Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: How do you manage “knowledge”

news.ycombinator.com

Post reply on HN
Print view View original

21–30 of 34 posts

Re: Ask HN: How do you manage “knowledge”

#21

Post by sudouser »

sp527 wrote: post #18

I've found the opposite approach to be more useful: elimination, selectivity, and focus. For example: Read only the best books, take notes on them, and absorb them fully. Book lists are good, but they tend to grow unbounded and promote anxiety. It's not difficult to get to a book list of 300-500. Once you're there, there's little reason to update it often. The focus should instead become chewing through the list you'…

this 1000x. accept you’ll never know everything so focus and trim and trim again on your interests. for technical stuff like setting up ubuntu machines a gist github work.

i don’t use bookmarks anymore it’s just too stressful trying to keep them updated, almost a full time job he! if I don’t read it in that moment it’s gone for good.

I use Evernote for temporary stuff but then it’s gone too after I’m finished with it or uploaded to a gist.

books in kindle are cleaned regularly too

Re: Ask HN: How do you manage “knowledge”

#22

Post by Jeff_Brown »

sp527 wrote: post #18

I've found the opposite approach to be more useful: elimination, selectivity, and focus. For example: Read only the best books, take notes on them, and absorb them fully. Book lists are good, but they tend to grow unbounded and promote anxiety. It's not difficult to get to a book list of 300-500. Once you're there, there's little reason to update it often. The focus should instead become chewing through the list you'…

Ah, now I realize this post is about how to process information, not just representing and storing it.

I once took it upon myself to map as much of my knowledge as I could. (I had heart surgery and wasn't sure I'd live a long time. Turns out I have.) I would collect lists of ideas, and then stare at them until I knew which categories they belonged in, and then split them up into the appropriate category. Long sentences I would compress into shorter ones, often two or three words (and maybe some symbols in an evolving language of punctuation). It's much easier to internalize a concept when it's written in few symbols.

I just looked through my Semantic Synchrony knowledge graph for an example. I found "?is thing = ?has logic". That's shorthand for "to wonder if a pattern is an entity is to wonder if it the pattern has a logic". Kind of abstruse, admittedly, but my point is the compressed version was much easier to internalize.

To create a knowledge base is to create a big collection of mandalas. Meditating on them is rewarding -- new ideas spring from them. There is a spectrum of cognitive processing states, from empty mind, to feeling an idea form, to having the words for it, to deciding it is worth writing down, to reading things (that you or someone else wrote) earlier deemed important. They are all valid and valuable.

Discovering writing was a big deal for humanity, because writing lets you collect a lot of good ideas, and then distill them into a more powerful form. We now have the capacity (with graphs) to process information nonlinearly. This lets us move faster, theoretically processing much more information than linear presentations like books permit.

I've written a little about the advantages of nonlinearity here[1].

[1] https://github.com/synchrony/smsn-why/blob/master/nonlinear-...

Re: Ask HN: How do you manage “knowledge”

#24

Post by mjmein »

I've been spending some time over the past few months developing a app for myself and family to help with this.

Basically, what I do is:

- make a list of books that I want to read

- over time, as I acquire the book, I have a tool to split up the book into separate chapters. These chapters are then loaded into the reading app.

- every day, I open the app and it chooses a chapter from a random book for me to read

- as I read, I highlight interesting, useful or difficult sections that I want to remember/study further

- instead of showing me a chapter, the app can also show a list of highlights from a previously read chapter. During this step it asks me to create little quizzes from the notes - either in the form of cloze replacements or simple q&a. Alternatively I can also create reflection questions or daily actions from the notes.

- in addition to the previous steps, the tool then also uses spaced repetition algorithm to review the little quizzes that I created from the books I've read

- every morning it chooses 10 of the actions that I identified and asks me how I am going to implement/practice them today

- every night it asks me how I practiced those actions

It does a bit more than this, but that's the main part. I've already seen great results from doing this, as I've always found the biggest problem with reading is that it tends to be too passive, and having the review and action steps helps alleviate that problem.

Re: Ask HN: How do you manage “knowledge”

#25

Post by ssivark »

What started off as a few small lists have expanded (over ~5+ years) into a full-fledged "notebook" with pages on various topics.A lot of pages in my Zim notebook are lists of resources and tips, organized by theme (eg: speaking/presenting, things to buy, digital currency, sci-fi to read) i.e. hard to come by nuggets whose collective perspective seems valuable. I can look up things by theme whenever I seek, and also add to those pages whenever I find a new nugget. This structure also implicitly encourages spaced repetition when used for ideas/concepts [1].

One reason I like this approach is that it allows me to de-clutter my working memory, eg: number of open browser tabs [2]. It helps me avoid FOMO. As Scott Hanselman said: "It's not what you read, it's what you ignore". I keep adding to my lists over time, and this also helps me get a sense for which themes/resources/recommendations keep popping up repeatedly, which helps me prioritize what to spend time/money on.

I use the Zim desktop wiki, on linux [3]. Also, I download a PDF copy of any thought-provoking online content that I really like, since I don't want to risk the links not being around years later.

[1] https://en.wikipedia.org/wiki/Spaced_repetition

[2] I loved Firefox tab groups, and am on the lookout for alternatives compatible with the new Quantum FF.

[3] http://zim-wiki.org/

Re: Ask HN: How do you manage “knowledge”

#26

Post by branweb »

Definitely agree with the synthesize, select, focus approach, though I find software can make this process cleaner / more enjoyable. You just have to guard against it (the software) becoming an end in itself. For suggestions, check out this HN discussion from a few years back: https://news.ycombinator.com/item?id=8270759 . Pay particular attention to Ixiaus' emacs org mode comment. That's pretty close to my ideal solution.

Re: Ask HN: How do you manage “knowledge”

#27

Post by contingencies »

Finer grained changes and git. When reading online or books, notes go in git. If you don't have time to type and just want to squiggle, try a wacom tablet and inkscape. If you have time, markdown all the things. Learn tools such as graphviz[0], mscgen[1], mermaidjs[2], syntrax[3] and wavedrom[4] to assist with clear communication.

[0] https://www.graphviz.org/

[1] http://www.mcternan.co.uk/mscgen/

[2] https://mermaidjs.github.io/

[3] https://kevinpt.github.io/syntrax/

[4] http://wavedrom.com/

Re: Ask HN: How do you manage “knowledge”

#28

Post by forapurpose »

dirktheman wrote: post #4

I use Evernote for exactly this. I organized it with a couple of notebooks: Filing cabinet, Family, Work, Inbox and Templates. I have about 20 tags that I can assign to each note and I rename all notes with the date in the name so a combination of tags, title and notebook alsways yields the result I want. A couple of takeaways: - Evernote Premium is well wort hit, it converts all PDFs and images with OCR so that the…

Can you get your valuable data out of Evernote? Will you be able to read it in 20 years?

Re: Ask HN: How do you manage “knowledge”

#29

Post by forapurpose »

I'm guessing you've considered it, but a wiki is a straightforward, highly flexible solution. Some wiki software supports structured data (alongside unstructured, of course), if you want that.

In my own experiments, I've found that the speed of the interface for input is very important. There's not enough time to input everything I would like; the longer input takes, the less knowledge I capture. Many wikis are weak in this regard; input requires a lot of reloads and scrolling, mainly because of the different modes for reading and editing:

1. Load the page on which you are inputting the data. Read it to decide where the new info belongs.

2. Click edit, wait for page reload.

3. In edit mode, scroll and read the page again to find the place you identified in step 1.

4. Add the new knowledge.

5. Click save, wait for reload.

6. Scroll and read to find the new knowledge on the updated page; verify that it's what you want.

7. If it's not what you want, return to step 2.

Re: Ask HN: How do you manage “knowledge”

#30

Post by myfriendslinky »

mjmein wrote: post #24

I've been spending some time over the past few months developing a app for myself and family to help with this. Basically, what I do is: - make a list of books that I want to read - over time, as I acquire the book, I have a tool to split up the book into separate chapters. These chapters are then loaded into the reading app. - every day, I open the app and it chooses a chapter from a random book for me to read - as…

This sounds awesome. I’ve been using a spaced repetition app within my reading process for a while now. But your process looks intriguing! For for the curious here’s my current process: http://www.opensourced.me/2018/02/11/how-i-remember-what-i-r...

Can your app be made publicly? I’d love to experiment with your setup.

Post reply on HN