TLDR: I’m using Vault Changelog and Incremental Writing Queue to create “black hole” notes. The Changelog keeps track of the last N notes (N=100 in my case) and IRQ adds all new notes to my Default Queue (DQ). Higher tension tells me I need to do either/both of fleshing/filtering out new notes and/or revisiting old notes to relieve the tension.
This is before going through DQ for 30m:
And after: (same settings, just looks smaller because I had to zoom out further)
I color my indices (Index notes) red and look at those first. There’s a lot to glean from the graph but one example is that even after relieving the tension, I can immediately see there’s still a lot of notes clustered between my Computer Science Index (upper center) and IRQ (“adult babies” as I call them below).
Problem
The main goal here is that I want the quality of my vault to increase, so I need to clean out and prevent “weak” notes. I had a problem of frequently creating notes that had few/no links with low-quality content and would never revisit them (this made worse by inadvertently filtering orphans from the graph). That incentivized me to over-link for the sake of having a dense graph and not “losing” notes, but the graph was useless at that point and I was mentally searching for links to add rather than having concepts pop up in my head organically.
Explanation
The Changelog concentrates notes by edit recency and the Queue concentrates by both creation recency and completeness. The Changelog also concentrates by creation recency if I’m editing notes slower than I’m creating them. That should happen less often as my vault ages since I’ll do more editing than creating, but if that’s a problem I can always lower N to 50 or 25 or whatever.
Notes clustered between the two are like “baby notes.” They’ve either been created recently or I’m still actively working on them. Notes clustered beside the DQ are like “adult babies.” They’re getting old but aren’t mature enough to have left the DQ (these are higher priority than the babies). Notes clustered beside the Changelog just serve to attract graph communities that I’m focusing on, since the more notes a community has in the Changelog, the more that community gets sucked in.
The DQ also automatically gives me some spaced repetition without creating an anki card. I see it once when it’s created and again when I cycle through the queue (I spend 30 minutes at the end of each day just going through the queue). This has helped me in re-evaluating whether I need the note in the first place and/or if it should be split up further, merged, or just better quality. Two other benefits I’ve noticed are that my daily notes and books I’m reading/listening to pop up. Seeing daily notes from days to weeks ago is always a helpful reminder of priorities and progress, and the book notes help me to test my recollection and confront whether or not I’m actually getting anything out of them, as opposed to writing notes as I’m reading/listening (I think I prefer to take notes after the fact at this point).
I anticipate that this will get more useful as time goes on. I don’t have a whole lot of “communities” (clusters of notes), and I just have the one Queue. In the future with more Queues and many more notes, if, say, I was neglecting the topic of Psychology, then that community would drift further from the center. Conversely, that will also be a sign of how mature the community is. I should be able to visually distinguish whether different communities are mature or neglected by how densely connected that community’s Queue is (secondary queues would also add another layer of spaced repetition as an added bonus).
Right now it’s experimental, but at the very least it feels good to relieve the tension in the graph and I’m remembering more of my notes so it seems to be helping.