You never have to tag a note for Second Brain to know it is related to something else
Most knowledge tools put the connection work on you. Tag this. Link that. File it in the right folder.
Second Brain v2 builds the graph automatically.
When you store a memory, the system extracts the people, projects, decisions, and concepts mentioned. It then scans the existing graph for related entries and creates links where it finds meaningful overlap.
You do not have to decide what is related to what. By the time you need to recall something, the graph has already traced the connections.
A few examples of what this surfaces in practice:
You store a note about a product decision. The system links it to a previous discussion about the same feature, a person mentioned in both, and an older decision that this one partly reverses.
You store a meeting summary. It links to the project it references, the last meeting about the same topic, and an open question that this meeting answered.
The links are not perfect. Some are too broad. Some miss things. You can remove or correct them. But the default behavior is that the graph grows on its own as you use it, and recall gets better the more you add.
After a day of using Second Brain, what types of automatic links surprised you most?
Replies
The most useful links would probably be the ones that show when a decision changed over time.
Not just this note is related to that project, but this newer note reverses or narrows an older assumption. That’s where manual tagging usually fails, because you don’t know which old context will matter until later.
I’d be curious how you rank links so the graph doesn’t become noisy as it grows.
I really like the philosophy here. Most tools make me think about structure before I can think about ideas. This flips that around. Have you found any patterns in the links users end up keeping versus deleting?
I can already imagine forgetting a conversation, then randomly discovering it again through an unexpected connection. Those little moments are what make products memorable. Have you had any surprising examples yourself?
This feels much closer to how people actually remember things. We don't think in folders we think in relationships. Does the graph prioritize recent memories, or are all connections weighted equally?
One thing I'm wondering if I have thousands of notes, does the quality of linking improve because there's more context or does it become harder to keep the graph relevant? Really interesting concept.
The fact that you admit the links arent perfect actually gives me more confidence. I would rather have editable suggestions than pretend AI is always right. Are corrections used to improve future linking?