For a long time, the following thought has been on my mind when approaching the study of certain activities/topics:
Take all available books on the topic you want to study.
Perform an initial filtering of the books – discard the obvious junk and keep the rest. For example, let’s assume that the final number of books will be 10-30.
Enter the Table of Contents of all selected books into Obsidian – either into one common file or into individual files. Canvas/Bases can be used.
All entered Tables of Contents are analyzed, and a general plan for studying the topic is drawn up based on them. Since the books are selected on one topic, some subtopics may be repeated in them, and in this case, one item of the plan may imply the study of a specific subtopic in several books at once.
Next, the topic is studied according to the plan, using all 10-30 books simultaneously.
The following questions arise:
How practical is this approach in general – is it too idealized?
How can the process of entering the Table of Contents of all books into Obsidian be optimized? Simple copy-paste is slow. In cases where the book is a scan, OCR can be used, but this, as with copy-paste where possible, is slow. Perhaps there are some book databases where the Table of Contents is recorded as metadata, and all the content can be copied from there? That would be a little faster.
What are your thoughts on organizing this approach in general? How to enter the contents of all books – individual files with content in checkbox format? How to draw up a final plan for studying the topic – also a separate file with items in checkbox format, and not come up with anything sophisticated? At the stage of drawing up the plan, it seems reasonable to place the Table of Contents of all books on Canvas to make it easier to review them and navigate them while drawing up the final plan for studying the topic. What about Bases – does anyone have any ideas?
Is it appropriate to use AI here? If so, for what exactly?
There is also an idea that, in addition to our own plan, it makes sense to turn to AI and ask it to draw up its own detailed plan for studying the selected topic. And to recommend information sources for each point of the plan, in which a particular subtopic is best covered.
Then you can compare both plans and adjust/supplement your own, as well as select the necessary information sources.
I tried this approach once before, but it didn’t quite work for me. It didn’t quite work because the preparatory stage itself takes a lot of time, and this type of study is focused on depth rather than speed, and at that time, speed and superficial understanding were more important to me than depth. But now I want to try this approach again.
The fragmentation, disorganization, and duplication of information is annoying – you have to not only study it, but also pre-select, filter, and organize it properly.
If anyone has any useful thoughts on what I’ve written, it would be great if you could share them.
Thank you.
You may want to divide the books into introductory / general ones and detailed / specialized ones, and do the introductory / general ones first to give you an overview and a better sense of your priorities with the more detailed / specialized ones.
That depends on your priorities, reading and note-taking speed, and available time. It may be worth noting that half-year college courses tend to involve fewer books than this, and even the full set of classes in that half year might not total 30 books. On the other hand, you don’t necessarily have to read every part of every book.
A simple way to answer the question is to try it.
Also if you haven’t researched ways to study things in depth, that would probably be useful. People have been doing and writing about it for a long time.
I wouldn’t expect this to be the case. How are you copy pasting and what do you consider slow? Do all the books you’re looking at have multi page tables of contents? Compared to how much time the whole course of study would take, this isn’t the first place I’d look for time savings. And the time you save may be offset by the time you spend investigating optimization. (That said, making a system more pleasant to use is of course worthwhile.)
What you are describing is similar to the “literature review” step of a Master’s or PhD project. However, in those cases, the list of things to read is far, far in excess of 10 to 30 items, and, often, contains journal articles and records from talks presented at conferences in addition to books. Yes, it is a slow process.
The obsidian plugin Bibliosidian is designed for helping one “build up a reference-backed knowledge graph PKM, one in which your ideas, writings, notes, etc. are backed up by references that are first-class citizens of your PKM, not just an external link to something”. It might be an interesting thing for you to look into?
I don’t have very good experience with “introductory” books. Because they don’t delve deeply into specific topics, the initial tasks remain unresolved, and I simply cannot use the information from them in practice. I try to avoid such books and go straight to narrowly focused ones. Even those often turn out to be insufficient, and I have to significantly refine/personalize the material.
I haven’t researched it. Are there any recommendations? Preferably books; I’m not very fond of articles, and even less so videos.
It’s just routine work that seems completely unnecessary and that I would like to avoid.
Not all books have tables of contents that take up several pages, but if my memory serves me correctly, SumatraPDF supports multi-page selection, and there are no problems with that.
The problem lies in “adjusting” the table of contents to the desired “format.” Every table of contents has its flaws, and sometimes fixing them takes a considerable amount of time, especially in large books with detailed tables of contents.
I haven’t heard about this, I’ll check it out, thank you.