Digital Gardens

A digital garden is a personal website where notes grow and change over time. Unlike blogs, gardens don’t sort by date. Unlike wikis, they’re personal and opinionated.

The term goes back to Mark Bernstein’s 1998 essay “Hypertext Gardens.” The modern revival started around 2015-2020 with developers building alternatives to reverse-chronological blogs.

Gardens vs Blogs

Blogs optimize for publication. You write, polish, hit publish, move on. The post is frozen in time.

Gardens optimize for growth. You plant a seed (rough note), water it (add links, refine ideas), and let it evolve. Nothing is ever “done.”

Joel Hooks put it well: “Chronologically sorted pages of posts aren’t how people actually use the internet.”

Core Principles

Topography over timelines. Notes connect by context, not date. A 2019 note on LLM prompting can link directly to a 2024 note on agent loops.

Continuous growth. Ideas mature publicly. Readers see the process, not just the result. This connects directly to learning in public.

Imperfection is expected. Gardens give you permission to publish rough ideas. “Seedling” or “budding” labels signal that the note isn’t finished.

Dense internal linking. Every note should connect to others. Andy Matuschak calls this “associative ontologies over hierarchical taxonomies.” Links are the garden’s root system.

Evergreen Notes

Andy Matuschak’s “evergreen notes” concept maps cleanly to digital gardens:

1. Atomic: One idea per note
2. Concept-oriented: Named for the concept, not the source
3. Densely linked: Every note connects to others
4. Written for yourself: Default to personal clarity over audience optimization

The goal: notes that stay useful across years and projects. Most people take transient notes that die after one use. Evergreen notes compound.

Practical Implementation

A minimal digital garden needs:

content/
├── notes/
│   ├── digital-gardens.md      # concept note
│   ├── learning-in-public.md   # linked concept
│   └── zettelkasten.md         # related method
└── _index.md                   # garden home (not chronological)

Frontmatter can track note maturity:

+++
title = "Some Concept"
status = "seedling"  # seedling | budding | evergreen
planted = 2024-03-15
tended = 2024-06-20
+++

Tools that work well:

• Static site generators: Hugo, Eleventy, Astro
• Note apps with publish: Obsidian Publish, Notion, Quartz
• Wiki-style: TiddlyWiki, WikiJS

The tool matters less than the practice. Start with folders and Markdown files.

Stock and Flow

Robin Sloan’s 2010 framework explains why gardens matter:

• Flow: tweets, posts, daily updates. Ephemeral.
• Stock: durable content that compounds over time.

Blogs push you toward flow. Gardens push you toward stock. Tom Critchlow: “I spend too much time in flow and not nearly enough in stock.”

Digital gardens are stock machines. Notes accrue value as you tend them.

Connection to Malleable Software

Digital gardens embody malleable software principles. You own the content. You control the structure. You can reshape the system as your thinking evolves.

This is why many gardeners use static sites over platforms. The garden should adapt to you, not the other way around.

What You Can Steal

1. Add status indicators to your notes. Three levels: seedling, budding, evergreen. Readers calibrate expectations.

2. Stop sorting by date. Group by topic, link by association. Kill the archive page.

3. Publish unfinished work. Write a “planted” date and let it sit. Come back when you have more to add.

4. Link aggressively. Every note should have at least 2-3 internal links. If it doesn’t connect, it’s orphaned.

5. Tend regularly. Block 30 minutes weekly to revisit old notes. Add links. Update outdated sections. Prune dead ends.

Sources

• Maggie Appleton, “A Brief History & Ethos of the Digital Garden” (2020)
• Andy Matuschak, “Evergreen Notes”
• Tom Critchlow, “Building a Digital Garden” (2019)
• Joel Hooks, “My Blog is a Digital Garden, Not a Blog”
• Robin Sloan, “Stock and Flow” (2010)

Next: Learning in Public