Numbering Analog Notes

Some thoughts after talking with Scott Scheper

I had a very beneficial talk with Scott Scheper yesterday, who is “reverse-engineering” Niklas Luhmann’s zettelkasten into an analog not-taking process he calls Antinet. I made a video of the talk, with Scott’s permission, and posted it as the first in a series of talks with people I may call “MakingKnowledge”, if it gets off the ground.

As I was thinking about the conversation, I started to ask myself, if I was going to try to classify my knowledge into broad categories (to build an Antinet), what would those be?

1. History

1. Africa

2. Europe

3. Asia

4. North America

1. Canada

2. US

3. Mexico

5. South America

6. Australia/Oceania

7. World

A. Cosmic time

B. Geological time

C. Early Humans

D. Holocene

E. Ancient history

F. Medieval history

G. Renaissance/Early Modern

H. Enlightenment/Scientific Revolution

I. Age of Revolutions

J. 19th century

K. 20th century

2. Science

3. Knowledge Work

4. Human Behavior

I'd want these to reflect the things I spend my time doing.

I don't think I completely believe in the model I've been using, which is jam everything into Obsidian without thinking much about how it fits. I don't think a structure is going to organically emerge without some cues as to how things should fit together. I think this is why I've tried creating buckets. A few months ago I tried fitting factoids into MOCs. I don't think I really added any value to the vault by doing that. So I stopped and removed all those folders, and put everything into the main list again. But then I started separating things again.

The numbering scheme seems like it's a key issue in a physical slipbox. I'm not particularly interested in having my top-level numbers conform to some type of academic code like CIP (where US History would be 54.0102). If history is "1", world might be "1.7" and North America might be "1.4". I wonder whether it would make sense in a case like that to then use letters for the periods? Where "A" meant cosmic time, "B" geologic time, "C" early humans, "D" Holocene, "E" ancient history, etc. The periodization is of course subjective and arbitrary, to some extent. But so is the top-level categorization.

This reminds me a bit of Leibniz and Waterhouse and Wilkins discussing a "universal language" in Stephenson's Baroque Trilogy. Seems like the point about that was that the effort was ultimately futile and got in the way of actually beginning to work on solutions.

The numbering system needs to be simple enough though, that I can easily add a card. If there's a lot of friction involved in trying to decide on "just the right" number, I'll never get on with writing the card. Maybe the point is, when I'm writing what Scott Scheper calls "Main Cards", I need to be able to very quickly decide where the idea will "fit" in the box before I begin writing the card. This will help determine the number, I imagine, because the new number will follow from an already-existing one.

The implication of this is that the "Main Card" represents a pretty fully-formed idea. It's not an observation I just want to store for later thought or a quote (or even a paraphrase!) from a book I'm reading. That would be something else. A "Fleeting Note" or a "Literature Note" in Ahrens' language. Seems like most of my notes in my Obsidian vault are probably Fleeting or Literature Notes.

But there will still be arbitrariness. If I happen to be interested in, for example, 19th century US History, that might be something like "1.4.2J" which would designate History, North American, US 19th century. If I then wanted to have a category for freethinkers (one of my research interests), that could be "1.4.2J/1" and the lumber industry could be "1.4.2J/2" (where the slash separates between the general topic and the specific). I'd need to make Index cards for each of these main sections right away, so I can remember them. Also, I'll need to jump time periods sometimes; like when my freethinker research extends into the 18th century (1.4.2H/1) or the 20th (1.4.2K/1).

I wonder if this is going to work? I guess there's no way to know but to try it. I don't think it solves all my needs. I still very much like the idea of making little forests of knowledge that might be more like wikis, where students could wander slightly less linearly, maybe by following keywords that cut through the data and group ideas in different ways. But that may be more of a user experience thing (writing output) than an internal organization scheme for me to accumulate and work with ideas.