Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Skill for your agent to visualize your gbrain and Obsidian

github.com

Post reply on HN
Print view View original

11–18 of 18 posts

Re: Show HN: Skill for your agent to visualize your gbrain and Obsidian

#11

Post by purpleflashing »

ElFitz wrote: post #10
purpleflashing wrote: post #2

I always wondered what people use these maps for. Does it help you find things faster in your notes?

The graph visualisation itself isn’t much use in my experience. But being able to tie related notes together, and see at the bottom of one which other notes reference it is interesting. Even more now that a LLM can take care of the actual tending and pruning.

Ah, I see, sort of like figuring out the boundaries of your knowledge base and seeing if you have missed any connections between concepts? I suppose it might be useful for learning/ideation. I should try something like that — it could be an interesting synthesis/writing exercise to try to connect concepts that are far removed in your own mental model.

Thank you for sharing!

Re: Show HN: Skill for your agent to visualize your gbrain and Obsidian

#12

Post by v_ignatyev »

purpleflashing wrote: post #7
v_ignatyev wrote: post #4

Earlier quoted context omitted.

Mostly to understand the structure of your knowledge base. To search across the KB it is more convenient to use Obsidian/gbrain itself. Suppose you've been collecting thousands of notes from meetings, feedback from colleagues or code-reviews, "insights" and tips from you wife over the years. Aren't you curious what this KB contains and how it has evolved over time? So, shortly it is a debug tool for your second brain…

> Aren't you curious what this KB contains and how it has evolved over time? Can you give a more specific example of what you have found in this data? I already know what KB contains and how it evolved — I was the one who put things into it, after all. Just to clarify — I am not being snarky or criticizing your project, I am genuinely curious. I like data visualization. P.s. Also, as an unrelated tangent, please feel…

It's more about "meta" data over the data itself. What seemed for me good enough (notes had hierarchy only around some topics of interest) ended up in being almost useless for AI agents. When I've seen for the first time how my notes were really structured (only two clouds and endless non-linked grid of random ideas and observations) I'd got an idea why it was useless for AI. Then I tried to remove some garbage myself, tried to do this with the help of agent, but ended up making a KB from scratch and cherry-pick notes from my old knowledge base. Now it looks like 4 separated clouds and only few random disconnected notes out there.

Last week I changed my job and this KB got quickly saturated with some information about company, contracts, people in company etc. This helped me to start off quickly on the new site, but the downside was that I concerned about what sorts of info had been stored in this KB. This Skill has helped me (along with basically `grep`) to ensure that there is no leakage and work is fully separated from anything else.

I put "wife's insights" into quotation marks, because, honestly, most of things contributed by my wife into this KB were really obvious or boring things which don't empower this KB in any way. She knows what it means ;)

Re: Show HN: Skill for your agent to visualize your gbrain and Obsidian

#13

Post by v_ignatyev »

RHab wrote: post #9

I wanted something like this for my knowledge wiki, nexidion: https://github.com/HabermannR/Nexidion Will look how hard it is to combine next week, thanks!

I am looking forward to getting feedback from you. I can't give any guarantees though))

Re: Show HN: Skill for your agent to visualize your gbrain and Obsidian

#16

Post by awllau »

Do people even use the visual/graph views at all or is it just optics? I'm thinking about it from the perspective of using LLMs to search your own database, and I imagine it reaches a scale where it's not feasible or effective for a human to be manually searching.

Re: Show HN: Skill for your agent to visualize your gbrain and Obsidian

#17

Post by ElFitz »

purpleflashing wrote: post #11
ElFitz wrote: post #10

Earlier quoted context omitted.

The graph visualisation itself isn’t much use in my experience. But being able to tie related notes together, and see at the bottom of one which other notes reference it is interesting. Even more now that a LLM can take care of the actual tending and pruning.

Ah, I see, sort of like figuring out the boundaries of your knowledge base and seeing if you have missed any connections between concepts? I suppose it might be useful for learning/ideation. I should try something like that — it could be an interesting synthesis/writing exercise to try to connect concepts that are far removed in your own mental model. Thank you for sharing!

> Ah, I see, sort of like figuring out the boundaries of your knowledge base and seeing if you have missed any connections between concepts?

Yeah. And also resurface relevant old notes, ideas, and web clippings I have forgotten. Two recent example[^0][^1].

> it could be an interesting synthesis/writing exercise to try to connect concepts that are far removed in your own mental model.

As well as different takes on or applications for those concepts, same for sources.

[0]: https://imgur.com/a/jhxUV9i

[1]: https://imgur.com/a/gmaq9Yw

Post reply on HN