Why Your Second Brain Isn’t Producing Insights — The Knowledge Graph Fix! — Transcript
Full transcript
- 0:00In this video, I want to talk about the second brain and why your second brain needs
- 0:05a knowledge graph. A second brain is a system that allows people to capture, organize, store,
- 0:12and retrieve information using an external tool like a knowledge management system, for example,
- 0:17Obsidian that you see right now on your screen. However, the problem with most approaches is
- 0:22that they focus too much on classifying ideas into different folders. And while this can be useful,
- 0:28in fact our human brains do not work in this way. They work based on associations on reasoning
- 0:36logics and rules. So this is why knowledge graphs are super useful. And if we can visualize our
- 0:41ideas like I have here with my obsidian folder as a knowledge graph and then apply graph theory
- 0:47metrics that allow us to identify what are the most important topics or clusters of ideas that
- 0:52we have inside our vault, how they're connected to one another, what are the gaps between them,
- 0:58what are the different concepts at the periphery of our thinking that could benefit from being
- 1:03developed further. You know this gives us much more capabilities especially if we talk about
- 1:08research or creative work. On top of that if you have a network structure visualized to you
- 1:14as a graph and represented as a graph then you can also identify diversity and this is what
- 1:19InfraNodus does here. I have the InfraNodus graph view extension here in Obsidian and I analyze the
- 1:24whole folder of business ideas that I have and then it shows me okay that I'm too focused here.
- 1:30You see I have the diversity measure and so it gives me an advice that I could diversify
- 1:35my ideas. So I could benefit from thinking about the clusters and ideas that are at the periphery
- 1:41of the graph and if the graph was too dispersed then it would encourage me to connect those ideas
- 1:46together better. So as you can see this structural approach and graph representation allows you to
- 1:52not only generate interesting ideas and find gaps but also identify how you can be thinking in order
- 1:58to harmonize your knowledge. And in this video I'm going to show you how you can do that and how
- 2:04you can apply this approach to your particular use case. Because what most people do, you know, when
- 2:10they have a folder like I have here for instance with business ideas that I've been writing down,
- 2:14sometimes using voice recorders, sometimes using text and I visualize it as a graph, right? Like I
- 2:20can see the main topics inside. If I click here in the InfraNodus extension, I can see what are
- 2:25the main topical clusters in this idea. So I get a really nice overview, but I can control how I will
- 2:31be thinking around this graph. Because if I just ask the AI to give me a summary of all the ideas
- 2:37in this folder, AI will do what it does best. It will generate a really predictable outcome that
- 2:43will be quite generic. Of course, I can do some prompt gymnastics in order to get it extract some
- 2:49hidden insights, but it's going to be hard and I have uh no control over how it does it. Even if I
- 2:55ask it to explain how it's thinking, it's just going to actually take me a lot of time. Here
- 3:00with the graph visualizing all the ideas here, I can directly see okay, I'm focused on workflow
- 3:07dynamics, on interactive legacy, on content generation, AI tools, contextual graphs. So if I
- 3:13know this whole context, I already understand what I'm talking about and what ideas tend to recur on
- 3:21the permanent basis. So for example, I understand that okay, I should really look into this uh idea
- 3:26of creating repository of different workflows that I've been thinking about for a while. Also here
- 3:33I have something about market insights and so on. And I can click on each of those clusters and then
- 3:38ask in front to generate a summary of what I was thinking about in relation to each cluster. And
- 3:44then I can also look at the actual statements and reiterate through them to understand where exactly
- 3:49I was writing uh things that relate to market insight. So it's really useful to get an overview
- 3:54and this is kind of like the first uh advantage of using the graph that you can get a nice overview
- 3:59but you control what you're looking at and you can choose exactly what you will be selecting and
- 4:04what you will be exploring not relying on AI to make this choice for you and produce some generic
- 4:10result and maybe forget something that is quite important to you. And then the second advantage
- 4:15of using the graph and having this structured representation of ideas is that you can click
- 4:20on the gaps here and sometimes you can actually use the 2D mode here. So you can see the gaps a
- 4:25little bit better and then you iterate through them and then you will see that it finds some
- 4:30gaps between the clusters that are furthest apart from each other in the graph. It means
- 4:35that you're thinking about these topics like I'm thinking about AI tools and uh I have this another
- 4:39idea of digital legacy but I'm not connecting them together. So for me a nice idea would be
- 4:45how I can connect those two clusters together. And if you use an analogy with social networks
- 4:50it's like you have two groups that kind of like to hang out together but never connect and maybe
- 4:55they would benefit from being introduced to each other. So maybe these two clusters of ideas could
- 5:00also benefit from being introduced. And there's a whole theory that shows that really great ideas
- 5:05lie exactly at those structural gaps between different topics. So I can identify those topics
- 5:10using the InfraNodus graph and then generate an interesting research question that would propose
- 5:15me how I can bridge those ideas together and then I can reiterate through them and generate some
- 5:21really interesting stuff that will get me thinking in this direction and help me connect my ideas
- 5:26further. One other thing I mentioned is that you can switch to the concepts view and then you can
- 5:32increase the number of concepts you see here. And then what I like to do is to also use the graph to
- 5:37identify what I have at the periphery of my ideas like some smaller stuff that will take me beyond
- 5:43this bubble that I'm in when I'm thinking about a certain topic. Right? So I can select those and
- 5:49also generate a business idea for example based on those clusters and that will help me take my
- 5:55ideas further beyond the limits of that particular knowledge graph. So is also a really interesting
- 6:01way to ideulate and finally like I showed you at the beginning you have this diversity measure
- 6:07that shows you what the structure of your network is. So it kind of gives you an advice that okay
- 6:11Demetri you're too focused on certain ideas here in this case you know I'm talking a lot about AI
- 6:17and marketing and if I look at the topics you know I can have like a more highle overview and maybe
- 6:23you would benefit from actually diversifying your ideas. So that means focusing on smaller clusters,
- 6:29looking beyond the periphery and kind of seeing how you can uh make them a little bit different
- 6:34from each other or maybe even erasing some of the central ideas and focusing on the ones that
- 6:40are less obvious. So this would be another really useful way to apply graphs to this thinking. So as
- 6:46you see, we're not really imposing categories on top of our knowledge, but we allow them to
- 6:52organically emerge from the bottom up. and we use a network science and graphs in order to make sure
- 6:59that they will be fully related to the content that we're studying. So that I find is very very
- 7:06interesting. And one last thing, if you want to see how that works, uh the easiest way to do it
- 7:11is to install the InfraNodus graph view plug-in to Obsidian if you're already using Obsidian. If
- 7:18you don't have Obsidian, you can also go inside InfraNodus and create a new graph and import from
- 7:23multiple sources that we have. So for example, if you have Evernote, you can import your Evernote
- 7:28notes here and then connect it to your vault and then you see I have the connection here. So I can
- 7:34import my notes from any folder and then once you do that you will have it visualized and then you
- 7:40can analyze it also using the graph inside InfraNodus. But if you use obsidian you can
- 7:46um organize your ideas into folders. So here I'm using categories. Okay. But it's because I want
- 7:51to have things slightly organized. But inside this organization I would like some ideas to
- 7:57emerge. So this is why I select uh this folder once I have the uh graph view plug-in installed
- 8:03and then I right click on that folder and I say okay open InfraNodus graph. And what it does is
- 8:07that it takes all those different notes all those different ideas I've been having. Some
- 8:12of them are more developed, some less and then it visualizes them as a graph and shows me okay what
- 8:16are the main topics inside how they connect to one another and then you apply this workflow that
- 8:20I demonstrated earlier to see uh how you could develop them further and you can use the built-in
- 8:27AI to summarize ideulate and also generate questions in relation to them. So, I hope you find
- 8:33it useful and I hope it will enhance your approach to building your second brain. And like I promised
- 8:39in the end of the video, I will link above to the video of Nate Jones where he explains how you can
- 8:45build a really nice stack using n8n and various tools to kind of like create your own system where
- 8:52you can capture ideas through Slack and then use categorization to split them into different groups
- 8:58and then uh retrieve them using notifications. I really like his approach. So I recommend you
- 9:03to check him out and also give a shout out from Dimmitri from InfraNodus as well. All right. So
- 9:09if you have any questions, let me know. I'll be happy to help you. And also if you have any
- 9:14comments or feedback, I will be happy to engage in the comments section below the video. Thanks.
About this transcript
This page contains the full transcript of Why Your Second Brain Isn’t Producing Insights — The Knowledge Graph Fix! by Nodus Labs, generated from the public captions YouTube serves with the video. The transcript has 1,792 words across 100 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.
What you can do with it
Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.
Free YouTube transcript tool
YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.