Feedback on my Zettelkasten workflow

[3mbry0]

[TLDR:]
I write ideas in bullet form, each bullet containing one atomic idea. I copy & paste each idea to their own separate note, so I’ll end up with a bunch of new notes. I try to find any connections between each of them and create links to related notes at the bottom of the corresponding note. I then make a short essay below each related-note link to make sense of its connection to the main idea of the corresponding note (a note with 3 related-note links will have 3 short essays).

---

Hello.

Have just recently been developing a Zettelkasten workflow using Obsidian. I would like some feedback from you folks, especially those that are more experienced with this. So here’s my current workflow:

When I find a video or article that contains note-worthy ideas, I write all of those ideas in one note and in bullet form. This one note containing ideas from one source I call a Reference Note.

In writing each idea, I put in mind the zettelkasten principles: I keep each bullet as atomic & autonomous as I can, and each containing only one idea; essentially, these bullets will become their own zettel and will each link to new separate notes.

What I do is copy & paste one bulleted idea from the Reference Note to a new note (zettel). I create a link in the new note to the Reference Note in case I want the source. I write tags and leave space for related notes at the bottom. Then I do the same for the rest of the bullets, each having their own zettel, so by the end of it, I’ll have the same amount of new zettels as the amount of bullets found in the Reference Note. I also link each bullet to their corresponding zettels. (Filenames are: summary of idea + zettelkasten code thing)

I then open all of the zettels I’ve made from the same Reference Note and see how I can connect them (since these notes are more likely to connect with each other than with ones from other Reference Notes).

I add links in each note to related notes. Then what I do is pretty much write a short essay below each link, connecting the idea contained in the main zettel and the idea of the linked/related zettel. And I do the same for the rest.

So far, that’s pretty much my workflow. I also try to find connections from other notes that I’ve made, but I don’t really have enough of them to cause a surprise.

Concerns

Right now, I don’t think I’m doing much work to really drill the ideas in my head, writing the essays allows me to understand the ideas but it barely forces me to remember them. Perhaps you can give suggestions.
My purpose for a PKM is not really for producing content like blogs/articles (though, I’d like to be able to do so if I want to), but I just… for a lack of better words… want a bigger brain lmao, I just want to have a bunch of ideas in my head and allow them to help me whenever there’s an opportunity.

I’ve seen Shu Omi’s video regarding the topic and I’m not quite understanding the permanent notes bit, I’d consider my essays from the related notes area my permanent notes I guess, but no separate note making sense of all the connected zettels since I might add new related ideas to the zettels in the future and I’d prolly have to rewrite the permanent notes if that’s the case; I also just don’t like making long notes.

I don’t utilize backlinking at all. I find that I link two notes to each other instead of just having one link to connect both of them, but usually it’s necessary for me to write a different essay for each of the two links, plus it’s easier for me to navigate from the related notes than through backlinks. Sometimes I do create one link to connect two notes when note A is connected to note B but not vice versa, but this is rarely the case.

I also plan on utilizing the IMF workflow once zettels reach around 100 in number.

Pardon me for the long and messy post. I’d very much appreciate feedbacks. Feel free to ask for clarifications as well. Thank you!

[Kuncy]

@3mbry0

I keep each bullet as atomic & autonomous as I can, and each containing only one idea; essentially, these bullets will become their own Zettel and will each link to new separate notes.

Alternatively, you can create Zettelkasten Tags as I use for the same - ie

• ie keep ‘Video Notes’(as I call them) in bullet point form.
• use hashtags beside and inside bullet points to give context in addition to having them near the title.
• Any time you feel that a certain bullet point has to be associated with one from another note use a Zettel-tag-pair ie ( two #Z20200623113127 tags)
  or two (#2020-06-23-1132-29) tags which can be easily generated using these
  AHK

;Zettel-tag 1
!z::
FormatTime, time, A_now, {#}{Z}yyyyMMddhmmss
send %time%
return

;Zettel-tag 2
!t::
FormatTime, time, A_now, {#}yyyy-MM-dd-hmm-ss
send %time%
return

[Kuncy]

@3mbry0

I’m not quite understanding the permanent notes bit

I had the habit of extracting annotations from books and saving them. Over the years i discovered that the extractions diminish in value overtime. Because extractions are only as goos as you remember the context set up in the book.In the absence of context, extracted annotations merely become quotes.

Permanent notes are a means of proofing your knowledge.

Basically every time you’ve understood an idea, you write down an atomic note with the intent of being read by another person, viz your future self . This means you will have to write down ideas as if you are coding - simple sentences, logical connections - avoid anything that would be taxing on your future self .

[Dor]

Using bullets like this seems like a hard work way of imitating Roam.

@Kuncy

Basically every time you’ve understood an idea, you write down an atomic note with the intent of being read by another person, viz your future self .

Yes. As a first or second stage.

The essence of the zettelkasten system is repeatedly going through the notes, making links as you recognise them, and writing further notes as you develop new thoughts. And the linking should be thoughtful and deliberate not semi-automated.

Also worth remembering that remembering and thinking can be in conflict if they are not in balance. Remembering more isn’t necessarily a good thing, though some examiners seem to believe so.

[minhthanh3145]

When migrating from Roam to Obsidian, I found that ideas as bullet points really don’t play well with obsidian, from an UX perspective. In Obsidian I prefer to write more naturally in paragraphs. That aside, I don’t look tend to look for connections within newly written notes, but with existing (old) notes, in order to fit the new information into my existing frameworks.

[3mbry0]

@Dor

Using bullets like this seems like a hard work way of imitating Roam.

Care to elaborate? Perhaps you can give suggestions on how I can make the process more effortless & effective.

---

@Kuncy

Basically every time you’ve understood an idea, you write down an atomic note with the intent of being read by another person, viz your future self . This means you will have to write down ideas as if you are coding - simple sentences, logical connections - avoid anything that would be taxing on your future self .

Ok, makes sense. So the ideas contained in each bullet/zettel are technically already the permanent notes? Assuming that I have written them down with the intent that other people would be able to understand. (I do also revise each idea once I’ve transferred them into a new note; adding or editing words while still keeping atomicity)

---

@minhthanh3145

In Obsidian I prefer to write more naturally in paragraphs. That aside, I don’t look tend to look for connections within newly written notes, but with existing (old) notes, in order to fit the new information into my existing frameworks.

I’m curious, what is your workflow for extracting ideas from videos/podcasts? The reason I use bullets is because there’s less effort in jotting down ideas while I’m watching a video or when I’m dumping whatever I remember from the video into bullet points; I dislike having to worry about writing notes while I’m still watching the video, I wanna allocate some time after watching it to work with the information I’ve gathered.

Plus, because I’m still tryna wrap my head around the zettelkasten method, I wanna follow the principles as much as possible, a paragraph would prolly not follow some of those principles, especially since I’ve just started writing notes so I don’t really have much old notes to connect to. And shorter notes are just easier to revisit, I find revisiting paragraphs or big notes in general a bit daunting.

That being said, writing paragraphs can prolly help me in actually wrapping my head around the ideas I’ve gathered. I just don’t know how I can actually go about doing that: where to start; when to connect ideas; and if I ever create new notes, how would I connect them to older notes (by adding new information or additional paragraphs to the old notes?). A bit overwhelmed with the idea of writing paragraphs, but as I’m writing this, I’m actually able to make more sense of it lol

[Dor]

@3mbry0

@Dor

Using bullets like this seems like a hard work way of imitating Roam.

Care to elaborate? Perhaps you can give suggestions on how I can make the process more effortless & effective.

With Roam all points are bullets and all bullets are linkable.

Obsidian has what is available in markdown documents - ie Documents and Headings. If, instead of writing in bullets, you wrote in Headings, then they would all be linkable. Just an easier workflow in Obsidian.
What I’d then do is a second stage note as a separate atomic note, linked to the heading in the original document.

[echej]

Obsidian works better with short paragraphs like Andy Matuschak notes

https://notes.andymatuschak.org/About_these_notes

[minhthanh3145]

@3mbry0

I’m curious, what is your workflow for extracting ideas from videos/podcasts?

Same as how I treat articles, books, etc: I pause and think about the presented ideas, and try to describe them in ways that I can understand. I will also try to use concepts that already exist in my Zettelkasten to try to reframe the new ideas.

I believe you should write short paragraphs and link as much as possible to existing notes (like Andy Matuschak notes), since Writing is the only thing that matters (from How to take Smart Notes). By writing down your conceptualizations, which are actual products of thinking, rather than recording the ideas, the resulting writing is much more useful.

If you are just starting out, then I’d admit Roam is very attractive in that it allows for seamlessly putting down ideas, regardless of the quality of these ideas. Such frictionless offering may also come with a price, which is your resulting notes get more cluttered with ideas that you recorded down just for the sake of recording them down.

This especially has been my experience with using Roam, and it requires periodic check-in to get rid or transform such shallow recorded ideas into something that actually is useful.

It’s also one of the reasons I like Obsidian more, and while I advocate for block reference feature, the initial period of using Obsidian that lacks such feature has allowed me to be more critical of what to put in the vault, which is also a good thing (not to dismiss the usefulness of block reference, it’s definitely a killer feature, used mindfully).

[Andrei]

@minhthanh3145

and while I advocate for block reference feature

For block references, I use H6

[minhthanh3145]

Good to know ! I don’t have a particular heading reserved for block reference, currently I just refer by headings in general.

[3mbry0]

@minhthanh3145 Other than connecting notes, do you still keep in mind the other Zettelkasten principles (especially atomicity) while writing your paragraphs?

---

@Dor

If, instead of writing in bullets, you wrote in Headings, then they would all be linkable. Just an easier workflow in Obsidian.
What I’d then do is a second stage note as a separate atomic note, linked to the heading in the original document.

What I write in the bullets wouldn’t really be the atomic note’s file name, but rather its content, that’s why I copy & paste the bullets, and once I’ve come up with the file name after, I manually create links on the original note. I use links instead of headings cuz links don’t have to be the filename as long as I add this thing | e.g.: [[file name|what I want preview to display]], idk how I can do that with headings. Though I can imagine adding a heading above the bullet and embedding the heading to the new note instead, idk.

If I proceed with the current workflow (using atomic notes and not paragraphs), I might go with your suggestion. But I’ll definitely keep the idea of linking headings in mind, regardless if I proceed with this workflow or not.

EDIT: I think I understand now, every idea I make in each bullet I just add a heading at the beginning and then perhaps I can embed that heading to the new note, which also acts as a block reference to the Reference Note???

---

What do you guys think about this: I do what I’ve been already doing, make “fleeting notes,” I guess is the term, to quickly jot down ideas (most likely in bullet form) from a vid/article. Then what I’d do is review all of the ideas I’ve written down, relearn them & find connections, then make paragraphs on one note based on what I’ve learned & understood from those ideas, instead of making new notes for each bullet. I’ll perhaps make more than one note if some of the ideas call for creating separate notes, if that makes sense.

---

I’ve been pondering since last night about the idea of writing paragraphs as opposed to what I’ve been doing with my current workflow. I did imagine such paragraphs to look like Nick Milo’s notes from the samples he has sent in the forum, and Andy Matuschak’s notes have also popped up in my head. Before, I’ve sort of disregarded their way of note-taking into my workflow as I thought they weren’t atomic enough and the idea of paragraphs was too overwhelming for me, but I think I’ll be taking the same approach soon.

I’m currently reading a bit of Andy Matuschak’s notes, especially the ones regarding evergreen notes, an idea that I’m also currently trying to grasp along with Zettelkasten. This whole time, I’ve been thinking of atomicity as what Shu Omi described: one idea with no more than 3 sentences. From what I understand from Andy Matuschak’s view, atomicity refers specifically to having one idea even if it is described in more than 3 sentences (in this case, it’s paragraphs). While he thinks that atomic notes should not be too broad, he also thinks that they shouldn’t be too fragmented, which seems to be what I’m having with this workflow; I think I will indeed have a hard time finding connections this way.

I’ll continue reading about evergreen notes, I’m still not sure of its distinctions from the Zettelkasten other than perhaps evergreen notes are longer??? My goal right now is to find a workflow that works for me, but the more I read about the topic, I feel simultaneously closer and farther from getting to the right workflow. While I’m understanding more as I read (especially thanks to the replies from this thread), I’m also getting more overwhelmed as I feel that there are a lot to keep in mind. That said, this is prolly just me feeling overwhelmed, and the next step is prolly to just try things out, though I fear the outcome of doing things the “wrong way.”

Again, sorry for the long & messy post lol, I guess this is why I tend to get overwhelmed with paragraphs.

[Dor]

@3mbry0

Then what I’d do is review all of the ideas I’ve written down, relearn them & find connections, then make paragraphs on one note based on what I’ve learned & understood from those ideas, instead of making new notes for each bullet. I’ll perhaps make more than one note if some of the ideas call for creating separate notes, if that makes sense.

This is pure zettelkasten.

Remember that paragraphs can be very short, and can be a bullet or bullets. The heading is what makes it easy to link in Obsidian.

Atomicity is a key concept. But the atom is what you decide. You can describe it (inefficiently) at length or in just a few words. Or equations.
Luhmann emphasised quality prose, but that was with an eye to publication - and he also advised being very selective about reading.

I think you have done very well in thinking it through, my only advice would be to fit a system around your own needs rather than someone else’s perfect concept.

[Klaas]

@Dor

…… he also advised being very selective about reading.

And that makes a lot more sense than Warren Buffett’s advice to read, read, read, 1000 pages/day, which is what he has done over the year, or so he totally unconvincingly claims.