Obsidian didn't just popularize zettelkasten — it taught a generation how to think in links.

The graph view, the backlinks, the daily notes, the emergent structure from atomic ideas — these weren't just features. They were training wheels for a new way of thinking. Linked thinking. Connection-first knowledge work.

But here's the thing: you don't need Obsidian anymore.

Not because the tool is bad. Because what it taught you is now embedded in how you work. And the LLM is your new linking engine.

What Obsidian Taught Us

Before Obsidian, most people took notes like they organized files: nested folders, hierarchical categories, one big document per topic.

Obsidian showed a different model:

• Linked thinking: Notes connect to other notes. Ideas emerge from connections, not categories.
• Atomic notes: One idea per note. Clear titles. No giant documents that mix everything.
• Zettelkasten: Not a filing system, a thinking system. You don't file — you link, and structure emerges.
• Backlinks: Every note shows what links to it. You don't just find notes; you find relationships.

The graph view was the visualization of something deeper. People got addicted to watching their knowledge base grow into a web.

This was the skill. And it's essential.

What AI Changed

Here's what's different now:

You don't need manual backlinks. The LLM can find connections across your entire vault without you explicitly linking. You write about a concept in five places. The LLM reads all five and surfaces the connections when you ask.

You don't need to remember where you wrote something. You ask the LLM to search your knowledge base. It doesn't just search keywords — it reasons over the content, finds related ideas, suggests connections you'd miss.

You don't need to build the graph by hand. The graph emerges from the LLM's understanding. Ask "what connects to this?" and it traces relationships across your entire corpus.

The skill — linked thinking, atomic notes, emergent structure — transfers. The tool — Obsidian — is optional.

Why Openness Matters

This only works if your notes are in a format AI can read.

Proprietary formats lock you out of AI-native workflows. If your knowledge is trapped in a database that only one app understands, you can't point an LLM at it. You can't ask it to reason over your corpus. You can't get the augmentation.

Markdown is the foundation:

• Open: Anyone can read it, including AI
• Readable: No decoding, no export step, just text
• Portable: Your notes work in any tool, any workflow, any AI
• AI-native: LLMs are trained on markdown. It's their native language.

Your PKM/PIM — personal knowledge management, personal information management — should be LLM-accessible. Not because you need another tool. Because you need your knowledge to work for you.

The New Workflow

1. Write in open formats. Markdown. Plain text. No proprietary lock-in.
2. Let AI reason over it. Point the LLM at your vault. Ask questions. Get connections surfaced.
3. Connections emerge without manual effort. You don't link every note to every related note. The LLM finds the relationships.
4. Your knowledge works for you. It surfaces when you need it, connects to what you're working on, extends your thinking.

This is cognitive augmentation in practice. The LLM doesn't replace your thinking — it remembers for you, reasons with you, surfaces what you'd miss.

Obsidian as On-Ramp

Obsidian is still great for learning the skill. If you've never thought in links, if you've never kept atomic notes, if you've never watched a knowledge base grow into a web — start there.

But the skill outlives the tool. When you switch to AI-native workflows, you bring the skill with you. You don't lose your notes — they're just markdown. You don't lose your structure — the LLM understands it better than any backlink graph.

What you lose is the manual labor of linking everything by hand. The LLM does that now.

This Is Bigger Than Notes

The same philosophy from The New Transparency applies here:

• Openness as default. Your knowledge should be yours, portable, reusable.
• Reuse via AI. The LLM reads your vault, reasons over it, surfaces connections.
• Transparency. You can see what you know because it's in a format you control.

This is the broader shift. Not just in software development, but in knowledge work itself. The LLM is your partner. Your notes, your calendar, your tasks, your history — all connected, all searchable, all reasoning-capable.

The Skill Is Essential. The Tool Is Optional. The Format Is Compulsory.

If you're still using Obsidian, great. Keep using it. It's a good tool.

But don't confuse the tool with the skill. The skill — thinking in links, keeping atomic notes, building emergent structure — is what matters. The LLM gives you that skill without needing the app.

What you need now is:

• Open formats (markdown)
• AI-native workflows (LLM reads your vault)
• Reuse (your knowledge works for you)

That's the shift.