I'll push back a bit here, Mia, as you know I'm a big fan, subscriber, and I've got WATSON, Morning Light and Greenhouse already built into my way of working. So this is from inside the tent. ⛺️ 😀
The filter is where I'd take a different line. "Does this change what I'm doing this week?" works, but I think it optimises for the short-term. The value of a system like this is what compounds over time, and that needs a longer lens.
I do have a filter, but it asks this: Does this connect to one of my core pillars? Themes and questions that keep coming up in my work. I took the idea from Richard P Feynman, who apparently kept a list of twelve favourite problems and held every new piece of knowledge against them. Not "is this relevant this week" but "does this fit something I'm always working on?"
In practice, I'll see a YouTube video that resonates, so it goes through a skill that pulls the transcript and drops it into the vault. Audiobooks get transcribed and added. Longer PDF books or papers are pushed through NotebookLM. Everything gets its own folder with a thematic index, a log, and a memory file. YAML frontmatter and tags auto-align to the right pillar. The Smart Connections plugin then surfaces related fragments automatically whenever I open anything in Obsidian.
The vault getting bigger is the point because every day, it gets more useful.
Two features I find genuinely valuable: the mastermind. Pick Feynman, Niels Bohr, and Iain McGilchrist as dinner guests, for example, then ask them a group question, the system draws on what it's built up about each of them and provides perspectives.
Or run a research profile on someone before a meeting (web presence, interests, where they connect to my work) and generate questions worth asking.
It isn't a case of just dumping everything in. But you need your pillars first, not just a filter.
Please do push back, it actually helps because I'm definitely biased about my own stuff sometimes, and either way we're learning from each other ❤️
There was a reason for that Friday question. I mostly run this through Hermes, my agent, and he has a really good persistent memory that learns from every interaction we have, which is exactly why I explicitly named him in the piece. I haven't tested the workflow with other AIs.
If I make the question broader, like how things tie back to my brand, it's always going to find something relevant, and that just clutters the second brain for me. I keep separate second brains for different things, and for anything outside the business, where the stakes are more "relaxed" (inspiration, learning, et), I actually use a process a lot like yours. Business wise, this would SCARE me 😅 I've done it before and still ended up with a pile
And when I ask Hermes the Friday question, he isn't rigid about it. He won't go "drop this, useless for Friday." He'll say this might matter, maybe next month, definitely not now. It leaves the decision up to me still, I can make the decisions to drop or keep muuuch faster!
Hope this makes sense. I'm going to add a clearer disclaimer though, so thank you for bringing this up ❤️
Been using it since it came out! My husband was all in with OpenClaw and then we switched to Hermes and it's awesome.
I like that I can use open source models, and some of them I like even better than Claude! Current fav is Qwen. It's super fast, helps me with all my local files, Obsidian, etc. Memory is a HUGE thing.
I chat with it on Telegram and what I kind of dislike is obviously the chat interface.. like, I have to have a convo at a time, but I'm gonna change that soon, hopefully.
My husband has way more insights on it, really he just builds the infrastructure behind all this so I can enjoy it haha so check him out, he goes into all this in great detail! https://vibestacklab.substack.com/
OMG, I can't process all this right now as my brain needs a break. But in a case of hilarious overlap, I just published a piece about organizing all your 1/2 formed thoughts or existing podcast/webinar/sales calls, etc. I swear I didn't see this first.
Also, I'm also one who's never succeeded at the 2nd brain no matter how much I try.
I haven't built an agent to process my capture system. But I'm going to come back and read this with a clearer head. You're more technical than I.
And, as a digital hoarder, I don't know that I can use "Am I going to use this this week" as the metric. It's a good idea but "what if I need it one day?" I know that's ridiculous. This is good food for thought.
Also, just to add - you can adapt this to your process. You don't need the "use this week" metric, but in my experience, if the answer to that is no, it's probably gonna collect a loooot of dust haha.
Checking out your article today! ❤️ Also, I'm with you - I am definitely not technical haha. Kudos to my husband who is. But this is just a matter of, hey I have 10 resources I want to add to my library to "do later" or check them out later, can you have a first pass at them and tell me how they match my goals?
This is interesting. I use ai for a lot of things but trusting it to filter input was not one of them. But then again it makes sense because the decision can actually be as easy as figuring out if it is relevant to what you’re doing right now. Btw I also never made a system work for me. My notes are a mess.
Exactly what you said. This is not such a high stakes situation, plus you can always pass the check! :) my notes were a mess too. I used to add SO MANY resources to my library, I just had to do something about it.
This blew my mind. Does it mean that you don't read the "input", or only read it if it passes the First Brain test? Is this a double edged sword of more focus but less potential inspiration sources?
Hey Jen, oh no, I do read it! Not fully though, it's like.. I come across an article let's say, and the title is interesting, I could POTENTIALLY want to implement this too maybe today, tomorrow, at some point ...
So naturally, I add it as a resource in my library, but it just goes through an extra check! Of course, this is for the more practical, "I might need to do this" stuff. Inspos shouldn't really fit into a second brain YET, so for those I have another process ❤️
Good post, as always, Mia. I think the second brain thing works for academics, because we develop ideas over months and years, and we revisit and build on them. For someone trying to develop ideas in a more practical and efficient way, I can see how it's not so useful. I say that because one of the main reasons my obsidian vault is actually useful to me is that its inefficiency creates serendipitous links between ideas, and because I often end up returning to older ideas that get reactivated by a new project. It helps me by surfacing ideas I didn't know I was looking for. However, I also had to develop customized methods for exactly how my brain works that probably would not work as well for other people or other use cases. That's where you come in, because you've designed a helpful tool that *everyone* can put to use.
One way the second brain part works for me is that I have a second Obsidian vault for personal life stuff, and when I'm depressed or anxious, it surfaces encouraging memories or thoughts that my brain was too negatively biased at that moment to be able to think of on its own.
Anyway, I did want to ask: since you mentioned that you have a library of anecdotes to use as analogies in your posts, did you ever find some form of second brain useful for surfacing that kind of content or linking it across domains? You do a lot of really strong cross-domain thinking, so I wondered if any kind of second brain was ever useful in that area, or did they come up short in that area too?
I love the idea of a second vault for positive stuff too! I have an Obsidian vault for nice messages and encouraging notes as well, how cool is that??
And yes, I think you’re right about academia, though I also wonder if it can get overwhelming there too, just in a different way, when you have a pile of resources and don’t know what to do with them yet. The customized methods you mentioned feel like WAY more than a second brain to me, not the standard second-brain method.
As for the anecdote library, not quite in the classic second-brain sense either :) The references usually make it in after they’ve already passed the first-brain stage I talk about in the article. So maybe it’s a second brain, maybe not, but this would make for a very, very interesting debate haha ❤️
Yes, I guess a lot of it boils down to the definition of “second brain.” Because yeah, the common methods you mention in the article wouldn’t work for me either. One of my tricks is that every note gets a title that is a complete sentence and is basically the thesis statement or claim that the note involves. That way, when I look through notes, I pretty much know what I’m looking at and how it could relate to other notes before I open it again. Plus, my process leads me to see a lot of my notes over and over, so I kind of have a mental map of the most important ones. I guess this is all quite different from the second brains you describe. Maybe I should write about my system some time!..
I tried to make mental maps work but I couldn't! And I like the titles idea, I do the same with one sentence description or "read this for when you want to..."
I'll push back a bit here, Mia, as you know I'm a big fan, subscriber, and I've got WATSON, Morning Light and Greenhouse already built into my way of working. So this is from inside the tent. ⛺️ 😀
The filter is where I'd take a different line. "Does this change what I'm doing this week?" works, but I think it optimises for the short-term. The value of a system like this is what compounds over time, and that needs a longer lens.
I do have a filter, but it asks this: Does this connect to one of my core pillars? Themes and questions that keep coming up in my work. I took the idea from Richard P Feynman, who apparently kept a list of twelve favourite problems and held every new piece of knowledge against them. Not "is this relevant this week" but "does this fit something I'm always working on?"
In practice, I'll see a YouTube video that resonates, so it goes through a skill that pulls the transcript and drops it into the vault. Audiobooks get transcribed and added. Longer PDF books or papers are pushed through NotebookLM. Everything gets its own folder with a thematic index, a log, and a memory file. YAML frontmatter and tags auto-align to the right pillar. The Smart Connections plugin then surfaces related fragments automatically whenever I open anything in Obsidian.
The vault getting bigger is the point because every day, it gets more useful.
Two features I find genuinely valuable: the mastermind. Pick Feynman, Niels Bohr, and Iain McGilchrist as dinner guests, for example, then ask them a group question, the system draws on what it's built up about each of them and provides perspectives.
Or run a research profile on someone before a meeting (web presence, interests, where they connect to my work) and generate questions worth asking.
It isn't a case of just dumping everything in. But you need your pillars first, not just a filter.
Please do push back, it actually helps because I'm definitely biased about my own stuff sometimes, and either way we're learning from each other ❤️
There was a reason for that Friday question. I mostly run this through Hermes, my agent, and he has a really good persistent memory that learns from every interaction we have, which is exactly why I explicitly named him in the piece. I haven't tested the workflow with other AIs.
If I make the question broader, like how things tie back to my brand, it's always going to find something relevant, and that just clutters the second brain for me. I keep separate second brains for different things, and for anything outside the business, where the stakes are more "relaxed" (inspiration, learning, et), I actually use a process a lot like yours. Business wise, this would SCARE me 😅 I've done it before and still ended up with a pile
And when I ask Hermes the Friday question, he isn't rigid about it. He won't go "drop this, useless for Friday." He'll say this might matter, maybe next month, definitely not now. It leaves the decision up to me still, I can make the decisions to drop or keep muuuch faster!
Hope this makes sense. I'm going to add a clearer disclaimer though, so thank you for bringing this up ❤️
Absolutely. I completely get it. How long have you been using Hermes, and how have you found it? What are the positives and negatives? 🤔
Been using it since it came out! My husband was all in with OpenClaw and then we switched to Hermes and it's awesome.
I like that I can use open source models, and some of them I like even better than Claude! Current fav is Qwen. It's super fast, helps me with all my local files, Obsidian, etc. Memory is a HUGE thing.
I chat with it on Telegram and what I kind of dislike is obviously the chat interface.. like, I have to have a convo at a time, but I'm gonna change that soon, hopefully.
My husband has way more insights on it, really he just builds the infrastructure behind all this so I can enjoy it haha so check him out, he goes into all this in great detail! https://vibestacklab.substack.com/
Brilliant! I’ll check it out. Thank you so much for all you do, Mia! 🤖🚀
Thank YOU for challenging me! 🙏 Really appreciate it.
OMG, I can't process all this right now as my brain needs a break. But in a case of hilarious overlap, I just published a piece about organizing all your 1/2 formed thoughts or existing podcast/webinar/sales calls, etc. I swear I didn't see this first.
Also, I'm also one who's never succeeded at the 2nd brain no matter how much I try.
I haven't built an agent to process my capture system. But I'm going to come back and read this with a clearer head. You're more technical than I.
And, as a digital hoarder, I don't know that I can use "Am I going to use this this week" as the metric. It's a good idea but "what if I need it one day?" I know that's ridiculous. This is good food for thought.
Also, just to add - you can adapt this to your process. You don't need the "use this week" metric, but in my experience, if the answer to that is no, it's probably gonna collect a loooot of dust haha.
It’s a good metric. I’m going to try it. 😂
Checking out your article today! ❤️ Also, I'm with you - I am definitely not technical haha. Kudos to my husband who is. But this is just a matter of, hey I have 10 resources I want to add to my library to "do later" or check them out later, can you have a first pass at them and tell me how they match my goals?
“Do they match my goals” is a good one. I have started asking AI is something matches my goals or I’m going off script.
And fwiw, I still think you have a technical brain, you frequently write about AI tech things that are five steps beyond my understanding. 😂
Hahah thank you, running to tell my husband this! 🤣🤣
This is interesting. I use ai for a lot of things but trusting it to filter input was not one of them. But then again it makes sense because the decision can actually be as easy as figuring out if it is relevant to what you’re doing right now. Btw I also never made a system work for me. My notes are a mess.
Exactly what you said. This is not such a high stakes situation, plus you can always pass the check! :) my notes were a mess too. I used to add SO MANY resources to my library, I just had to do something about it.
This blew my mind. Does it mean that you don't read the "input", or only read it if it passes the First Brain test? Is this a double edged sword of more focus but less potential inspiration sources?
Hey Jen, oh no, I do read it! Not fully though, it's like.. I come across an article let's say, and the title is interesting, I could POTENTIALLY want to implement this too maybe today, tomorrow, at some point ...
So naturally, I add it as a resource in my library, but it just goes through an extra check! Of course, this is for the more practical, "I might need to do this" stuff. Inspos shouldn't really fit into a second brain YET, so for those I have another process ❤️
Makes total sense. Probably worth me being a bit more thoughtful about separating resources vs inspo too! Thanks for the guide :)
Good post, as always, Mia. I think the second brain thing works for academics, because we develop ideas over months and years, and we revisit and build on them. For someone trying to develop ideas in a more practical and efficient way, I can see how it's not so useful. I say that because one of the main reasons my obsidian vault is actually useful to me is that its inefficiency creates serendipitous links between ideas, and because I often end up returning to older ideas that get reactivated by a new project. It helps me by surfacing ideas I didn't know I was looking for. However, I also had to develop customized methods for exactly how my brain works that probably would not work as well for other people or other use cases. That's where you come in, because you've designed a helpful tool that *everyone* can put to use.
One way the second brain part works for me is that I have a second Obsidian vault for personal life stuff, and when I'm depressed or anxious, it surfaces encouraging memories or thoughts that my brain was too negatively biased at that moment to be able to think of on its own.
Anyway, I did want to ask: since you mentioned that you have a library of anecdotes to use as analogies in your posts, did you ever find some form of second brain useful for surfacing that kind of content or linking it across domains? You do a lot of really strong cross-domain thinking, so I wondered if any kind of second brain was ever useful in that area, or did they come up short in that area too?
I love the idea of a second vault for positive stuff too! I have an Obsidian vault for nice messages and encouraging notes as well, how cool is that??
And yes, I think you’re right about academia, though I also wonder if it can get overwhelming there too, just in a different way, when you have a pile of resources and don’t know what to do with them yet. The customized methods you mentioned feel like WAY more than a second brain to me, not the standard second-brain method.
As for the anecdote library, not quite in the classic second-brain sense either :) The references usually make it in after they’ve already passed the first-brain stage I talk about in the article. So maybe it’s a second brain, maybe not, but this would make for a very, very interesting debate haha ❤️
Yes, I guess a lot of it boils down to the definition of “second brain.” Because yeah, the common methods you mention in the article wouldn’t work for me either. One of my tricks is that every note gets a title that is a complete sentence and is basically the thesis statement or claim that the note involves. That way, when I look through notes, I pretty much know what I’m looking at and how it could relate to other notes before I open it again. Plus, my process leads me to see a lot of my notes over and over, so I kind of have a mental map of the most important ones. I guess this is all quite different from the second brains you describe. Maybe I should write about my system some time!..
I tried to make mental maps work but I couldn't! And I like the titles idea, I do the same with one sentence description or "read this for when you want to..."
I'd love a post on your process! :)