I have a messy library. It has grown out of hand. But I’ve recently discovered how to navigate the quagmire that is my mental map representation. But first a bit of context:
In my library, all of my atomic permanent notes are placed in one folder. The structure of the library is created by links between notes, a global index that all permanent notes are linked to directly or indirectly, and topic notes (MOCs). The beauty of this is that it is a joy to create new notes, and connect them with ideas already represented in the library (most notes has 3+ connections). To maintain an overview, I create a topic note whenever I felt a need for it. But this became increasingly difficult as the library grew, and when it reached 500+ notes, it became so mentally taxing that I almost stopped creating them all together. This was obviously unstainable (sustainability is a prerequisite for a permanent library), so the issue needed to be solved. How can I avoid being overwhelmed by the number of notes and connections when I try to create structure?
I spendt a couple of months trying to create structure by introducing a tree-like structure based on the Breadcrumbs plugin. I hoped that this could visualize my library in a more structured manner, which could help me navigate the weeds. But this proved to be too much of a chore to set up manually for 500+ notes, and none of the various attempts mitigated my sense of drowning in ideas and connections.
The other day, however, I sort of stubled upon a solution that not only solved my problem, but supercharged my ability to use the library for thinking and writing. But first I need to point out an important principle: when you start a new project you never start with a blank screen / sheet. it’s easy to visualize a student sitting at home in front of a blank screen, trying desperately to force the first sentence in the essay. The entire point of my Obsidian library is to facilitate the emergence of old ideas when I need it. That is, help me retrieve the ideas that are relevant for the current project. So the current solution for achieving this, is by creating all the relevant topic notes I feel are needed to start working on my projects, using the excellent and powerful plugin Graph Analysis ( obsidian://show-plugin?id=graph-analysis ).
The procedure is simple, really:
- Create a new topic note, e.g. “Topic note on perspectives on learning”
- Insert a link to one or more notes I know are relevant for the topic. Check the global index for that keyword, unless I remember one right away. Example: “Learning is creating connections” and “First hand experiences is important for learning”
- Fire up the Graph analysis (GA) plugin, and run an analysis.
- Insert the suggestions from the analysis (that is, has a high score) that seems relevant. This can also include suggestions about older projects and relevant research articles, but I find that advantageous. Create the structure of the topic note the fly as you see fit. See if different types of analyses gives different suggestions. Consider if GA might have missed something.
This should only take a couple of minutes. I spendt about 15 minutes generating three or four pretty comprehensive topic notes which I needed for a paper. The biggest bottle neck seems to be creating the links from the GA suggestions in the topic note.
Remember that my goal was to find a way to avoid a sense of drowning. But in practice, this does not only give a sense of surviving, but rather a sense of discovery from reviving old ideas, and a boost from seeing new connections. The speed of this procedure is important; it helps me not loose track of what the initial aim of creating the topic notes was. Usually this is part of some sort of project I’m working on, and the boost procedure aids the project. Creating structure is no longer providing friction, but lubricant.
A different unexpected positive effect is that the GA analysis emphasizes the holes in my library. This adds an extra motivation for reading to fill that missing piece. And I should add a disclaimer: I’ve kind of recently discovered this solution, so I’m wearing my rose tinted glasses, but am expecting to find some new issues as the personal hype settles.
I also believe there are some prerequisites for this type of procedure to work (that is, features with the structure of the Obsidian library that must or should be in place for the procedure to work). Note that this is personal assumptions based on my experience, so take it with a grain of salt:
- As GA is primarily looking for links between notes, connections are important. That all notes are connected directly or indirectly (trough the index) makes sure no group of notes are ignored by GA.
- The ideas should be organized in atomic notes.
- Relevant ideas are connected with links.
- It’s easy to understand the content of the various notes suggested by the GA analysis. This means it’s an advantage if:
- The usage of key terms is consistent.
- There is descriptive titles in the notes.
So to conclude, this is probably sort of just a rediscovery of the power of generating topic notes (MOCs), that many already know and have internalized. But I hope some see some new possibilites based on my experience, either with topic notes in general, or the power of Graph analysis.