Note-taking apps are designed for storage, not insight – can AI change that?
theverge.com
theverge.com
The issue with the daily note is that it doesn’t sort by content. That’s why I just constantly make notes based on topic of discussion. Whenever I have an idea, I take a voice memo on my phone. Then I send that voice memo to whisper AI for autotranscription, and a little script on my server formats it for Obsidian MD. I used to try to write from scratch. Now I realize that writing from scratch is a terrible method of writing.
I mostly use Obsidian for taking notes during D&D (arguably its killer app), but I think "session" notes are a good enough analogue to "daily notes".
To that end, these notes serve as an excuse to force you to list your individual notes in chronological order (which makes it easy to trigger further memories/details that came next), which then prompts you to spin out any specific-enough information into atomic notes, adding backlinks which provide context and organic categorization that might not otherwise be obvious.
My point was that I think the authors problem is the daily note rather than the note taking app.
Rather the benefit seems to mostly stem from freeing up mental space by removing whatever the thought was that you want to cling to, and putting it somewhere else that "feels" permanent. That feeling is important because it allows your brain to drop it, rather than feel like you have to continue to think about it out of fear of "forgetting".
You have to understand where that differing POV connects with other POVs. The bridge between the two is what matters here; not the content.
Most of the work is in thinking through things for yourself, organizing information, and synthesizing some outcome. Internet browsing is like junk food, and doesn't do a good job substituting for that.
Also, you mention "thoughts from others are the only way to adjust your own bias." I definitely agree with that, but I want to get thoughts from people who I know and trust. If you adjust your biases based on the cacophony of the internet it's much more likely that you'll get swept up in other people's biases instead
An online resource using a lot of context specific phrases, that l'll need to spend time to define and understand individually, just to build up that greater understanding of a topic is a bit of a time sink. It's helpful to have a note on something in one's own voice that helps you return to that frame of mind.
This type of notetaking has been around forever. Ten years ago I was using TiddlyWiki exactly that way. Click the button to create a new journal and organize them by date.
> The second is known to note-taking nerds as “bidirectional linking.”
Also strange to attribute this to Roam. It was around long before Roam. TiddlyWiki had it forever. I never saw much use to it, but it was there, even if the author didn't know about it.
> In short: it is probably a mistake, in the end, to ask software to improve our thinking.
It's tough to draw that conclusion based on the arguments up to that point. Software is really helpful. I could use a plain text editor, but quality search, hyperlinks between notes, automatic naming of notes, ability to make task lists with checkable boxes, etc. is a clear win for me.
You can do it with index cards or you can use software to practice the method and grow your note collection. I now prefer Zettlr (https://www.zettlr.com) after using Joplin (https://joplinapp.org), which are both FOSS.
One of the core strategies of the Zettlekasten Method is to link notes to each other. That's how knowledge grows: connections and synthesis (internalization/application of the connections)
Here's a 3-year-old video on the method that serves as a good primer: https://www.youtube.com/watch?v=wFZHuWLA09M
Maybe somebody might find it useful:
Everything is just files in a directory, so I use Syncthing to have access to it everywhere.
It's also got a thing that lets you associate files with a given note, but the tooling around that isn't very good yet. I've been slowly working on a client for the Acme text editor, and I've also wanted to make a proper Go GUI client.
(man I need to re-do the CLI with some library that does built-in help, because some commands like `zk orphans` [which finds notes that have no parents] aren't documented anywhere)
I make liberal use of tags and find that this basically works great - 50 to 100 tags seems to be all I could ever need for both personal and work stuff, which number is quite scalable. However, I had always intended to progressively adopt Zettlekasten, which does sort of seem like the gold standard.
Unfortunately, the way Obsidian works is that you can only really conveniently link to the filenames of other notes, rather than directly to a heading therein. That UX makes it a non-starter for Zettlekasten, for me.
Thought I’d leave this here in case it’s of use to anyone, and would be very interested to hear from anyone who’s landed on a good Zettlekasten setup in Obsidian.
Zettlr looks excellent, and since they both run on a plain folder of markdown files, I might even trial running both together for a bit.
It's barely mentioned and I don't even remember what post I saw this in on the Obsidian forum.
[1] https://forum.obsidian.md/t/use-h1-or-yaml-property-title-in...
I wrote a piece about it: https://www.dsebastien.net/2022-05-01-zettelkasten-method/
Then, there's real value to taking notes that for me is most pronounced when I write them with a real pen and paper. I may never actually go back to these, but the act of writing them down solidifies things for me. Things like WWDC sessions, online courses. The value is the act of deliberately recording. There's a kind of brain/physical connection that happens.
Ultimately, intentionality matters. There are some projects where the note-taking itself is the point. The notes are the end product; you're producing something like a catalogue of thoughts. Hypercard.
But, it's easy to get lost in that if what you're really trying to achieve is 'over there' somewhere. It can be a way of avoiding the work entirely.
Yes, this is me. I've gone through all sorts of ways to "improve" my note-taking, including various note-taking apps. But they all actually reduce the value of note-taking for me, because the valuable part of note-taking is that act of taking notes itself. If I'm not writing them long-form by hand, as opposed to tapping or typing, that value is eliminated.
An exception to all of this is when attending lectures or presentations -- in that setting, taking notes always reduces my comprehension of what it being presented, because I can't take notes and listen at the same time. For me, it's similar to photography -- if I'm taking photographs of a place or event that I'm at, the act of taking the photographs removes me from that place or event and I can no longer fully take it in.
Did that in uni when cramming for some exams. Writing (a good part of the course) by hand in the days prior to the exam helped a lot.
I had a few teachers who would twist this a bit by saying "you can bring a single note card" or similar. Once you've had to figure out how to stuff a calc-exam's worth of help into a 3x5 card, you'll damn sure know more than when you started!
I'll often make myself little docs or cheat sheets for some onerous process at work, and again - getting to the point where I have a doc that's foolproof even on my stupidest mornings does a lot to make sure I know the material well.
But quite often, you get replies in the comments saying that _all_ note-taking and organization is a procrastinating waste of time. ("Why can't everyone just memorize everything like I do???") Which is wrong too.
Notes are just a tool for knowledge work. They are necessary but not sufficient. If you're not writing things down SOMEWHERE every day, you are leaving a LOT of mental capacity on the table. All the while setting up Future You to relearn the same things over and over again.
(Final tangent: apprentice woodworkers and blacksmiths would craft their own basic tools when just getting started, I think most apprentice software devs should do the same.)
I rarely ever attended class in my major. I was notorious for it. I had 3 jobs. I would ride the bus an hour and a half to school and spend most days sleeping in the student union. I'd go xerox her notes right before the test, study them a few times during lunch breaks and afterward at work, and then go take the exams.
She took such good notes that I almost always had higher grades than she did. Studying notes is more important than taking them, but having great notes means As if you understand what you're reading.
So far its been good and an improvement, thanks to md im taking great notes fast and efficiently and im putting more thought into connections. But all the videos and posts claiming its changed their lives or something is a bit much. I cant imagine it being that much more useful outside of school.
On a side note, any other useful programs/techniques feel free to share.
I’m also experimenting with Asana to track class assignments, but I’m currently trying to get them to let me buy “only” one seat of Business level, since the “workload” tool seems immensely valuable to keep my work spread out evenly through the semester.
Would you have clicked on the video if the title was "My second brain has been kinda useful, I guess"?
I feel approaching journaling with a goal in mind, instrumentally, may undermine the process. Better to just till the mental soil and see what happens. Like going on a hike or a bike ride.
edit: healthy gamer is the channel, and I really enjoy when he talks the science behind some of these topics
Then, I switched to spending a little bit of time organizing my notes before taking lots of them. This changed the game for me considerably.
It definitely doesn't make me smarter but it does keep me more organized. To the point where I'm not in a conversation remembering something I missed from earlier. This does make me more effective in those conversations and more able to focus on the people I'm conversing with.
And that can definitely make me _seem_ smarter...
It's not true that note-taking and well designed apps to do so don't make us smarter. Being smart means "having or showing a high degree of mental ability" (point 1 of 7 in https://www.merriam-webster.com/dictionary/smart). Applications which support our information management increase our mental abilities.
Since I take notes of every meeting, I still know days or even years later what was decided and for what reason, and I can check the arguments afterwards to see if they are still valid and, if necessary, make sure that they are returned to if circumstances change. If I don't take any notes or only keyword notes, I can't do that because after two hours you only remember a fraction of the meeting, and after a day full of meetings you hardly remember any details. I've done big enough projects for long enough and participated in meetings of 2 with up to 50 people in a room, and have been able to observe very well how what kind of notes affected the aftermath of the meeting on individuals.
"I define intelligence an innate ability to juggle information well, which is independent of memory. If you're trying to think, but have forgotten some key information and not have taken notes, you're not less intelligent, only less informed; therefore, note taking cannot make you more intelligent, only better informed."
Yawn ...
> I define intelligence an innate ability to juggle information well, which is independent of memory ...
Well, if people start using their own definitions, then we should not wonder that they come to unusual conclusions.
Like, my template record for a company has slots for the C-suite that link to records for persons, and slots for parent and child companies. Because of that, whenever I add a new company to my database, I'm prompted to add that information. I don't have to, but what's important is that I don't have to worry about forgetting to deliberately make that choice. And in the future I can count on being reminded of what I omitted.
It can already happen on Twitter that people complete each other's thoughts but the Twitter timeline is not the medium to publish half-finished thoughts that could tarnish one's reputation.
Still, there is some joined thinking. HN shows that it's fun to pool notes on poplar topics. The difficulty lies in managing the attention on notes. Popularity voting doesn't work to navigate note collections.
The headline is about our smartness, but it doesn't mean us as a group. There is no joined thinking in his note-taking apps.
He is right when he talks about his isolated thinking:
>But the original promise of Roam — that it would improve my thinking by helping me to build a knowledge base and discover new ideas — fizzled completely
Where should new ideas arise in old notes when nobody sees them? Old notes just remind us that something is important. Revisiting them later, with new connections from new memories, it's possible to have new insights.
There is an easier way to access new memories: When notes are published, other brains, with many more, and different memories, can find solutions or advancements.
Historically, scientists only publish polished results, apart from exceptions like Hilbert's problems. Even those were remarkable problems. Who would dare to publish their minor nuisances or shallow observations (unless they come with a nice picture)?
The author hopes that AI can find answers. Why should we wait for AI when there are already intelligent humans? The problem lies in managing access to our thoughts. We don't want to allow everybody to see each of our notes.
>One interpretation of these events is that the software failed: that journaling and souped-up links simply don’t have the power some of us once hoped they did.
>In short: it is probably a mistake, in the end, to ask software to improve our thinking.
Software could improve our joined thinking. We need an acceptable format to publish our notes.
The most an AI should do is find relevant notes on a topic I search. That's it. I don't need its robotic voice summarizing my grotesque ideas in a painless, unpassionate way. The last thing I need is more of my ideas seeping into more of my computers in hopes of somehow making me smarter.
Personally, I'd just like an assistant that will automatically label my mails. Per project, utility bills, accounting stuff, all in their own bucket.
The per project stuff is the problem because just using the sender won't work, and there are even emails referring to multiple projects. Which i'd like filed as multiple copies in multiple folders. Can the current LLMs sort that?
On my local machines not on someone's cloud thank you very much.
I'd pay maybe a Starbucks coffee per year (not month).
"Hey what notes did I take this week related for my job? What are the actionable things I owe others?"
My hunch is that generative pre-trained transformers can help provide insights, and I'm interested in trying it out. For those struggling with their note-taking systems and stuck with exact/regex matching, the ideas in this article, while not new to me, would provide some ideas. Leaning on "AI" to be the answer to a mess of random notes probably won't get the value this article is hyping.
But I admit that the Zettelkasten method has had marginal benefit for me, possibly because I'm not working that way all day every day; and simply copying and pasting articles from the web into your knowledge base isn't going to accomplish much of anything for anybody.
I don't know what kind of formatting you're looking for, but most of the apps I'm familiar with are markdown or just plain text. I use Vimwiki in markdown for my notes, but it can also be exported to HTML, which I don't use often.
Since a few weeks I started playing around with embeddings and the results are pretty mind-blowing although it’s just an improved search for now. One unexpected outcome for example is that search now includes snippets with similar meaning but in different languages.
If you waste ~3k words to argue such an evident point as "note-taking tools can't make you smarter", of course you should be surprised if people would spend even 47 seconds skimming through your article.
So for me, main benefit is extra information which I retained while/because I was writing it down. I stay away from commercial apps, just org(roam) mode since emacs is not going out of business
In smartness terms they give me:
- good reference when I need them
- details on something when I want
My brain is still the same, no IT tool so far can extend my neural processing, but the overall results are pretty positive.
My note against much other "apps" is:
- not integrated with anything else, so you can't link things like mail, financial transactions etc
- not easy to run automation on top of notes (something more complex than simple agenda reminder)
That's not much a "note app" issue but a system design issue. Emacs is a fully integrated tool, like classical systems, in the modern era for business reasons anything is just an isle with all the flexibility and usability issue of the case.
The Notetaking Cold War (2020) - https://news.ycombinator.com/item?id=37266200 - Aug 2023 (14 comments)
2. Learning requires a risk of loss. Without some kind of loss being averted, your mind won't retain nor focus on the subject at hand.
This is why TV and entertainment or fiction rarely help us learn anything.
BENEFITS
The biggest benefit of note taking is 1. Writing the notes helps your instantly retain - hand written notes much more so 2. you miss about 20%-80% of verbal information when its presented first tiem - especially subjects we aren't familair with.
Add in a "random note" plugin, and voila, I'm exploring the recesses of the mind, bringing ideas from the back burner to the front, marking old tasks as done, completing half-finished earlier thoughts.
My personal notes are pretty much the last thing that needs to be AI-fied into a generic Quora summary.
Apple still hasn't learned how to do online stuff...
But, how you use it can make you smarter,
or suck you into a cult.
> Why note-taking apps don’t make us smarter
Because... they're not supposed to?
To me, this sounds like "Why pen doesn't make us smarter" or "Why lamp doesn't make us smarter". Why was the author expecting it to make us smarter anyway? Where did they get that expectation?
This is such a strange comment.
"They’re designed for storage, not sparking insights. Can AI change that?"
That's the sub-header. The followup. The next line.
> Because... they're not supposed to?
Tell that to the note-taking apps that are adding AI to their apps in order to "make us smarter" as some would say. To improve our knowledge, to think bigger, to brainstorm for us.
> Why was the author expecting it to make us smarter anyway?
The headline doesn't say that at all. In fact, the article starts off by making it very very very clear that they aren't doing this.
> Where did they get that expectation?
They didn't. That's your conclusion.
> To me, this sounds like "Why pen doesn't make us smarter" or "Why lamp doesn't make us smarter".
They sound nothing alike. The sounds are completely different. You should get your hearing checked. =)
In my experience, thinking more is the only thing that will reliably make you think better. And to think more, you need space and time, not a breathtaking density of inputs and outputs surrounding your thoughts, which seems to be the note-taking industry's solution.
ie.: a simple walk in a park or a calm and silent place
Wanna get fit: watch what you eat, burn more calories than you intake, exercise, probably minimize alcohol intake.
Wanna learn stuff: intake new information from a reliable source, write it down (in anything), reprocess the information in your own words, and review often enough for you to pick it up.
Anything I'm missing?
> burn more calories than you intake
That's harder than it sounds. Most people incorrectly measure how many calories they intake and incorrectly measure how many calories they burn.
Even for folks who correctly calculate both, they then have to change behaviors -- usually behaviors that have been developed over decades. Again, harder than it sounds.
This is why so many people keep seeking out weight-loss solutions: because the best way isn't easy.
While most folks today can pick any note software (or notebook) off the shelf and happily jot things down, a relative few folks are trying to do a _lot_ more with their notes and are going to be frustrated with the generic tools that most people use. For those few folks, the easy way isn't the best way.
And this isn't new: depending how we categorize it (har), folks have been trying to refine their "note-taking" for hundreds or thousands of years. I don't think they were looking for a panacea; they were heavy thinkers who wanted to improve their process.
I love this article, because I've had to discover this independently.
The specific point that I realized was the following: You know the whole mindmapping thing with the bidirectionality and the fancy graphs and so forth?
That's REALLY appealing to us for the following reason: We, as humans, are already very good at it. It's literally how our brains work.
But also, consequently, that's why the apps suck at it -- we do this WAY better than they likely EVER will. That was one of those "once I realized that, I was enlightened" things in this space for sure.
I've just kind of found that, good bookmarks/article tools (for me that was Shaarli, and later Shiori) and then just making notes to myself ALL THE TIME, even if I don't read them as much, does a much better job of this. I'm the real computer here :)