The first is that when I do take notes, I over-write them. I’ll find myself nearly word-for-word copying the information I need. This is a habit from my years in education where I would basically transcribe lectures as a way to maintain attention as a strategy for ADHD. Therefore, my notes end up a muddled version of the source material and I’ll just use the source material for reference, making the initial note take a big waste of time from the standpoint of having an easy reference available.
Another problem I run into is that when I’m very attentive to a task or piece of information and want to take some note on it or reference notes from elsewhere, switching contexts to my notes is disruptive.
Are there any resources you recommend for effective note-taking?
Jokes aside, I do find note taking to be a disruptive context switch a lot of the time.
The best way I found to get around this when reading non-fiction is to 1) read a chapter/article through once from beginning to end without taking any notes or even paying particular attention (optional step if time poor, but genuinely helpful), 2) reread more attentively now that I have some bird's-eye view of the author's intent and structure of the chapter/article, occassionally taking little mental notes or asking myself questions, but not usually taking written notes, 3) go through the text paragraph by paragraph, taking copious notes (optionally blended with the next step if time poor), 4) write section summaries, what was the author's intent in writing this section, what arguments did they present, what evidence did they use to strengthen their position, and does anything immediately leap to mind that is relevant, or challenges what the author is saying?, 5) read through the text one last time, double checking you've understood what the author is saying before writing a short summary of the text, presenting the same things as the section summaries, but at a greater distance.
Obviously, this can be extremely expensive. To the point that you can't possibly use it for every or even most texts that you encounter. However I have used this technique at a graduate level to understand in detail texts that left other students completely flummoxed. So if you are in a situation where a single text is key to you understanding something you need to learn, this is extremely effective.
Edit: I forgot to mention, that you should take notes in full form prose both as if you were explaining it to someome who wasn't reading the text. Furthermore, reread notes immediately after and ask yourself questions to check you've really understood the botes you've taken. This is why reading the text top to bottom at least once before any note taking is important in my view, because it helps you avoid ignoring lack of understanding on the assumption that the text will clear up something later on, when it may require active attention from you.
Furthermore, feel free to hope around and reference the text while taking notes and double checking your understanding, but I recommend taking the notes in strict linear order in most cases.
Step one: Collect as much as possible. Don't worry about what's important or not, you don't know enough to make that judgement yet.
Step two: Make it shorter. Paraphrase a direct quote, remove unnecessary info, make it more useful to your future self, anticipate what you'd be looking for in the future, you're your target audience. The thing that works the best for me is limiting myself to a very short bullet point list (3-5 items) right at the top of the note. That way, if I need to remind myself of more than that, all I have to do is just scroll down.
Step three: Never open a note without making some changes to it, even if it's just fixing punctuation or re-ordering things. That way, the most important notes are always significantly better, more complete than the ones I don't particularly care about. But, I don't feel obliged to chase that perfection from the start, if it happens, it happens organically.
Depending on your source material, you might be able to use things like Zotero and Obsidian to annotate and export them into text files, for PDFs and webpages even I think. Just coming out from helping someone setup the workflow.
But if it is more like code or video, I am not certain. For me Emacs org-mode has been a huge help. I can insert links to the source content like code either directly to files or GitHub permalinks and write down my notes, have code highlights and even in some cases run the code and have results inserted back into docs.
personally, the best resource for me is GPT-4. It's my second brain, a rubber duck, living notes that benefit from being over-written especially since they recently added memories. I write as much as I can on paper and pencil, I look it over and I "chat" with GPT-4, keeping different chat conversations for different topics and later I can ask it things like "it's been a while, summarize everything from the beginning and give me suggestions for where to continue" and things like that.
There's more value to note-taking than just having a reference afterwards. The act can be helpful with retention. I don't reference a lot of my notes.
I'm not saying you should do these things. I'm just offering the palette that I've evolved and saying that this somewhat works for me. Everything here came about as an adaptation in response to the pain of not doing it.
1. I only take verbatim notes in meetings where I'm not in charge of the pacing. Even when doing that, I try to notice the things I'll need later: commitments to actions, unresolved questions, specific decisions. I bold those so they're easily findable.
2. If I'm in charge of the pacing, then I only write down what I need to. Experience has taught me that we meet for different reasons and I try to focus on the things I'll need later: decisions, actions, open questions, key frames of the discussion. For the latter, I'll make grids or lists as a representation of the frame and share my screen so everyone can see and we can build the artifact together.
3. I have found it impossible to say something and take notes on it simultaneously, so when I'm a participant I have to remember to end my thought with "ok, now give me a sec to write that down."
4. Sometimes we're meeting to build something together: a schema, a plan, a proposal. In that case, I'm not taking meeting notes. I'm editing the schema/plan/proposal and keep track of actions and unresolved decisions in the doc. The document is done when there are no more actions left to do and no more decisions to make (or we can live with the unmade decisions).
5. I use a filesystem to organise my meeting notes, and fulltext search over it. One file per meeting generally.
6. These are Word files, ymmv. First thing in the file is title-styled description of the meeting purpose (sometimes just a company name if it's us meeting them). Second thing is list of participants, broken down by company if it's that kind of meeting. Then it's bulleted list notes.
7. The filename is the title and date. I can search filenames by project and see in the list of results what date the meeting happened (may not be last modified timestamp on the file).
8. Can't emphasize enough taking notes with the end in mind. Experience has taught me that there are some things I need to write down: the things I say about how we'll do things that I need to have a record of so I can say "no, we told you this would happen"; product and company names; any actual needs/requirements/problems from customers or staff; any commitments explicitly made or not made (and listen for commitments being glossed over or implied, and ask questions to make them explicit so they can be recorded).
9. If you're calendared/scheduled then book time after meetings to deal with the notes and actions and questions from that meeting. Otherwise you just accumulate open loops that sap your will to live. Share the notes to participants as soon as you can (I often do it at the wrapup of the meeting).
10. I'll assemble notes on product or technical things to understand them. There I look for what problem they try to solve, the moving parts I'd need to know, the key bits of logic, dependencies, etc. These get published on the intranet because if I've had questions, other people will too. They're never perfect, almost always still unresolved questions, but they're good enough to answer a lot of questions for me and others. I book meetings to go through them with people who hold wisdom in that area, and we answer questions and uncover new areas together. I'll screen- or document share so they can see what we're working on. I know these work -- I forget I've written some and I'll have a question, go search the intranet, and discover that I answered that question for myself 18 months ago and present me thanks past me.
11. I can't really help with context switching. As I took more and more notes, I got better at it. The trick is to be listening at the meta level, to recognise things that need to be recorded. Using numbered/bulleted lists keeps the amount of context in my doc almost to zero. If I'm transcribing a meeting, I'm just adding something new. If it's one I'm running then I might have a bulletpoint for Next Steps and record explicit actions there as they're uncovered. Also, see my final para in this post.
12. If you do record Next Steps, say WHO and if possible BY WHEN. And if the WHO is you, add it to your own todo list in the post-meeting time or it'll be forgotten. Any time I have more than one todo list, I've lost. (Not saying I always win even when I have one todo list!)
Meeting notes example: when we meet Quality Foods to talk about integrations on May 1st, I'll make a file called "Quality Foods Integrations 2024-05-01" and it'll start with:
<Title>Quality Foods Integrations 2024-05-01</Title>
Attendees: Nat, Alex (Ontempo), Alice, Bob, Charlie (Quality Foods)
* Looking at two different systems, one for finance and one for marketing.
* Finance = Microsoft Business Central, using a third party specialist to configure it.
* Don't know the integrator yet.
* Marketing = MailMuch.
* <b>No API docs.</b>
* We have BC integration already. MailMuch is new.
* Explained the flow of an integration: scoping, requirements, iterative development. Time and materials.
And finally, a note is not a general purpose winning tool. I still find I've missed things, wish I'd written down things, can't find notes from a meeting I'm sure I was at, etc. You always will. Don't look on perfection as the goal: they should help most of the time, and if there are occasional lacunae that's still better than not having them.
I continued this practice at college and credit my success at school and college in large part to the confidence I created in the "memory" that my written notes gave me.
I continue to feel that I have a very poor memory and so note-taking has become a lifelong skill|tool.
Hand-written notes are best for me to actually remember (without recourse to my notes) but I almost entirely swype|type notes today because of the benefits of accessing them anytime (any device), searching etc.
I try to keep my note-taking DRY and prefer to update|correct older notes rather than recreate content.
https://en.wikipedia.org/wiki/Don%27t_repeat_yourself
Here's my current practice:
1. I write technical blogs partly to share but equally to document for my own purposes; I have a blog called "snippets" (co-opted from a Google practice) where I document tips-and-tricks from coding (e.g. how to set up Protocol Buffers proto_paths when using Visual Studio code)
2. I keep arbitrary notes in Google Keep. My shopping list was migrated to Keep too and, I write trip packing lists and the humans' names at the dog park (keyed by their dog's name, of course) there too.
3. Google used to have (!?) an excellent practice that everyone in a meeting recorded meeting notes in real-time in Google Docs. I would comment-ping people with to do items.
4. Until recently, I used Google Docs for record my programming notes. Every day, I would add a new H3 header for the date with "Actions", "To Do", "Miscellany" sections. I have 3 years of one project's development recorded this way and it served me well. However, as the document grew in size, it became unwieldy and I was forced to split it into two documents (before|after) which made it less useful. I've begun (though it's also imperfect) using Markdown and a Hugo static site to record the same content. Now, each day is a dated Markdown file. This scales better. I prefer using Markdown. It's just as accessible but it's not as good for searching.
5. I contribute to Stack overflow and I've a GitHub repo for every answer I've ever worked on. When I write code or explore a technology, just as everywhere else, I document the solution (as a README) and commit it to the repo. This is mostly for my benefit because it allows me to quickly find technologies that I've worked on.
My coding velocity is slowed by my sometimes obsessive desire to make notes but, note-taking has paid me so many dividends over the years. For example, I rarely feel anxiety that I know I've used an obscure tool, programming language or some such but am unable to find a record of this; reading my extensive notes gives me a thorough play-by-play of what to do.
org-mode|org-roam-directory/
- archive/
- longrun/
- yyyy
- finance
- work
yyyy contains daily notes mm-dd.org and "binders" that are notes I plan to move in the longrun eventually. Anything goes in daily notes, than get linked/org-transclude-d elsewhere as needed.
longrun contains for instance notes about a piece of hw, a keyboard or a car, it doesn't matter, it's a kind of summary of many daily notes to gives me any insights related to the hw, also people notes namesurname.org, also anything else that might be useful for more than one year.
archive contains archived years (so far zero, since I use this approach since 3 years) but also my old notes, some longrun notes no more relevant so archived etc. The sole difference than the other trees is that it's not part of org-agenda for performance purposes.
finance essentially is in part files tangled in a time-based hierarchy from org-mode daily notes, used with fava and beancount CLI + a bit of org for reporting, summarizing etc. Essentially my life financial status and history, linking anything from taxes and bill, mails and pdfs etc, all org-attached as needed.
work is a separate taxonomy for work, meant to be shared eventually (not so far, but I still can if a day it will be needed) with colleagues or dropped entirely.
As a bottomline Emacs is also my desktop (EXWM) and in notes I have almost anything from NixOS to zsh and Emacs itself configs, tangle-d from notes. Org-mode is information and meta-information with relevant visualization, linking anything and so on. org-attach handle 99% of my files accessed via org-roam-node-find/counsel-rg or some other search&narrow tool.
I've hesitated around 2/3 years before demolishing my decades old hyper-curated home taxonomy but in the end I regret NOTHING, keeping anything in notes, thanks to Emacs, is the way to go.
Secondly, I'm going to skip the mechanical recommendations (the 'how-to' stuff). There's a great book called _How to Take Smart Notes_ that a lot of folks here rep these days. I quite enjoyed it. _Getting Things Done_ is also good. Either book will help you in this regard far more than I could. Instead, I'm going to focus on emotional dimension.
Claim: you can't build a habit unless you 'cathect' to the ritual and the tools.
Thus, you need tools that you like using, that you connect with, and you need a place you enjoy using them in, and you need a time every day that is generally available for this practice.
(Note: a lot of people (myself included) are initially frustrated that they cannot bring themselves to like _very_ inexpensive things. If you happen to love cheap ballpoints and scratchy, acidic paper, you're very lucky; otherwise, you're going to be hunting for the perfect notetaking implements for the rest of your days. Adjust to this idea. If it helps, you may reflect that paying extra for nice paper and pens is a trivial expense, if they get you writing, because writing is important.)
So, first of all, get yourself a notebook that you really like. Some folks do a single-notebook thing, but I prefer a small notebook for on-the-go and a larger one as a 'thinking surface'. (As it happens, I also like _very_ large thinking surfaces, and keep drafting paper and a roll to keep it in.). Make sure the paper quality is reasonable. I'm particularly fond of the dot-pitch notebooks by Statology, and Kleid also makes some amazing products. So do Clairefontaine and Rhodia (and those are generally easier to track down.)
You're also going to need a pen that pleases you to touch and use. If you, like me, appreciate fountain pens, but don't want to spend a fortune, I'd recommend, for example, a TWSBI Mini, which will still run you about $70 USD. (Skip the TWSBI Eco at $40 -- it's worth the extra $30 for the Mini.) I also quite like a Platinum Plaisir at about $15, if $70 is too much for you.
Get yourself a bottle of nice ink as well (I recommend an Iroshizuku ink by Pilot).
Some other accessories that making writing a joy: a leather (or silicone!) inkblotter; a great playlist or, in my case, a public radio station that has an amazing show on at _just_ the right time every day; and maybe, if you're so inclined, a small amount of whiskey. (Emphasis on the 'small' -- you don't need to go full Hemingway on this one.)
Now, prepare the rest of the ritual. Remember, the goal is to trick yourself into spending a certain enjoyable amount of time every day writing, reading back, organizing, and reflecting on your notes.
And that's why you'll find me in my study, every day at about 8pm, talking a brisk walk around my mental garden. If you don't have a study, a kitchen table will do. If you haven't got that (I've been there,) find a nearby café that feels like home.
And that's how I keep notes in 2024.
P.S., if you find your handwriting is terrible, well, you probably need to use it more! Consider also calligraphy as a means of refining it.
I think you’re right I need to just get back to that system, anything digital is too ethereal to really stick to.
As for pens I’m a hi-tec-C guy btw
Obsidian also is good for storing _backups_ of my notebooks, which I make with my smartphone camera. It's pretty nice to be able to thumb through a handwritten notebook from a few years ago.
the Neovim era was great for establishing the basic folder structure I use, but ran out of gas when I needed to introduce notes which weren't UTF-8.
The Hi-Tec C is a reasonable pen, and I wouldn't disrespect it, but a mid-market fountain pen is as far as I can get from my calligraphy practice and still feel comfortable; I am also probably overinvested in the specific qualities of inks that I use.
Still, I permit the roccoco aficionado-mind to flourish, as, again, it is in the service of a greater good. Even a fast ship needs ballast.
---
It's all depends on your need. Only you can devise a suitable system that will keep you taking notes. As for me, I realized that while I have good memory, what I retain was a bird view. I recall by "revisiting" the concepts again through whatever contextual information I stored (the layout of the page, the place I was, the logical reasoning,...) I did good in school, but mostly because the exams reminded me of the books, and the questions surfaced the answer (I was not great at reciting).
I found I was forgetting key information about some stuff, because I was too busy understanding it (trying to make a reasoning model about it) that I was skipping over information. Any problem solving session needed research and that usually means a lot of googling. But search engines have not been good lately (seo garbages), so I decided to keep everything closer. My notes system is mostly for keeping interesting facts about things I read, did, or thought. Spaced Repetition Systems (SRS) could work, but I was never fan of learning things by heart (again school) and believed in understanding and practicing (you only do what you like to do or need to do, so that act as a filter of what's important)
So I keep hoards of PDF, have two database of interesting links (Things I need to read and things that are interesting). My notes are tidbits of information: Documentation, Project Planning, Books extracts, summary. I only do the barest of organization (I use Bear which use tags and can do full-text search). But what is more important is revisiting, and rewriting. Everything is an idea and a new exposition can spark new ones. I'm not memorizing anything, they're just there for references and brainstorming.
---
As for the note taking itself, if it's a short piece (web articles), I try to read it as a whole first, then go back to take quotes of what I liked (and the source). Maybe I add a short observation. For longer piece of science fiction, I convert it to PDF (because of better annotation support) and just highlight what elicits a reaction. Then I reread the chapter (or two or three if they're short) and only focus on what I marked the first time. I either copy the whole thing over, dismiss it, or do a short summary. If the book is a map, what I do is trace the interesting landmarks and trace a route through them. As I said before, I don't care about memorizing, I just note what was interesting/important for me in the book. Another reread later, and I'd probably highlighted different things.
> Another problem I run into is that when I’m very attentive to a task or piece of information and want to take some note on it or reference notes from elsewhere, switching contexts to my notes is disruptive.
My advice would be to make it quick. If I'm reading on my computer. I'd copy-paste the interesting fact (just create a new note and add the date somewhere, maybe). If it was a physical book, I'd just grab a piece of paper, or my phone to take a picture. If it was on my phone, I screenshot it. I just want to store it quickly as possible and get back to it later. If I want to reference something, but the current task is reading and not collating information. I just add a quick task on the note (ex. <this> reminds me of <that>, check it out".