Ask HN: How do you take notes when reading pdf textbooks not in your field? | Hacker News Reader

Ask HN: How do you take notes when reading pdf textbooks not in your field?

9 pointsmarkus_zhang14 comments

Hi experts,

Happy holidays!

Context: My field is data engineering and the pdf textbook I'm reading is of Microcomputer Architecture (the Mick Bit slice book) from my laptop.

Issues: I'll list the two issues below.

*Issue 1*: How to effectively take notes when read from *pdf sources*? I feel the best way is to insert a *BLANK* page after each *TEXT* page (by saying text page simply means any page with texts), use an e-pen to hand-write in the BLANK page while circling/labelling contents in the TEXT page. I can probably insert the blank pages using a Python script but I'm not sure which e-pen and writing board I should purchase for my laptop (Lenovo T-460S). I believe handwriting is the only fast, reliable way to take long notes.

*Issue 2*: How to effectively take notes for *technical textbooks not in my field*? I just completed the nand2tetris hardware part and had no other training in computer engineering, so the Mick Bit slice book is much more advanced. But at least the opening 20% of each chapter is very well accessible so I only intend to read those parts. I hate to admit, but I actually don't know how to take effective notes in a non-academic environment. There are two questions bugging me:

    1- When should I take notes? Taking them too early, then I found myself erasing previous notes frequently. Taking them too late, I found myself forgetting previous contents frequently.

    2- How much note should I take? The book is for advanced users so I have to evaluate a lot of Boolean logic by myself. Should I include all of them in notes? Should I copy down a lot of contents from the textbook into notes?

Afterthoughts: Maybe at the age of 40 I still don't know how to study effectively. This doesn't surprise me, as back in university I tended to gloss over details and proofs and grab for good marks (I did get 3.7+ in both undergraduate and master). But now this is a hobby so I want to stop cheating myself. If anyone (especially someone who struggled with the same problems) can point me to a resource to re-learn how to study I'd appreciate the help. I learned about Pomodoro Technique but it doesn't suit topics that need hours of focus.

I've been spending some time studying several different subjects so I'll try to describe my approach. This is mostly just a hobby for me so I don't have any real criteria for evaluating how well it works but so far I feel reasonably satisfied with it.

For every subject I create a subdirectory for downloaded resources and an emacs org mode file for my notes. Unlike you I much prefer typing notes to handwriting. For math you can use latex and display previews inline in your org buffer. The only disadvantage is for drawings. I don't currently have a good way to capture those.

Generally my first step with a new subject is to try to find good online resources for it. That usually ends up being a collection of websites, PDF files that I download and YouTube videos. I often include links to websites and videos in my org file.

Next I try to get an overview of the subject and what particular topics I think I'll need to get an understanding of to reach my goals. I typically create sections or subsections in the org file for those areas.

When watching or reading content questions usually come up about things I don't fully understand. I try to write those down in the org file and at some point find information that helps to answer those questions, which of course I also note.

I don't try to take lots of notes or copy everything in the resources I read. I'll just note points that seem interesting, surprising, important, difficult and/or easily forgotten.

I'm not sure any of this will help you but that's my basic methodology.

One other thing you might want to look into are spaced repetition apps like Anki. I haven't personally made much use of them yet but I think they should help with subjects where there's a lot of material that needs to be memorized.

> For every subject I create a subdirectory for downloaded resources and an emacs org mode file for my notes.

This is an interesting approach. I was tempted to learn Emacs just because of the org mode, but then I figured for my particular purpose (reading and taking notes of pdf textbooks) the org mode is not fast/good enough. But for other topics this is definitely a very good approach.

> I don't try to take lots of notes or copy everything in the resources I read.

Just curious, how do you reference the source material in the notes? I figured that's the most difficult thing for electronics resources, so one either 1) Simply copies down the source material referenced in the notes, or 2) Somehow electronically write the notes in the source material. Neither is quite satisfactory to me, thus the thought of adding blank pages in pdf files.

Thanks again for the reply, and happy holiday!

> Just curious, how do you reference the source material in the notes?

It hasn't been much of a problem. With PDF's you can either copy/rewrite the actual content or note page numbers and/or chapter and section headings.

Usually I'm able to find good enough source materials that those become my primary references anyway and the notes are just for some specific points that aren't so well explained.

Nonetheless I find that taking some notes helps focus the learning process.

It also depends on the type of material you are learning. For many subjects there will be some kind exercise or other work product that captures most of the skill based knowledge. For example if you're learning a programming language you'll usually be writing a lot of small programs. For math, if you're studying it fairly seriously, you'll probably have worked exercises, which I like to do in latex, either in the org file or in separate files.

I've never tried to learn how to design a CPU but I think designing one on an FPGA is enough like programming that I would treat it the same as learning any other programming language or software technology. I would probably jump right into to doing practical circuits in VHDL or Verilog as soon as I had the basic knowledge needed. I wouldn't spend a lot of time on learning the theory since that will probably become much easier to understand once you get into actual projects.

> The only disadvantage is for drawings. I don't currently have a good way to capture those.

How about Markdeep? https://casual-effects.com/markdeep

I think these are good, but still too slow comparing to hand-writing.

I highlight important points, then take a snapshot of the highlight, paste it into an outliner, and add my thoughts about the point to the same item in the outliner.

My reader can also display other file types and webpages, so I treat them in the same way.

I wrote my reader-outliner but can't mention it here as that leads to angry comments, and I have been banned twice here for doing so. Look in Apple's app store under Education.

Thank you~~it sounds very interesting, I'd like to check it out. I'll take a look in the app store but might come back to bug you with the name.

Here's help. Its name is three words. First word begins with e. Second word begins with w. Look for "Academic Writing". Having to do this is so silly and frustrating.

I first finish reading the textbook (or the parts I'm interested in), highlighting and taking notes on the side bars along the way. Then I go back through that marked up pdf and extract notes from that. The reason I stopped taking notes while reading are:

1. There was mental overhead from both understanding the material / organizing it into notes at the same time.

2. On the first pass everything seems noteworthy and the bigger picture is not clear. This leads to notes that are too long and unfocused.

One other thing I've found useful is to use of a combination of highlight colors / underlines to make the process of taking notes later easier or avoid notes altogether.

Thanks man! I found the same reasons as you did and thus was unable to conform to a desired system for note taking. I also had the tendency to overestimate my abilities and time to study hobbies which adds into the frustration.

I am in a very different field so I am not sure if it's any helpful to you but here is what I do. Open PDFelement on the right side of my screen and Obsidian/Craft on the left. The idea is that my notes should be sufficient enough that if I want to review the material I don't need to open the textbook anymore. This means sometimes I copy snippets from the textbook to my notes, which is always properly indicated. Other than that my notes are basically my thoughts and ideas of how I understand the subject I am reading about. I take notes as I read and if I ever feel that I wrote too much I delete the redundant sentences. Longer notes are not always better so I try to find the right balance.

The most important advice is that you don't want to rewrite the textbook. Seems obvious, I know. Focus on how you understand the subject and write about it. My notes are in a sense small essays that are highly inspired by the textbook chapters.

As for some misc study tactics. My absolute #1 during my MA was setting a hardcore daily limit on how much I studied and during that time I made sure I am 100% focused on what I was studying. I quickly realized that being fully focused for 2 hours brought me vastly superior results than let's say giving my 60% for 5 hours. More time spent working does not equal better results.

The brain's default mode is to forget as fast as it can. That is what I remember from an Elon M interview with the badgalaxyboy mma podcast The Full Send. If you have a strong why case your brain will lower the bar to remembering.

One old approach to memorizing is the memory palace. I've heard there are institutes dedicated to studying this kind of thing.

I don't take many notes. After reading 284 of 362 pages I have set down 606 lines of text in gemini mode on emacs. I group ten pages of reading to one section. The resulting gemini file is viewable through a gemini browser. I have the option to further process my notes to org mode or Texinfo in emacs.

Captured drawings can convert to Inkscape SVG or PDF. I use Graphviz to connect the dots of ideas.

Learning is easier to retain and improves if you teach what you have learned from the Master. At a guess, short-term and long-term memory, cognition, group consciousness are keywords to search in the field of psychology and information research for more information. Marvin Minsky observed in one of his lectures on Ai that children of varying grades, stages, ages learn well from each other together.

To internalize and digest a new idea you can exercise while you think.

> Issue 1

Give liquid text [1] a try (freemium). It has a similar approach to the white page "system" you mentioned, and much more.

> Issue 2

I have this rule: The clearer the end goals, the easier it is to take notes.

[1] https://www.liquidtext.net/

Thanks! I'll give liquid text a try.

Regarding the rule, I think I do know the end goals, but I tend to follow the "full academic path" as I don't quite know what the path looks like. My end target is to implement a softcore (of a real processor) in FPGA so naturally my "instinct" told me to read textbooks for computer engineering courses after I completed the nand2tetris hardware part.

Happy holiday btw~~

Reply on news.ycombinator.com