93 comments

[ 4.3 ms ] story [ 219 ms ] thread
Clairefontaine engineering ruled spiral notebook. There are various sizes but I prefer the larger one. Good paper feel, inks don’t bleed and it survives riding about in a backpack with the laptop. Everything goes into it so there’s only one item to remember. Pages are dated, key entries are highlighted. Bits can be scanned. If I know when something happened, I can scan backwards to restore state. Even random scribbles go in, and the gridded pages help drawings.

“If you don’t write it down, then it never happened…”.

Yeah, some do it with a keyboard, but I find it hard to draw, scribble or whatever when I have to switch devices or associate drawings with text. Eventually, they’ll figure out a tech solution to all this but notebooks cost like $10-20, survive impact and even survive a bit of liquid.

No, I didn’t think this up myself. Merely watched a boss never lack info on what went down last meeting or last week.

> notebooks cost like $10-20, survive impact and even survive a bit of liquid.

There are even liquid tolerable notebooks that are more like textile and even those are pretty inexpensive.

Paper-feel can be a big deal, though.
If you really need the "if you don't write it down, then it never happened...", then get a real bound notebook with numbered pages. It's too easy to tear out a page in a wirebound book.
I’ve used these when there’s a legal requirement but the spiral lies flat when folded in half which is a win on my cluttered desk. Fortunately, my coding has usually not been where the exact time of creation is critical…
Sure, but as long as it is digital plain text files.

I've tried the minimal paper setup and it got to the point where:

- Recall was terrible (both in-notes, i.e. no quick search and physical notebook misplaced at the most incovenient times)

- Insane amounts of duplication (no easy linking to old notes)

- Unshareability (couldn't lend notebook to colleague as it was personal, had to type it anyway)

- No redundancy (dog ate notes)

I now have a digital text file system. Although it is not a "NASA project", I can add searchability with basically any tool, easy linking (with tags or wikilinks to dates, topics, etc), copy-pasting and proper backup. It is portable across devices and OSes and allows me to perform simple (and complex) analysis with simple scripts if I need.

Being simple text, I expect this system to scale for the rest of my life.

Actually I think that creating a "perfect" note taking system reduces to finding an ideal compression algorithm for human knowledge, which Hutter showed is equivalent to solving general AI.

So you're wrong, note taking is at least as hard as rocket science, probably much harder since algorithmically rocket science is a pretty straightforward control problem plus some tuning.

You first blew note taking out of all proportion as an axiom "reduces to finding an ideal compression algorithm for human knowledge", and then deduced that "is at least as hard as rocket science".

There's no reason we should view note taking, the activity, as the pursuit of finding "an ideal compression algorithm for human knowledge". Finding just "a compression algorithm for human knowledge that suits us" will do.

When we do note taking just want to be able to find and remember something down the road. That's what we need to solve, for our preferences and requirements, and being happy with something that works good enough.

The same way that when we play guitar we don't want to be "the best possible technical/emotional/harmonic/melodic player that can possible exist", just to be able to play songs to have fun, play in a band, or be good musicians.

Yes, but it depends "what suits you". It can be arbitrarily difficult and there are similarities that go along the lines of fundamentally challenging CS problems, which is why everyone and their plushy-toy seem to be obsessing over it I assume.

Hacker news people are probably curious nerds.

HN people are also notorious bikeshedders.
I've recently simplified my note-taking workflow but there are complexities. IMHO, a note taking application needs a way to at least do the following with notes:

* View and edit

* Organise

* Search

* Sync

All in one solutions can be nice but also lock you in. It is as the name says: all or nothing (well all or figure out how to migrate your notes to another application). My current workflow is to use different tools for different jobs. The downside is it's more to set up and it may not be suitable for everybody.

* I use any markdown editor to view and edit notes

* I use the filesystem to organise notes by broad categories (directories are never more than two deep)

* I can use any tool I like for searching notes (my markdown editor(s) has one but I can also all use ripgrep or whatever tool I like)

* I sync using my preferred cloud storage provider

My markdown editor is doing most of the job of a note taking app but the point is I can seamless switch for something else without much hassle. Or use different ones on different platforms.

This does require search tools to do a lot of heavy lifting when broad categories aren't enough to find something. But I've yet to find a better way (I've not had much success managing tags, they get messy fast).

Obsidian[1] is an all-in-one solution that doesn't have much in the way of lock in. It fundamentally runs on Markdown files sitting in your filesystem (and you can even "reveal in Finder" from the UI). It has its own custom indexing to make things quick. They offer sync as a service, but people also use other sync solutions. They have a mobile app version to access notes elsewhere.

Beyond just folders for categories and tags, Obsidian also supports easy linking between notes (and shows backlinks for the current note).

In the end, it's great: should I decide Obsidian is no longer right for me, I have a bunch of Markdown on disk that I can take elsewhere.

[1]: https://obsidian.md/

From trying note-taking multiple times, I think my perfect note-taking systems involves markdown files that I sync myself. And then an AI goes over them, understands what I wrote, and the context, and organizes and interlinks those files. Turns out, the 2nd step is a bit harder than rocket science.
This post is practically devoid of content. Why do you care what other people spend their time on?
Why do you care what other people care about?
Their posts end up on HN and make the signal-to-noise ratio worse.
As someone who has spent hours watching some "creative" person painfully "transition" from Notion to Obsidian, I can only echo this pithy response.
I tried to transition from years of loose txt files to obsidian.

When I opened the folder the app showed me 1-2 files from years ago.

Apparently it only supports .md files.

I literally laughed out loud and deleted the app from my computer.

You've got a damn knowledge map but handling txt files is beneath you?

That app is doomed to obscurity.

> That app is doomed to obscurity.

Quite arrogant to assume that an app not supporting /your/ specific note-taking workflow means it's "doomed to obscurity". It's a popular app with plenty of users, a healthy plugin community, and a team of six building it. I think that pretty obviously proves you wrong.

If you were actually interested in trying it out, you could've simply run a quick find command.

find . -name "*.txt" -exec rename -v 's/\.txt$/\.md/i' {} \;

So simple you can fix the problem in one line but me pointing that out as bad design is arrogance. Gotcha.
Converting all files from .txt extension to .md extension would be enough to start. This I what I did.

The only annoying thing was that in my text files I often used indentation to group stuff and in markdown indented text is treated as code block.

Moreover, it's just advocating a specific extreme to a broad set of real-world situations in "just do it" fashion.

There's usually a balance between planning and execution. Note taking is no different.

There's no simple always solution to "plan meticulously" vs "just do it". Every choice in life is a cost-benefit analysis.

Personally, I spent a little time to setup orgmode. Works for me.

It's also tautological. Of course, we should use the note-taking approach that works for us.

Who would agree with using a note-taking approach that does NOT work for you?

One of the ways to figure out which note-taking approach works is to try the note-taking approaches from other people... It could be 18 multi-step note taking approach or it could be dead-simple. Unless you try, you won't know which one works for you.

This post is like a litmus test for people who reply to the title instead of the content.
What content? There is literally more content, and more thoughtful content, here on HN in response to the article!
It depends. We’ve seen this “thoughtful content” of HNers posting in detail about their personal notes-taking systems ad nauseum in uncountable threads already. No one asked for them to do it again here.

The whole point is that an elaborate system for taking notes is unnecessary and likely detracts from the actual work getting done.

I'd reckon everyone is seeing greener grass on the other side, it's hard to believe that Markdown is more time-consuming than plaintext. The entire point of a "notes system" is that it works well for you. If you only need to take weekly notes, you won't need organization at all. If you're studying a variety of academic topics, then yeah, you're going to want a notetaking schema of some kind, even if it's only hierarchical folders.

I'll still agree that notetaking doesn't need to be complicated, but... I don't see Markdown or Zettelkasten as complicated at all. It's like telling someone "version control isn't rocket science" and encouraging them to uninstall git in favor of versioned tarballs.

The OP doesn't actually recommend people do that. It just says that note-taking need not be overcomplicated. No one said Markdown is time-consuming.
Then I guess I have to agree with the HN consensus, here. This is a pointless post with no call-to-action, it purely exists to jab the Pathos stick into the side of people who could care less.
I see it as existing within the context of cultural trends. For years productivity porn and life hacks were in vogue. But the market for it has gotten big and bloated, the content overwhelming and over-monetized, so there's a cultural backlash against it now. The post mostly serves as a cathartic cry.

For something more substantiative, satire can be more effective: https://www.youtube.com/watch?v=_o7qjN3KF8U

Though that video is from- 2018?!

> For years productivity porn and life hacks were in vogue. But the market for it has gotten big and bloated, the content overwhelming and over-monetized, so there's a cultural backlash against it now.

I agree. How is this a critique of any of that? They aren't calling out over-monetized SaaS products; they're calling out the entire market of notetaking apps. Their rhetoric doesn't support the idea that productivity porn is harmful, but rather that simplicity is paramount and everything else is superfluous. Which is simply wrong. Maybe it's easy for a solo-dev to piss on knowledge bases, but well-organized notetaking systems is what keeps the corporate world going round.

Unless the author has a better, simpler alternative to the organization products people already use, this article simply reads as whining. Who is even the intended audience?

Yep, I know there's no content.
If by that you mean that people that saw the content didn't participate coz there is just nothing there, sure
The one and only article linked to by this one is arguably worth posting here, and in fact, it was, about an hour ago (I'm guessing as a response to the vacuity of this one.) Unlike this one, and somewhat ironically, it seems to have been ignored so far.

https://news.ycombinator.com/item?id=32887140

That's exactly what I meant :).
Let me mirror that with a sarcasm:

Code typing is a practice of typing code, really. Why some people building their code editors like NASA engineers building the next rocket to send us to the next habitable Earth?

You don't need to meticulously craft your code editor, stop overthinking.

> You don't need to meticulously craft your code editor, stop overthinking.

You don't need to, in the sense that you can craft code in anything, but surely anyone who is an expert in their editor of choice—whatever that editor may be—would feel a huge productivity hit if forced to use another, formally equivalent, editor.

You missed:

>Let me mirror that with a sarcasm

It is a remark to the post's argument. Indeed someone doesn't need a note-taking system but their productivity may increase with one. (Whatever a note-taking system is to someone.)

> You missed:

> >Let me mirror that with a sarcasm

I didn't miss it, but I guess I misunderstood its scope:

> Let me mirror that with a sarcasm:

> Code typing is a practice of typing code, really. Why some people building their code editors like NASA engineers building the next rocket to send us to the next habitable Earth?

> You don't need to meticulously craft your code editor, stop overthinking.

I thought that the second paragraph was the sarcastic one, and the third paragraph was the sensible rebuttal. On re-reading, interpreting the second and third paragraphs both as sarcasm makes more sense.

I agree with the general spirit of the article - so many developers are obsessed with finding the best system and quite a few even write their own. But I have not really seen people who spend that much energy on notes to be any more productive...
For computer note taking, I love vim wiki. I used to use a Google doc but I really like how quick it is to open and navigate in vim.

For paper note taking I like a good notebook with a fountain pen. Writing has become enjoyable once I re-discovered fountain pens for myself.

For me, syncing my Apple accounts on my mobile while keeping them separate on my work and personal computer has been a game changer because it was easy to do, it requires no thought or maintaining
Adobe Illustrator and other infinite canvas platforms provide a compelling case against this. Forcing yourself to draw pictures of what you’re taking notes on especially in studying software systems where the ideas are abstract and not necessarily visual is really useful. The added ability to nest concepts inside other concepts like Russian doll note taking is also pretty interesting. You can also screen clip whole articles and paste them next to your drawings and notes. Infinite canvas note taking is probably the most useful thing I’ve ever discovered in terms of learning technical systems.
I have had text file + folder system for over 4 years now, and I feel it iteratively gets better and better.

The upper level (text file) category determines the high level structure, e.g. 01. todo; 03. ideas, concepts; 06. thoughts; 12. humour; 04. politics; 19. food; 05. self; 20. mathematics; 21. philosophy, etc, and the file itself contains high level remarks/thoughts, depending on how necessary it's to have subcategories. The folder is named after the text file category and it contains subcategories and material (research papers in pdf, pictures, sketches, etc). The system structure moves recursively downwards, although in most cases I only need 2 levels. For example "19. food" has subcategories such as 01. vitamin content; 02. electrolytes; 03. meal design, etc, and these could have their own subcategories.

I feel my system is like an externalized version of my ego (including extensions). As I put more content into the system, I become more aware of myself and my shortcomings and the thought/emotional patterns I have, and as I make changes to the system I develop/restructure my own thinking. It works really great, but I still haven't figured out 1) how to deal with the material in version control, because I'd like separate the text files that I put into git from sizeable material files; 2) how to automatize the backup process; 3) how to store sensitive data in git automatically (it would be nice if I could encrypt files in a meaningful way and still be able to have a measure of the size of the diff).

Nice. Sometimes I wonder why we don’t go full circle and bring back gopher. Your filing structure brought that idea back.
I'm down for purely text-based web experience :-).
"Write better notes by writing notes". Thanks..?

No idea what this blog post is getting at. Maybe they're trying to point out that consistently writing notes will produce higher quality notes than changing styles every 6 weeks?

I love space ships and drawing space ships. As a results my notes have always been full of space ships and that becomes a metaphor. Big important ideas get big spaceships next to them. Connected ideas get space ships flying back and forth between them. Popular ideas get fleets of space ships drawn next to them. Controversial ideas get highlighted with space battles. People who see my notes and try to make use of them are often critical, but this has always helped both to engage me with note taking and to enjoy going over and really understanding notes when reviewing material. So for me note taking really is rocket science.
Finally, a notes-taking system on HN that makes some damn sense.
Yeah this much better than insert here some obscure usually German sounding word note system.
Isn’t the above the Raumschiffnotizen system?
Why can I only give a +1 to brilliant posts like this?

I had to laugh out loud. Thanks for making my day.

Agree 100%. The rest of them just make me feel bad about myself.
Yeah, I'm quite proud at how my doodling skills have improved now that I take "notes". Keeps me from zoning out, and while my pen is out I find i actually do write stuff with it
> And if you need an example of dead-simple note-taking, James Clear stores his notes in a multi-hundred Google document

If you click the link, you see that he uses other tools for notes, including Asana and Evernote. Google Docs is for writing a book.

So, looking through my notes here, rocket science is: $dv = -v_\text{ejection}(dm/m)$

Then I ran Dijkstra's algorithm across my notes (903 pages, 90 folder, 18MB), and the shortest pathway between rocket propulsion and communications (which is presumably the closest concept to "note-taking" represented in my obsidian vault),

Communications and locomotion are both applications of physics -- rocket propulsion being one specific model.

By virtue of the fact that both tasks are physics, I would claim the inverse and say that note-taking is indeed rocket-science, and therefore complicated as the common metaphor would suggest you're implying.

But then I loosened constraints (essentially 6 degrees-of-separation between communications and rocket propulsion), and came to the following realizations:

- Ink, toner, and the advent of electronics like the laser printer, are owed at-least in part to the broad effort in science of the space race.

- Additionally, since the flames of a rocket reach temperatures that would exceed human limitations, and the human limitation that speech is primary form of communications means that good note taking is a way to overcome a key human limitation I've identified.

- And since presumably, human communications regarding models of physics resulted in the physical laws and principles that resulted in us producing the thrust force and exceeding the escape speed imposed due to the gravitational interaction using combustion -- note taking caused rocket science.

Just my calculus on the situation.

I have noticed here for a period of time some people posting blog posts about their way to take note, I think it's about directing traffic to their blogs, and building klout, probably the biggest contribution of those posters is to make sure you know which blog url you can blacklist as they have nothing to say and just build clickbait, so I am thankful for that
> Get what works for you.

Yes, I agree. What works for me however is a fairly complex system (although the complexity is hidden in everyday usage) that took me several hundred hours to build, write and refine.

Taking notes is not so hard. Feel free to use any paper in your reach or any software capable of storing text. The hard part is actually using these notes. Getting something out of them. Relating them to other notes. Actually thinking "hey I have note for this?!" months later and executing on that thought. And then still have an input mode so casual that you also still input notes into your system.

Of course you can elect to care or not to care about that, same as you can elect to not care about properly sharpened knifes, identifying birds by sounds or child rearing practices.

But as with most things, if you care to do it well, it's anything but trivial.

For me, the goal of taking notes on paper is to internalize the information. The act of writing it down is what helps to get there, not the fact the information is written down.

IOW the best note taking system for me would be such that after I took notes I could just shred the notes and retain all the written down information in my head.

Honest question, do you think that's something that can be achieved?

Personally I find that without at least some Spaced Repetition (or similar methods) there is no way to remember quotes, github repos or word definitions I jotted down once.

Very open to suggestions, though :)

No, I certainly don’t ;) though it’s the theoretical upper limit of note taking efficiency, something to aspire to.
The point of notes is to use information that presented fleetingly, to drive further activity, inform decisions and guide expectations.

The complexity must reside somewhere... but not during the moments I am taking notes.

Hard agree.

But I feel like the author is "criticizing me" (not personally of course) because of the complexity that comes after the initial writing-down, since that is still part of my note-taking system, no?

I use Roam Research. It's just the tool I use. I use it because there was hype for it.

I delved deep into meta note-taking with RR. Watched videos, participated in book clubs, built a zettelkasten, etc.

While meta note-taking is not note-taking, I found the time spent meta note-taking meaningful. It is a reward on its own. The benefit, is that I now feel comfortable using my tool. And using it, I do.

While I also prefer for each of my passing seconds to contribute to the production an artefact, if it is not the case then it is fine too. That is the distinction between project and hobby.

I started as note-taking hobbyist and now I guess you could call me a note-taking pro :)

Do not hate the hobbyist.

I simply use Gmail drafts. Good enough for taking basic notes on PC as well as my phone.
It really sounds like the author is intimidated by someone in his personal life and is trying to mock them into doing less.

Also, that “people be like” grammar sounds so offensively dumb. You loose all credibility for having anything smart to say instantly.

Sentence 1: ok, valid take.

Sentences 2,3: really funny (assuming it was deliberately ironic), considering "loose" -> "lose", and using an adverb to end a sentence, while critiquing grammar... ;b

Well loose being a typo: you got me. But what’s this noise about ending a sentence with an adverb?

“People be doing” is dumb, objectively.

Ooh, being picky about grammar is fun!

The absolute positioning of the adverb doesn't bother me, but I do read it as being bound to "say" rather than "lose" which is what I assume you meant.

Note taking is actually a big part of rocket science: Step 1. Take notes on rocket science Step 2. Discover and distill from notes the first principles that help decouple yourself from analog thinking Step 3. Engineer the most efficient rocket that can practically get you there
The post is short so I don’t want to read too much into the authors intentions, but it seems like their logic may be suffering from a blind spot - that patterns and systems that look like unnecessary complexity from the outside might actually be in service of simplifying the use of those notes for the writer.

You can’t just look at a system you don’t understand and deem it too complicated. You’d need some objective measure of how efficient and effective a note taking system is at doing its job - e.g., the speed a user can later find a particular note, or the success rate the user might have at querying their database of notes and getting back a quality answer.

I think he writes to people who have issues with starting due to decision paralysis due to multiple options. Complicated note systems are fine if they come from natural evolution of users "note enviroment", but a hell if you don't have any notes yet, and thus no idea if you should organise them by tags, date, lenght, topic, graph of links or their combination.