's home base, or in the notes folder. Give it a title that says what it claims, as in [[A note is a unit of thought|a note is a unit of thought]], and set draft: false when it's ready to appear on the site.

Settings

Site settings, such as the title, the accent color, and whether [[Notes open beside each other|panes open beside each other]], live in src/config.ts. In Zeku, Site settings shows them as a form, so you can change them without opening the file.

The vault's home is an Obsidian Base, a database view of every note, page and special page, grouped by folder.

Links work the way they do in Obsidian

David V. Kimball — Mon, 05 Oct 2026 00:00:00 GMT

Link one note to another the way you already do in Obsidian: type [[, pick the note, and keep writing. On the site, that link goes to the same note it opens in Obsidian, and in a wide window it [[Notes open beside each other|opens beside the current note]].

See [[Kaizen]] for the idea of small, steady changes.

Coffer reads links with Obsidian's own rules, so you never have to think about them:

A name finds that note in any folder: [[Kaizen]]. Capitals and the .md don't matter.
When two notes share a name, the one in the same folder wins, then the one with the shortest path. Write part of the path, such as [[ideas/Kaizen]], to choose the other.
[[Kaizen#Origins in Manufacturing]] goes to a heading, as in [[Kaizen#Origins in Manufacturing]], and [[#Heading]] goes to a heading in the same note.
[[Kaizen|small steps]] shows your own words: [[Kaizen|small steps]].
Markdown links work too, such as [small steps](Kaizen.md), if your vault is set to use them.

A link to a note that doesn't exist yet, or one that isn't published, shows faded and does nothing, the way Obsidian shows an [[Unwritten note]].

Embed a note

Put ![[Note]] on a line of its own to show another note's text inside this one, or ![[Note#Heading]] for a single section. This is the section on narrow screens from the note on sidenotes:

![[Sidenotes keep context beside the text#On narrow screens]]

Images, audio, video and PDFs embed the same way, such as ![[mountain-in-japan.jpg]]. Add a width to size an image: ![[mountain-in-japan.jpg|240]].

![[mountain-in-japan.jpg|240]]

Backlinks come for free

Every note ends with "Links to this note", which lists the notes that link to it and the paragraph each link sits in. You never maintain that list. It comes from the links you already wrote.

Sidenotes keep context beside the text

David V. Kimball — Sun, 04 Oct 2026 00:00:00 GMT

A footnote makes you leave the sentence you're reading, find the note at the bottom of the page, and find your way back. A sidenote sits in the margin, level with the line that refers to it, so a glance is enough.[^1]

In a note, write sidenotes as ordinary Markdown footnotes. Obsidian shows them as footnotes while you write, and the published note shows them in the margin.[^2] Nothing else changes in how you write.

On narrow screens

When there's no room for a margin, on a phone or inside a pane, a sidenote folds into the text. Its number stays in place, and tapping the number opens the note right below the line. Tap it again to close it.[^3]

When to use one

Use a sidenote for something a curious reader might want and everyone else can skip: a source, a caveat, a small example. If the point matters to the argument, it belongs in the sentence.

[^1]: This is a sidenote. On a wide screen it sits in the margin to the right of the column. [^2]: The syntax is [^1] where the note is referenced and [^1]: Your note. anywhere below it. [^3]: Edward Tufte's books, and the Tufte CSS project, made sidenotes familiar on the web.

Notes open beside each other

David V. Kimball — Sat, 03 Oct 2026 00:00:00 GMT

On a screen at least 1,024 pixels wide, following a link from one note to another doesn't replace the page. The linked note opens as a new pane to the right, and the note you came from stays where it was.

Keep following links and the panes stack up. Older ones slide under newer ones and leave a narrow spine with their title, so you can always see the trail you took. Click a spine to bring that note back.[^1]

Why read this way

Reading linked notes one page at a time forces a choice: follow the link and lose your place, or stay and lose the thread. Panes remove that choice. You can look at what a link points to and come straight back, which is how you'd read a stack of index cards on a desk.

This only works because each note is small. A pane is the width of one comfortable column, the same width as a single note on its own page. See [[A note is a unit of thought|a note is a unit of thought]].

The stack lives in the address

The address bar keeps the whole stack, so you can copy it and send someone the exact trail you followed. The browser's Back button closes the last pane you opened. Every note still has its own address too, so a link from outside always lands on a single note.

On a phone or a narrow window, links behave normally and open one note at a time.

[^1]: The idea comes from Andy Matuschak's working notes, which popularized stacked panes for reading linked notes.

A note is a unit of thought

David V. Kimball — Fri, 02 Oct 2026 00:00:00 GMT

A note in this notebook holds one idea. Its title says what the idea is, as a short claim rather than a topic: not "Habits" but "small changes outlast big ones". A title like that tells you what you'll find before you open it, and it reads well in the middle of a sentence when another note links to it.

Keeping each note to one idea is what makes linking worth it. A long essay can only be linked as a whole, and the reader has to dig for the part that mattered. A small note can be linked exactly where it's relevant, the way [[Kaizen|kaizen]] links to the habit research it builds on.[^1]

Notes grow by being linked

A new note rarely arrives finished. It starts as a few lines, then picks up a second paragraph when another note needs it, then a counterexample a month later. The updated date on each note shows when it last changed.

Over time the links matter more than any single note. Following them is how you find connections you didn't plan, which is the whole argument of [[Building a Second Brain|building a second brain]].

Where this comes from

The practice is old. Commonplace books, Niklas Luhmann's slip box, and Andy Matuschak's working notes all keep small, linked notes rather than finished essays.[^2]

[^1]: Links between notes work as they do in Obsidian, as described in [[Links work the way they do in Obsidian]]. [^2]: Luhmann kept about 90,000 index cards over four decades and credited them with most of his books.

Formatting Reference

David V. Kimball — Mon, 09 Mar 2026 00:00:00 GMT

This note demonstrates all the markdown, extended markdown, and formatting features available in Coffer. Use it as both a reference guide and a showcase of what's possible.

Basic Formatting

Text Emphasis

Bold text using **bold** or __bold__
Italic text using *italic* or _italic_
Bold and italic using ***text***
~~Strikethrough~~ using ~~text~~
==Highlighted text== using ==text==
Inline code using backticks

Headings

Heading 1

Heading 2

Heading 3

Heading 4

Heading 5

Heading 6

# Heading 1
## Heading 2
### Heading 3
#### Heading 4
##### Heading 5
###### Heading 6

Callouts and Admonitions

Coffer supports Obsidian-style callouts. Each type has its own icon; cautionary types take the accent color.

Basic Callouts

[!note] This is a note callout. Use it for general information that readers should be aware of.

[!tip] This is a tip callout. Perfect for helpful suggestions and best practices.

[!important] This is an important callout. Use it to highlight critical information.

[!warning] This is a warning callout. Use it to alert readers about potential issues.

[!caution] This is a caution callout. Use it for dangerous or risky situations.

Custom Titles

[!note] Custom Note Title You can customize the title of any callout by adding text after the callout type.

[!tip] Pro Tip Custom titles help you provide more context for your callouts.

Collapsible Callouts

You can make callouts collapsible by adding + (expanded by default) or - (collapsed by default) after the callout type:

[!note]+ Expanded by Default This callout starts expanded and can be collapsed by clicking the toggle button or the title.

[!warning]- Collapsed by Default This callout starts collapsed and can be expanded by clicking the toggle button or the title.

[!tip]+ Collapsible with Custom Title You can combine collapsible functionality with custom titles for more control over your content organization.

Extended Callout Types

[!info] Info callouts provide additional context or details.

[!success] Success callouts highlight positive outcomes or achievements.

[!question] Question callouts can be used to pose questions or highlight areas of uncertainty.

[!example] Example callouts are perfect for showcasing code examples or demonstrations.

[!quote] Quote callouts can be used to highlight important quotes or references.

Callouts with Formatting

[!example] You can use markdown syntax in callout content like italics, bolded text, or links.

Media Content

Images

Single Image

Multiple Image Layouts

Coffer automatically arranges consecutive images in responsive grid layouts. Place images together without empty lines between them to create these layouts.

Two Images Side by Side

Three Images in a Row

Four Images in a Row

How to Use Multiple Images / Gallery

Simply place multiple images together without empty lines between them:

![Image 1](image1.jpg)
![Image 2](image2.jpg)
![Image 3](image3.jpg)

On mobile devices, all layouts switch to a single column for better readability.

Linked Images

[![Glacier](glacier.jpg)](https://obsidian.md)

Lists

Unordered Lists

First item
Second item
- Nested item
- Another nested item
  - Deeply nested item
Third item

Ordered Lists

First step
Second step
1. Sub-step A
2. Sub-step B
  1. Sub-sub-step
Third step

Task Lists

[x] Completed task
[ ] Incomplete task
[x] Another completed task
- [ ] Nested incomplete task
- [x] Nested completed task
[ ] Final incomplete task

Links and References

External Links

Here's an external link.

Internal Links

Link to another note by name, as in Obsidian.

For example: [[Obsidian Embeds Demo]] or [[Kaizen|small, steady changes]].

[[Obsidian Embeds Demo]]
[[Kaizen|small, steady changes]]

Here's an internal link to a heading:

[[Kaizen#Origins in Manufacturing]]

Reference Links

This is a reference link and this is another reference link.

Code Blocks

Inline Code

Use const variable = 'value' for inline code snippets.

JavaScript

function greetUser(name) {
  console.log(`Hello, ${name}!`);
  return `Welcome to our blog, ${name}`;
}

const user = "Developer";
greetUser(user);

Python

def calculate_fibonacci(n):
    """Calculate the nth Fibonacci number."""
    if n <= 1:
        return n
    return calculate_fibonacci(n-1) + calculate_fibonacci(n-2)

# Example usage
for i in range(10):
    print(f"F({i}) = {calculate_fibonacci(i)}")

CSS

.button {
  background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
  border: none;
  border-radius: 8px;
  color: white;
  padding: 12px 24px;
  transition: transform 0.2s ease;
}

.button:hover {
  transform: translateY(-2px);
}

Bash/Shell

#!/bin/bash
echo "Setting up development environment..."

# Install dependencies
pnpm install

# Start development server
pnpm dev

echo "Development server started on http://localhost:4321"

Tables

Basic Tables

Feature	Supported	Notes
Markdown	Yes	Full CommonMark support
Callouts	Yes	Multiple types with icons
Math	Yes	LaTeX math with KaTeX rendering
Diagrams	Yes	Mermaid diagram support
Galleries	Yes	Auto grid layout for images

Advanced Tables

Language	Use Case	Performance	Learning Curve
JavaScript	Web Development	Great	Easy
Python	Data Science	Good	Easy
Rust	Systems Programming	Excellent	Hard
Go	Backend Services	Great	Medium

Blockquotes

Simple Quotes

The best way to predict the future is to invent it. -- Alan Kay

Nested Quotes

This is a top-level quote.

This is a nested quote within the first quote.

And this is a quote nested even deeper.

Back to the top level.

Footnotes

Coffer turns footnotes into sidenotes, which sit in the margin on wide screens [^1]. You can add multiple footnotes throughout your content [^2].

[^1]: This is a footnote. On desktop screens wider than 1280px, footnotes appear as sidenotes in the right margin.

[^2]: A second footnote. They're numbered automatically.

Text with a footnote [^1].

[^1]: Footnote content here.

GitHub Repository Cards

Dynamic cards that pull repository info from the GitHub API on page load.

https://github.com/davidvkimball/vaultcms

```embed
https://github.com/owner/repo
```

Buttons

<div class="btn-group"> <a href="https://astro.build" class="no-styling no-underline" target="_blank"><button class="btn-primary" name="button">Primary</button></a> <a href="https://astro.build" class="no-styling no-underline" target="_blank"><button class="btn-secondary" name="button">Secondary</button></a> <a href="https://astro.build" class="no-styling no-underline" target="_blank"><button class="btn-outline" name="button">Outline</button></a> <a href="https://astro.build" class="no-styling no-underline" target="_blank"><button class="btn-ghost" name="button">Ghost</button></a> </div>

<a href="https://astro.build" class="no-styling no-underline" target="_blank"><button class="btn-large" name="button">Large Button</button></a>

<div class="btn-group">
  <button class="btn-primary">Primary</button>
  <button class="btn-secondary">Secondary</button>
  <button class="btn-outline">Outline</button>
  <button class="btn-ghost">Ghost</button>
</div>
<button class="btn-large">Large</button>

Horizontal Rules

HTML Elements

You can use HTML directly. Here are some examples:

<details> <summary>Click to expand</summary>

This content is hidden by default and can be expanded by clicking the summary.

</details>

<small>Small text for fine print</small>

<sup>Superscript</sup>

<sub>Subscript</sub>

Keyboard Shortcuts

Press <kbd>Ctrl</kbd> + <kbd>C</kbd> to copy and <kbd>Ctrl</kbd> + <kbd>V</kbd> to paste.

Use <kbd>Ctrl</kbd> + <kbd>K</kbd> to open the command palette.

Special Characters and Symbols

Copyright: ©
Trademark: ™
Registered: ®
Arrows: ← ↑ → ↓
Currency: $ € £ ¥

Works with Obsidian

All of these formatting options render in Obsidian as well, with some differences depending on the theme you use.

Quick Reference

Bold: **text** or __text__
Italic: *text* or _text_
Code: `code`
Highlight: ==text==
Strikethrough: ~~text~~
Links: [text](url)
Images: ![alt](url)
Lists: - or 1. for ordered
Tasks: - [ ] and - [x]
Tables: Use | to separate columns
Quotes: Start lines with >
Callouts: > [!TYPE]
Footnotes: [^1] with [^1]: text below
Horizontal rule: ---
GitHub cards: ```embed with GitHub URL
Buttons: <button class="btn-primary">Text</button>

Further reading: [[Obsidian Embeds Demo]]

Obsidian Embeds Demo

David V. Kimball — Mon, 09 Mar 2026 00:00:00 GMT

This note shows the advanced embed features available in Coffer, including media embeds, auto-embeds, math rendering, and Mermaid diagrams.

Auto-embeds

YouTube Videos

YouTube URLs are automatically embedded as responsive video players. Just paste a standard YouTube link as a markdown image:

![Astro Suite for Obsidian](https://www.youtube.com/watch?v=3zeqJ5tqmaQ)

Twitter/X Posts

Twitter/X post URLs are automatically embedded with theme-aware styling that matches your site's light/dark mode:

![Why doesn't everyone use Astro?](https://x.com/davidvkimball/status/1933196479801536736)

Media Embeds

Coffer embeds local media files directly in your notes using standard markdown image syntax. Place the files in the note's folder and reference them.

Video

![Demo video](video.mp4)

Audio

![Demo audio](sound.wav)

PDF

![Sample document](document.pdf)

Math

Coffer renders LaTeX math with KaTeX, in both light and dark mode.

Inline Math

Use single dollar signs for inline math: $E = mc^2$ or $\int_0^{2\pi} d\theta x+e^{-i\theta}$.

Display Math

Use double dollar signs for centered display math:

$$ \begin{vmatrix}a & b\ c & d \end{vmatrix}=ad-bc $$

$$ f(x) = x^2 + 3x + 2 $$

Common Mathematical Notation

Fractions and Superscripts

Fractions: $\frac{a}{b}$, $\frac{x^2 + 1}{x - 1}$
Superscripts: $x^2$, $e^{i\pi} + 1 = 0$
Subscripts: $x_1$, $H_2O$

Greek Letters

$\alpha, \beta, \gamma, \delta, \epsilon, \theta, \lambda, \mu, \pi, \sigma, \phi, \omega$
$\Gamma, \Delta, \Theta, \Lambda, \Pi, \Sigma, \Phi, \Omega$

Mathematical Symbols

Summation: $\sum_{i=1}^{n} x_i$
Product: $\prod_{i=1}^{n} x_i$
Integral: $\int_{-\infty}^{\infty} e^{-x^2} dx = \sqrt{\pi}$
Limit: $\lim_{x \to 0} \frac{\sin x}{x} = 1$

Matrices and Vectors

$$ \begin{pmatrix} a & b \ c & d \end{pmatrix} \begin{pmatrix} x \ y \end{pmatrix}

\begin{pmatrix} ax + by \ cx + dy \end{pmatrix} $$

Complex Equations

$$ \nabla \times \vec{E} = -\frac{\partial \vec{B}}{\partial t} $$

$$ i\hbar\frac{\partial}{\partial t}\Psi(\vec{r},t) = \hat{H}\Psi(\vec{r},t) $$

Math in Callouts

Math works perfectly within callouts:

[!note] Mathematical Proof The Pythagorean theorem states that for a right triangle: $$a^2 + b^2 = c^2$$

Where $c$ is the hypotenuse and $a$ and $b$ are the other two sides.

[!tip] Integration by Parts The formula for integration by parts is: $$\int u , dv = uv - \int v , du$$

This is particularly useful for integrals involving products of functions.

Advanced Mathematical Typesetting

Aligned Equations

$$ \begin{aligned} f(x) &= ax^2 + bx + c \ f'(x) &= 2ax + b \ f''(x) &= 2a \end{aligned} $$

Cases and Piecewise Functions

$$ f(x) = \begin{cases} x^2 & \text{if } x \geq 0 \ -x^2 & \text{if } x < 0 \end{cases} $$

Set Notation

Natural numbers: $\mathbb{N} = {1, 2, 3, \ldots}$
Real numbers: $\mathbb{R}$
Complex numbers: $\mathbb{C}$
Set union: $A \cup B$
Set intersection: $A \cap B$
Subset: $A \subseteq B$

Obsidian Compatibility

All math notation works identically in Obsidian and your published site:

Inline math: $...$ syntax
Display math: $$...$$ syntax
LaTeX commands: Full support for standard LaTeX math commands
Greek letters: Use \alpha, \beta, etc.
Symbols: Use \sum, \int, \infty, etc.

Diagrams

Coffer renders Mermaid diagrams in the site palette, in both light and dark mode.

Flowchart

graph TD
    A[Start] --> B{Is it?}
    B -->|Yes| C[OK]
    C --> D[Rethink]
    D --> B
    B ---->|No| E[End]

Sequence Diagram

sequenceDiagram
    participant Alice
    participant Bob
    participant John
    Alice->>John: Hello John, how are you?
    loop Healthcheck
        John->>John: Fight against hypochondria
    end
    Note right of John: Rational thoughts prevail!
    John-->>Alice: Great!
    John->>Bob: How about you?
    Bob-->>John: Jolly good!

Class Diagram

classDiagram
    class Animal {
        +String name
        +int age
        +makeSound()
    }
    class Dog {
        +String breed
        +bark()
    }
    class Cat {
        +String color
        +meow()
    }
    Animal <|-- Dog
    Animal <|-- Cat

State Diagram

stateDiagram-v2
    [*] --> Still
    Still --> [*]
    Still --> Moving
    Moving --> Still
    Moving --> Crash
    Crash --> [*]

Entity Relationship Diagram

erDiagram
    CUSTOMER ||--o{ ORDER : places
    ORDER ||--|{ LINE-ITEM : contains
    CUSTOMER }|..|{ DELIVERY-ADDRESS : uses

User Journey

journey
    title My working day
    section Go to work
      Make tea: 5: Me
      Go upstairs: 3: Me
      Do work: 1: Me, Cat
    section Go home
      Go downstairs: 5: Me
      Sit down: 5: Me

Gantt Chart

gantt
    title A Gantt Diagram
    dateFormat  YYYY-MM-DD
    section Section
    A task           :a1, 2024-01-01, 30d
    Another task     :after a1  , 20d
    section Another
    Task in sec      :2024-01-12  , 12d
    another task     : 24d

Pie Chart

pie title Pets adopted by volunteers
    "Dogs" : 386
    "Cats" : 85
    "Rats" : 15

Git Graph

    gitGraph
       commit
       commit
       branch develop
       checkout develop
       commit
       commit
       checkout main
       merge develop
       commit
       commit

Complex Flowchart

graph TD
    A[Christmas] -->|Get money| B(Go shopping)
    B --> C{Let me think}
    C -->|One| D[Laptop]
    C -->|Two| E[iPhone]
    C -->|Three| F[Car]
    D --> G[Code on it]
    E --> H[Use it]
    F --> I[Drive it]

Responsive Design

All diagrams are responsive and work well on mobile devices, with proper scaling and overflow handling.

Building a Second Brain

David V. Kimball — Sun, 08 Mar 2026 00:00:00 GMT

The idea of a "second brain" is simple: offload what you know into a system you trust, so your actual brain can focus on thinking instead of remembering.

Why It Matters

Most of us consume far more information than we retain. Articles, podcasts, conversations, ideas in the shower. Without a capture system, it all evaporates.

A good PKM (personal knowledge management) setup gives you:

Capture without friction
Organization that scales
Retrieval when you need it most

The term was popularized by Tiago Forte[^1], but the underlying idea is ancient. Commonplace books, zettelkasten, and marginalia all served the same purpose: extending human memory beyond its biological limits.

[^1]: Tiago Forte's book Building a Second Brain (2022) formalized the PARA method and the concept of progressive summarization.

The Tools I Use

I've written more about my specific setup in [[My Complete Note-Taking Workflow]]. The short version: I use Obsidian as my primary tool, with a vault that syncs across devices.

The key insight is that the tool matters less than the habit. Pick something, commit to it for 90 days, and iterate.

What Makes a Good Tool

Not every app works for this. The ones that stick tend to share a few traits:

Local-first storage. Your notes live on your machine, not behind a login wall.
Plain text or Markdown. Future-proof formats you can read anywhere.
Linking between notes. The ability to connect ideas is what turns a folder into a knowledge graph.
Low friction capture. If it takes more than a few seconds to create a note, you won't do it.

Connecting Ideas

The real power comes from linking. When you connect a note about one topic to another, unexpected insights emerge. This is the core of networked thought.

Your second brain becomes a thinking partner, not just a filing cabinet.

Progressive Summarization

One technique that helps: when you revisit a note, bold the most important sentences. Next time, highlight the bolded ones. Over time, the signal rises to the top without you needing to rewrite anything[^2].

[^2]: Progressive summarization is a layered highlighting technique. Each pass distills the note further without losing the original context.

Common Mistakes

[!warning] Avoid these pitfalls Perfectionism kills PKM systems faster than anything else. Don't reorganize. Don't build elaborate folder hierarchies. Just capture and link.

A few patterns that derail people:

Over-organizing too early. Structure should emerge from use, not be imposed upfront.
Tool-hopping. Switching apps every month means you never build a critical mass of notes.
Treating it like a task list. A second brain is for reference and thinking, not for tracking to-dos.

Getting Started

Pick one tool (Obsidian, Notion, even a folder of text files)
Start capturing everything interesting for two weeks
Review weekly and start linking related ideas
Let the structure emerge organically

A 30-Day Challenge

Week	Focus	Goal
1	Capture	Write at least one note per day
2	Link	Connect every new note to at least one existing note
3	Review	Spend 30 minutes reviewing and refining
4	Create	Produce one piece of output from your notes

By the end of the month, you'll know whether the system works for you. The notes you've accumulated become proof of the concept.

The principles here overlap heavily with the [[Kaizen]] philosophy of continuous improvement. Small daily deposits compound into something significant.

The goal isn't a perfect system. It's a system that makes you think better.

Ikigai

David V. Kimball — Fri, 06 Mar 2026 00:00:00 GMT

Ikigai (生き甲斐) is often translated as "reason for being" or "that which makes life worth living." It's the sense of purpose that gets you out of bed in the morning.

The Four Circles

In popular use, ikigai is pictured as the overlap of four areas:

What you love (passion)
What you're good at (skill)
What the world needs (mission)
What sustains you (livelihood)

The sweet spot isn't one of these alone but where they meet. Work or life that feels meaningful, useful, and aligned with your strengths.

The Western Misinterpretation

The Venn diagram version of ikigai is actually a Western invention[^1]. In Japan, the concept is broader and less career-focused. A retired grandmother who tends her garden every morning has ikigai. A child building with blocks has ikigai. It doesn't require monetization or world-changing impact.

[^1]: The four-circle Venn diagram often attributed to ikigai was created by Marc Winn in 2014, combining ikigai with a separate framework by Andres Zuzunaga. It's useful but not authentically Japanese.

The original meaning is closer to "the feeling that life is worth living." Neuroscientist Ken Mogi describes five pillars of ikigai[^2]:

Starting small
Accepting yourself
Connecting with others and the world
Seeking small joys
Living in the here and now

[^2]: Ken Mogi, The Little Book of Ikigai (2017).

Finding Your Own

The idea is less about a single "calling" and more about paying attention to the small things that give you satisfaction: a walk, a conversation, a project that uses your skills, or the rewards of [[Kaizen|continuous improvement]].

[!tip] A practical exercise Write down three moments from the past week where you lost track of time. What were you doing? Who were you with? Those are clues.

In that sense, ikigai is something you notice and nurture through how you spend your time, not something you discover once and keep forever.

Ikigai and Knowledge Work

For those of us who work with ideas, ikigai often shows up in the process itself. The act of writing a clear sentence, solving a tricky problem, or helping someone understand something new. These are small moments of purpose that compound over time.

Building a system that supports this kind of work, like a [[Building a Second Brain|second brain]], is itself a form of ikigai. The system serves the purpose. The purpose feeds the system.

Kaizen

David V. Kimball — Thu, 05 Mar 2026 00:00:00 GMT

Kaizen (改善) means "change for better." It's continuous, incremental improvement. A mindset and a practice: improve a little every day instead of waiting for the perfect plan or the big push.

Origins in Manufacturing

Kaizen emerged in post-war Japan as a management philosophy at Toyota[^1]. The idea was radical in its simplicity: every employee, from the assembly line to the executive floor, should look for small improvements every single day.

[^1]: The Toyota Production System, developed by Taiichi Ohno and Eiji Toyoda in the 1950s, is widely credited with formalizing kaizen as an industrial practice.

The principles were straightforward:

Small experiments over grand plans
Quick feedback over long review cycles
Steady refinement over dramatic overhauls

These same principles later influenced lean manufacturing, agile software development, and countless productivity systems.

Kaizen for Personal Work

The same idea applies to habits and creative work. One small edit, one short session, one clear next step is enough. Consistency matters more than intensity.

The 1% Rule

If you improve by 1% each day, after a year you're 37 times better[^2]. The math is compelling, but the real insight is psychological: small improvements don't trigger resistance. Your brain doesn't fight a five-minute writing session the way it fights a two-hour commitment.

[^2]: This is the compound interest metaphor applied to habits: 1.01^365 = 37.78. James Clear popularized this framing in Atomic Habits (2018).

Practical Applications

Area	Big change (hard)	Kaizen approach (sustainable)
Writing	"Write 2,000 words a day"	Write one paragraph after morning coffee
Exercise	"Go to the gym 5x/week"	Do 10 pushups before your shower
Reading	"Read 50 books this year"	Read one page before bed
Note-taking	"Reorganize my entire vault"	Link one note to another each day

Why It Works

The appeal of kaizen is that it's sustainable. Big, sudden changes are hard to maintain; small steps compound without burning you out. You don't need to be twice as good tomorrow. Just a bit better than yesterday.

[!note] The compound effect Most people overestimate what they can do in a day and underestimate what they can do in a year. Kaizen respects this asymmetry.

Kaizen and Purpose

There's a natural connection between kaizen and [[Ikigai|ikigai]]. The daily practice of improving, even in tiny ways, creates a sense of forward motion. That feeling of progress is itself a source of meaning.

Building a [[Building a Second Brain|second brain]] is kaizen applied to knowledge. Every note captured, every link made, every review completed is a small improvement to your thinking infrastructure.

My Complete Note-Taking Workflow

David V. Kimball — Wed, 04 Mar 2026 00:00:00 GMT

After years of trying different systems, I've landed on a workflow that actually sticks. Here's the full breakdown.

Daily Capture

Every morning I open my daily note and jot down what's on my mind. Throughout the day, I capture anything interesting. Fleeting notes, quotes, article highlights, random ideas.

The rule: capture first, organize later. If I try to categorize in the moment, I lose the thought.

What I Capture

Not everything deserves a note. Over time I've developed a filter:

Ideas I might act on. Even half-formed ones.
Things that surprised me. Surprise means my mental model was wrong, which means learning.
Connections between topics. "This reminds me of..." is always worth writing down.
Quotes that resonate. But only if I can articulate why they resonate[^1].

[^1]: A quote without your own commentary is just someone else's thought. The value comes from your reaction to it.

Weekly Review

Every Sunday I spend 30 minutes reviewing the week's captures. This is where the magic happens:

Promote fleeting notes to permanent notes
Add links to related ideas
Tag anything that connects to active projects
Delete anything that no longer seems interesting

This is closely tied to the principles I wrote about in [[Building a Second Brain]]. The review habit is what turns a pile of notes into a knowledge system.

The Review Checklist

I keep a simple template for the weekly review:

Open this week's daily notes
Star anything worth keeping
For each starred item: create a proper note or add it to an existing one
Link the new notes to at least two existing notes
Check the "unlinked" list for orphaned notes
Archive the daily notes

Tools and Setup

My stack:

Tool	Purpose	Why I chose it
Obsidian	Writing and linking	Local-first, Markdown, incredible plugin ecosystem
Readwise	Book/article highlights	Auto-syncs to Obsidian
Physical notebook	Meetings and sketches	Some things are better on paper

I publish my notes using a static site generator. The same principles that make code maintainable (clear structure, minimal duplication, good naming) apply to knowledge bases too.

Folder Structure

I keep it minimal:

vault/
  daily/          # Daily notes, auto-generated
  notes/          # Permanent notes, linked
  projects/       # Active project folders
  references/     # Source material
  templates/      # Note templates

The key insight: folders are for types of notes, not topics. Topics emerge from links and tags, not from where a file lives[^2].

[^2]: This is the difference between a filing cabinet (hierarchical) and a wiki (networked). Both have value, but for knowledge work, the network wins.

Note Types

Not all notes are the same. I use three tiers:

Fleeting Notes

Quick captures. A sentence or two. No formatting required. These live in daily notes and get processed during the weekly review. Most of them get deleted or absorbed into permanent notes.

Literature Notes

Summaries and reactions to things I've read. Always include the source and my own commentary. "The author argues X. I think this connects to Y because Z."

Permanent Notes

The building blocks of the system. Each one is:

Atomic: one idea per note
Self-contained: makes sense without context
Linked: connected to at least two other notes
Written in my own words: not a copy-paste

What I've Learned

The biggest lesson: your note-taking system should reduce anxiety, not create it. If you feel guilty about an inbox of unprocessed notes, your system is too rigid.

[!tip] The golden rule of note-taking If your system makes you feel guilty instead of productive, simplify it. Delete the rules that create pressure without creating value.

The [[Kaizen]] approach applies perfectly here. Don't try to build the perfect system in a weekend. Make one small improvement each week. After a few months, you'll have something that genuinely serves how you think.

Keep it simple. Link generously. Review regularly. That's it.