Andy Matuschak's Evergreen Notes

Andy Matuschak has an interesting system he’s calling Evergreen notes. Here are some of his key points:

• The purpose of notes is better thinking
  • not better notetaking
  • It is easy to fool yourself by taking ineffective notes (may be no notes, or too many notes, or the wrong kind)
• Write about what you read because Writing forces sharper understanding.
• Properties of a good notetaking system: atomicity, densely linked notes, associations over hierarchies, notes are APIs for concepts, do not leave notes disconnected (either across different systems or within a system)
• Understanding requires effortful engagement
• Lots of short cycles > few long cycles
• “Tools for thought should be evaluated in the context of intrinsically meaningful purposes”
  • What meaningful purpose, external to the system or tool itself, is being served?
  • This is a common pitfall with spaced repitition software and communities.
• “An enabling environment significantly expands its participants’ capacity to do things they find meaningful and important.”
  • Examples: schools/universities, corporations/companies, Y Combinator, good software environments, Minecraft
  • Antiexamples: curricula/learning objectives, MOOCs, many games
• Books don’t work

Other links:

• How can we develop transformative tools for thought?

Ironically, what I’ve done here is contra most of Andy’s philosophy. He warns that collecting information, like making a list of bullet points as I have done, is not as effective as it feels. I have developed some fairly rooted habits of information collecting (e.g. Internet bookmarks and real books, and comparatively weaker habits of information engagement and processing, though I’m better off than I could be.

What struck me about Andy’s notes website was the novel format he used to display notes. I actually do not like the UX very much, but the underlying philosophy is interesting and at its core, useful. This was summarized above:

Properties of a good notetaking system: atomicity, densely linked notes, associations over hierarchies, notes are APIs for concepts, do not leave notes disconnected (either across different systems or within a system)

This notebook, so far, does not interact with itself. It feels dead – one entry exists pretty much independently of the others. Andy’s notebook, by contrast, just feels alive. It felt like I had found myself in a corner of Andy’s own mind in a way that an article would not have accomplished.

I’m really thinking about trying to build something like this for myself, and maybe even write my own software to do it as a way to build my chops. Other options, since Andy’s software is not (yet?) publicly available:

• I could go and pay for Roam Research
• vimwiki and vim-zettel
• more ideas here and here
• Obsidian