In my lists of Pros and Cons for sticking with the Google Pixel ecosystem, one of the Cons is the fact that Google definitely does not want you to have this valuable capability. If you stop looking things up, then you won't be looking at their search engine, their ads, and their recommendation algorithms. Every platform wants you to do that. It's why bookmarks in Google Chrome lack useful features like tagging. It's one of the reasons why so many vendors try to lock your data inside their walled gardens. Apple is well known for walled gardens, but for the most part, you can be sure they will let you change your default search engine without much hassle. They won't care so much if you want to use something like SearXNG to prioritize your knowledgebase first, however, the App Store, Apple Music, and Apple TV are the same story as Google - any attempts to influence the search results in your favor will be actively fought against.
Thank you for your insight and comment. I’ve witnessed this behavior but never thought to question it until now. It’s amazing how simple and devious it can be.
Extend that to marketplaces too, all their search UIs have dark patterns forcing you to see their “recommendations” instead of being able to manipulate the results like you want.
Also love using Obsidian for this! Small suggestion, use the `aliases` property for alternative titles, I usually use them for a title that means the same thing but uses different keywords. Makes it easier to search for a note.
Although usually a bottom-up approach using automatically updating `Map of Content` notes (Bases) work well for me for finding content.
In most cases, I just add a blog post for such things.
For example, Syncthing on Debian notes [1] or using Spleeter AI to remove background sound from a long audio track [2]. This is why I switched back from static site to a Wordpress-like site [3], so that I can quickly publish notes from my phone.
I've used Zim for this over the last 20 years. Its a mess but with search I can find what I'm looking for if I was good enough. For especially tricky trouble shooting I'll put in each theory and strike it out so at least if I stop documenting I'll see what didn't work before I gave up or succeeded and quit documenting.
I find one possible answer to the question “How to make yourself actually do it” is to start by getting into the routine of keeping an engineering notebook - if you are already in the habit of jotting down stream-of-consciousness notes on whatever you are working on at a given time, then Obsidian’s feature to “extract highlighted text into a new note” feature makes it blisteringly easy to file away things you are likely to want to repeat in the future.
I’m on the other side of this - I’d like it to happen passively and in an automated fashion. Right now I’m playing around with porting concepts from zettelkasten (card based handwritten knowledge systems) to openclaw memory.
Rather than just coalescing to markdown files, the memory-zet plugin looks for actionable durable information and files it inside the existing zettelkasten system with embeddings - a quick no-LLM step (well 300m parameter query embed, it’s fast) is run against incoming chats or as a tool - this returns cards (zettels).
Zettels are somewhat unique in that the original methodology included a post-writing categorization and linking step - I have the system doing this as well. Result - cards can give you a (possibly cyclic) directed graph of connectivity. I built it for ‘centaur’ mode, so I can edit, link, unlink, move, etc through a nice little web interface.
The auto links are not the same quality I would make. But they are genuinely useful; upshot is for anything incoming, the LLM can see information directly about the query (if we have it), stuff that relates whether or not it embeds similarly, and can follow up links if they look promising with a fast tool call.
I made this memory system my daily driver yesterday; so far it is a significant improvement over the core memory extension (write to markdown files, don’t worry about compaction bro, it will be fine)!
It’s already building out people and organizational card bases for things that come in via email and whatsapp - this is a dream, basically. I think it will scale over time - but it’s at least scaling nicely over a few days of work right now.
I use a couple of slim A7 notebooks, one is like a diary, it gets stored when it's full. The other is a hard copy of my memory of how to do things like squishing PDFs. I rarely have to look at things twice as the act of writing it down gives me enough context to remember it. But it's invaluable when I need it.
Tried Evernote and tagging and so on and it turns out cataloging stuff is hard, and the lazy recourse is to over-tag, and then I end up doing a brute force search.
In my lists of Pros and Cons for sticking with the Google Pixel ecosystem, one of the Cons is the fact that Google definitely does not want you to have this valuable capability. If you stop looking things up, then you won't be looking at their search engine, their ads, and their recommendation algorithms. Every platform wants you to do that. It's why bookmarks in Google Chrome lack useful features like tagging. It's one of the reasons why so many vendors try to lock your data inside their walled gardens. Apple is well known for walled gardens, but for the most part, you can be sure they will let you change your default search engine without much hassle. They won't care so much if you want to use something like SearXNG to prioritize your knowledgebase first, however, the App Store, Apple Music, and Apple TV are the same story as Google - any attempts to influence the search results in your favor will be actively fought against.
Thank you for your insight and comment. I’ve witnessed this behavior but never thought to question it until now. It’s amazing how simple and devious it can be.
Extend that to marketplaces too, all their search UIs have dark patterns forcing you to see their “recommendations” instead of being able to manipulate the results like you want.
Also love using Obsidian for this! Small suggestion, use the `aliases` property for alternative titles, I usually use them for a title that means the same thing but uses different keywords. Makes it easier to search for a note.
Although usually a bottom-up approach using automatically updating `Map of Content` notes (Bases) work well for me for finding content.
In most cases, I just add a blog post for such things.
For example, Syncthing on Debian notes [1] or using Spleeter AI to remove background sound from a long audio track [2]. This is why I switched back from static site to a Wordpress-like site [3], so that I can quickly publish notes from my phone.
[1]: https://huijzer.xyz/posts/149/setup-a-syncthing-service-on-d...
[2]: https://huijzer.xyz/posts/146/installing-and-running-spleete...
[3]: https://github.com/rikhuijzer/fx
I've used Zim for this over the last 20 years. Its a mess but with search I can find what I'm looking for if I was good enough. For especially tricky trouble shooting I'll put in each theory and strike it out so at least if I stop documenting I'll see what didn't work before I gave up or succeeded and quit documenting.
I find one possible answer to the question “How to make yourself actually do it” is to start by getting into the routine of keeping an engineering notebook - if you are already in the habit of jotting down stream-of-consciousness notes on whatever you are working on at a given time, then Obsidian’s feature to “extract highlighted text into a new note” feature makes it blisteringly easy to file away things you are likely to want to repeat in the future.
I’m on the other side of this - I’d like it to happen passively and in an automated fashion. Right now I’m playing around with porting concepts from zettelkasten (card based handwritten knowledge systems) to openclaw memory.
Rather than just coalescing to markdown files, the memory-zet plugin looks for actionable durable information and files it inside the existing zettelkasten system with embeddings - a quick no-LLM step (well 300m parameter query embed, it’s fast) is run against incoming chats or as a tool - this returns cards (zettels).
Zettels are somewhat unique in that the original methodology included a post-writing categorization and linking step - I have the system doing this as well. Result - cards can give you a (possibly cyclic) directed graph of connectivity. I built it for ‘centaur’ mode, so I can edit, link, unlink, move, etc through a nice little web interface.
The auto links are not the same quality I would make. But they are genuinely useful; upshot is for anything incoming, the LLM can see information directly about the query (if we have it), stuff that relates whether or not it embeds similarly, and can follow up links if they look promising with a fast tool call.
I made this memory system my daily driver yesterday; so far it is a significant improvement over the core memory extension (write to markdown files, don’t worry about compaction bro, it will be fine)!
It’s already building out people and organizational card bases for things that come in via email and whatsapp - this is a dream, basically. I think it will scale over time - but it’s at least scaling nicely over a few days of work right now.
This sounds fascinating, if you ever write more about it I'd love to read it!
I am using a combination of Tomboy (desktop), Tomdroid and Markor (mobile) to record info i need later. Simple and effective.
I use a couple of slim A7 notebooks, one is like a diary, it gets stored when it's full. The other is a hard copy of my memory of how to do things like squishing PDFs. I rarely have to look at things twice as the act of writing it down gives me enough context to remember it. But it's invaluable when I need it.
Tried Evernote and tagging and so on and it turns out cataloging stuff is hard, and the lazy recourse is to over-tag, and then I end up doing a brute force search.