How can the structure be built from the bottom up with the Folgezettel?
Clarification: although I express myself as if I were already writing my notes using Folgezettel, this is not the case, I just can’t express myself better in English.
If I understand how the Folgezettel works, it’s pretty much like a mindmap, or a HTML list (like Dynalist, Workflowy, Roam, etc.) where multiple items can be linked to an item based on a parent / child relationship. What I like about Folgezettel is that I can be sure that starting from the very first note in the Zettelkasten, I will somehow find the note I am looking for, as the structure / hierarchy of the notes forces me to put each new note in the right place as if I were placing them in a table of contents. For example, I have a note about operating systems, so I put new information related to this topic in a branch that starts with that note. The problem is that the branch does not always start from the right note.
For example, I got to know Zettelkasten, started taking notes from it, later I learned about the Evergreen Notes methodology and the concept of Personal Knowledge Management and because I felt they were related to Zettelkasten, I put them in a branch from Zettelkasten’s first note. I also wrote a structure note that summarizes these and I feel like all the others should start (as branches) from this structure note.
So for now, I don’t understand exactly what Folgezettel is all about if I create structure notes (the higher point of view) later than the (lower level) "experience" notes, so I have to go deeper and deeper in the branches to see a higher level of the big whole. I feel like I’m looking at a reverse table of contents. Can anyone explain how to use the Folgezettel effectively? I’m almost entirely sure that the question has already been asked and answered on the forum, so it’s a big help if someone gives me a link to the right topic.
It looks like you're new here. If you want to get involved, click one of these buttons!
@bimlas I'm probably not the right person to be giving an opinion on this, as I hardly understand it myself. However, I will leap into the fray anyway. Perhaps because of my limited understanding, I just keep it simple in my own mind:
Maybe to try to answer your question, I just work on creating Zettels - in a somewhat random way - whatever occurs to me on a particular day, I write. But I do spend a lot of time connecting a Zettel that I've just written to other Zettels and adding some tags. Thus the Folgezettels just evolve on their own.
Any of the more experienced people on the forum, feel free to tell me this post is a bunch of nonsense. I didn't just make it up, but of necessity, I've kept it simple, as that is how I work when trying to understand concepts (KISS).
my first Zettel uid: 202008120915
According to this, I could not express myself. I will try to explain my question through an example:
Suppose I am a young child who now knows the animals and wants to store his experience at Zettelkasten. He writes a zettel about the dog, the cat and the mouse. He goes to the zoo, where he meets the giraffe, the turtle and the seal. The latter two are not similar to previously known animals because the former were furry, but the seal and turtle were not. Based on these, he creates a structure zettel for furry animals and another for non-furry animals. Because he wrote these two structure zettels after the animal zettels, he cannot branch the seal and turtle from the structure zettel of non-furry animals. This is why I feel as if it is an inverted table of contents: structure zettes with a higher perspective are deeper in the hierarchy than what they are about.
So in chronological order, these zettels were made:
What would it look like if I used Folgezettels? And how can you "bottom up" create categories?
@bimlas I'm not sure I appreciate your question. If I'm missing the point, please ignore the following.
If I was doing this, I'd create a Zettel for each of the 6 animals listed. At some point, if I wanted to connect several animals together, I'd do so. I wouldn't bother with the structure zettels - as I mentioned before, they embed unnecessary (and often too much) "structure" into your ZK. Tags are better, in that each Zettel can have multiple tags. So I could tag "dog", "cat", "mouse" and "giraffe" with "#furry_animals" (if I wanted to). I could tag "dog", "cat" and "seal" with "#long_nose_animals" (if I wanted to). Easy enough to find the members of either group and one animal can belong to several groups. That method introduces no structural problems; the ZK stays flexible and open to other kinds of relationships through either tags or zettel connections.
Later on, if I wrote a zettel about "horses", I could add whatever tag or I could connect it to whatever other zettle I wanted. I don't care at all about a structure note, because I don't have one.
The only time I use structure notes is when I want to make a line of thought more explicit, i.e., more quickly find all the zettels in the line of thought. But I don't nest structure notes inside structure notes - maybe that is where you are running into a problem.
A ZK is supposed to be flexibly connected. Your system sounds a lot like folders, sub-folders, etc., and wondering how to move a note from one sub-folder to a different one.
Week 1 - Child creates "Note 1" listing three animals he is interested in: dog, cat, bunny. He then write a little bit of information on the note card about where he came across them, such that the note fills up. He then adds his first index entry titled - "Animals - 1"
Week 2 - Child learns about iguanas, giraffes, and leamurs while watching the nature channel. The best place to mention them and keep track of them of animals he likes would be behind the note he already created. So he goes and creates a second note, titled "Note 1a"
Week 3 - Child learns about seals, tortoises, and bobcats. He goes ahead and creates a new note listing them, titled "Note 1b".
Week 4 - Child comes across a very cool youtube series all about tortoises. He knows he already mentioned tortoises in one of his notes, so he fishes that one out "Note 1b", then creates a new note to add the new information hes come across. He titles this note "Note 1b1".
Week 5 - Child learns about vegetables. He is especially fascinated by carrots. He goes to add a note for carrots to his collection but it isn't an animal, so he starts a new note, titled "Note 2". He adds it to the index "Carrots - 2". At the end of the note he writes "tortoises (see note 1b1) love carrots"
Week 6 - Child watches a nature documentary on Netflix that has some interesting information on the savannah. So he goes and writes down some information on it. Where to put this new note? Well he mentioned the savannah in note 1a, so he goes and creates a "savannah note" titled "Note 1a1". He thinks for a second and realises that when he next wants to pull up whatever information he has on the Savannah, he won't remember to look at 1a1. So he goes ahead and adds it to the index, writing "Savannah - 1a1".
Time has passed and he has created a long list of notes listing creatures hes come across and then sub notes on the ones he likes a lot. He notices that he has a great liking for sea creatures, so he goes and creates a structure note. He titles it "Note 3", then includes links on it to all the notes he has on sea creatures from the note 1 series. He then finishes off by adding the new "structure note" to the index. "Sea Creatures - 3"
@bimlas I hope that helped a bit and didn't make it more confusing. Let me know if you need any further clarification.
@bimlas I think you might have the wrong idea about Folgezettel. They are not intentended to provide any structure because it is irrelevant where the place of a note is.
I am a Zettler
Thank you very much for your detailed answer! I was expecting exactly such an explanation!
What is not clear is that if the animals belong to the Note 1... branch, but I summarize them in the Note 3 zettel, then according to Folgezettel, these two notes are not related.
Note 1 (#Animals): dog, cat, bunny
Note 1a: iguanas, giraffes, and leamurs
Note 1b: seals, tortoises, and bobcats
Note 2 (#Carrots): vegetables
I know that Folgezettel is also just a link that is no different from the links in the text, but in this sense I don’t see any real use for it. Do you think Folgezettel is more useful in any way than date-based naming? Does this help you find a note? Or should I use register and structure zettels only?
I know that I have the wrong idea about Folgezettel. That’s why I asked how to use it correctly.
I was hoping it would help classify the notes, so when I add a new note to Zettelkasten, I’m forced to put it in the appropriate place because of Folgezettel. But if I understand correctly, it doesn't help anything in practice and was it really only important for physical, paper zettels?
I avoid using tags because I think they only make the system more complicated because we use keywords as tags, even though the text of the note will most likely already contain all related keywords, so we can use full text search instead of tags. Luhman did not have the option of full text search, so he had to use keyword tagging. But that’s another thread if you think we can talk about it.
Ah, I think I didn't make my point very clear. I mean that you seem associate Folgezettel with at least some kind of order. My point is that Folgezettel provide no order.
I quote Luhmann:
This seems to be a rather direct (as in "word for word") translation. The meaning is more like "not important" or "irrelevant" (like meaning "irrelevant" but masking it in an understatement for either humor or tone).
@Nick made the perfect example. It shows how the resulting order is not systematic but quite arbitrary. It could be done like this or in a different way. Imagine the kid learning about specific animals for a long long time and many other topics as well. It is entirely possible that a list of animals could be burried in 3a5d4 and link to some notes with just a digit (e.g. note 8) and some notes that are in the depth of his Zettelasten (e.g. note 5h3k2h5hk2h5h3k). The list of animals in note 3a5d4 gets to long and he starts another note listing not the specific species but the genera which in turn lists the species. However, the the taxanomic ranking is not as clean as it might seem.
Think of a tree that grows in reaction to the inputs of his environment: Animals eating from it, wind, sun, nutrients etc. The result can be something very beautiful:
But you can't make many useful planks out of it. Self-organising systems don't grow optimal for external use but for internal stability.
It demonstrates my point with relying on structure notes. They are more like clay that can be formed and formed and formed until you get it right. Or just make bricks and build a solid house instead of needing the creative power of Leonardo da Vinci to build a house out of trees like the one shown above.
I am a Zettler
@bimlas so the thing I did not mention about my example is that it is centered around Luhmann and his zettelkasten. The whole reason he has that weird numbering system with the arbitrary branching is because he faced organizational restrictions by using a physical, paper based system. The weird system was his clever work around.
We do not have the same organizational restrictions with a digital zettelkasten. Because of that it means that we don't want to copy Luhmann's organizational system exactly, but instead look at the outcome he achieved and the underlying principles that led to such an outcome.
To answer your direct questions:
Correct, in my opinion, people have over thought it. At the core, just link your notes together, such that when you come across a note you will also see related notes that you may have forgotten are related.
No, this idea stems from peoples misunderstanding. A lot of people think folgezettel is a structural way to represent information. That it is not, it is an organizational method to facilitate easy linking of information. We do not have this issue in digital systems, so we don't use folgzettel. Folgzettel is the solution to a physical spatial problem, there is no such issue for digital systems.
For some people who use the term differently, it might, because we commonly use hierarchies to remember information. And the "other way" people treat folgzettel is just a hierarchy or heterarchy. In my opinion this is a misunderstanding of folgzettel and the reason for it in the first place.
Register I don't think is really needed when we have search. I keep an index for my digital notes and I find that I rarely use it. I end up just using the global search, which makes sense. Index/Registers are traditional finding tools, to save you from flipping through every page of a book. But a computer can flip through every page of a book very quickly, making the index useless.
Structure Zettels can help with finding notes but that isn't their primary purpose (for me at least). They are more for showing you related information. As mentioned above, because we think in hierarchies, the structure zettel can still help you find a note if the traditional global search fails you.
@bimlas @zk_1000 @Nick @Sascha
This is a great discussion for me, as a newcomer to Zettelkasten. It's clarified a few things that were slowly simmering in my brain before but now have nicely risen to the surface. One example: I could never really understand the relevance of Folgzettel until I realized, from @Nick and @Sascha comments, that they were a feature only of a paper ZK and are not needed and are irrelevant in the digital world.
In fact, in the digital world, all of the normally used features of a ZK, such as searches, zettel connections, tags, and structure notes, all have a great degree of built-in flexibility that different people can learn to use in ways that suit their own workflow and way of thinking. Of course, we always want to see how other people use their tools - it tends to improve our craftsmanship as well.
@GeoEng51 yes, in creating ones own digital note system there are a couple of questions you want to ask:
While creating your own system occasionally reflect on
According to this, Folgezettel is really redundant. I will try to describe what my workflow is like and why I felt the need for Folgezettel:
This is how I interpret the "bottom to up" categorization.
What I felt was necessary for Folgezettel is not seen in this example, because the child writes about several animals at once, so they can be grouped and placed in the structure immediately after they are written. Suppose the child does not yet know any animals. When he gets to know the dog, for example, he writes a note about it, but he doesn’t know exactly where to put that note, he doesn’t know what context to put it in because there’s no context to it, so he can’t link to anything. That’s why he leaves it in the inbox until he finds something like it, like a cat. The more he learns about an area, the more he sees the groups and starts to see the topic from a bird’s eye view, so he can now write structure zettel on it.
Folgezettel seemed useful because without having to pinpoint the position of a note, I could place it among the earlier zettel so that I could find it later, I don't have to keep it in my inbox.
Am I doing something wrong or is this workflow okay?
Yes, I think of structure zettels similar to the table of contents in the book, because simply using full text search, I don’t always find the note I’m looking for. I don't think there is anything more accurate than links. I find the table of contents a great help in finding the right entry points (even though I use the links to navigate after finding the main topic). Here is an instructive read on the subject: The Weakness of Full Text Searching
I think I was enlightened and understood exactly what benefit folgezettel originally had: Because Luhmann tagged the branch opener zettel and because folgezettel connects multiple zettels in an easy-to-follow way, that tag not only applies to the branch opener, but also the zettels in its branch.
I think @Nick explained that too, but I only understood now. In his example, Note 1 received the #Animals tag which applies to everything in the Note 1... branch because they are tied to that zettel.
@Sascha said "@bimlas I think you might have the wrong idea about Folgezettel. They are not intentended to provide any structure because it is irrelevant where the place of a note is."
@Nick said "A lot of people think folgezettel is a structural way to represent information."
Based on the fact that folgezettels also form a kind of table of contents in Luhmann’s Zettelkasten, so I still feel like there’s something useful about folgezettel, I just don’t see exactly what it is yet.
If it is less important where a zettel goes, how could he create such an ordered tree without pre-designing it like a table of contents?
https://strengejacke.files.wordpress.com/2015/10/introduction-into-luhmanns-zettelkasten-thinking.pdf -> page 46
It looks and feels like a table of contents and that is the issue, as it leads people to thinking it is. Table of contents is a structural tool and his zettelkasten isn’t 100% structural. Sometimes it is, but often times the branches are arbitrarily, as in the branch has no structural relationship as a new chapter would in a book. The new branch of the zettelkasten is put there because he has to put it somewhere, so why not. This is also when you add a branch to the index. Because it’s not structurally related to the sequence it is being branched off of, you need some way of finding it again, so you create an index entry for the branch. If the branch is structurally related to the main branch it’s coming from, than you don’t create an index entry because it will naturally be easy to find again.
The reason you want to use the zettelkasten like a table of contents is because we think hierarchical and use hierarchies naturally for everyday cognition.
Thank you so much for your patience in trying to really understand the point with me.
I think I now understand why most people in the digital Zettelkasten ignore folgezettel and what they use instead:
When adding a new note, instead of visiting the branches for its topic and finding the one that best fits the new note, we can simply add the tag of that branch to the note itself. This connects it to other notes on the same topic in the same way as the numbering indicates the connection in folgezettel.
One last thought to summarize what folgezettel means to me based on these: nested tags.
I think it's worth using folgezettel in a paper-based Zettelkasten (in the digital world, however, the tags themselves are enough): when you create a new note that you can't classify under other notes because it belongs to a new "topic," give it an ID (e.g., Note 1) and a tag (e.g., #Animals). If you write a note again on this topic, you can also add the #Animals tag to the new note, but then each note should be included in the index for the corresponding tag, which is cumbersome for 90,000 notes. Instead, use numbering to indicate that two notes belong to the same tag, e.g. notes starting with 1... belong to the #Animals tag. If you create a note that is in the #Animals topic but is too special, place it within #Animals (e.g. Note 1a1) but add it to the index as a separate tah (e.g. #Savannah). Henceforth, Note 1... denotes #Animals, and 1a1... denotes #Savannah. This gives you a structure-like layout, but these are really just nested tags.
@bimlas I'm not sure if you are understanding me as I intended, as I'm not sure what tags have to do with folgzettel. Typically you want to place thematically related material next to each other. That is why when you write out a note, you think about where it would fit in your zettelkasten.
Arbitrary branching is just saying that not all note branches have to be thematically related to the original branch they come from. They often are, but don't have to be.
Please read this message again in light of the aboves.
If the tags in the example above are represented according to the folgezettel structure, it looks like this:
** Note 1a1 (#Savannah)
Because of this, I think folgezettel is about nested tags partly.
I think categorizing the notes is only one side of the coin and folgezettel can really help in the research process, maybe just as it appears on https://strengejacke.files.wordpress.com/2015/10/introduction-into-luhmanns-zettelkasten-thinking.pdf (starting from page 19):
Along your research, you write a separate note about each experience. If you have new experiences with a previous experience, you will start adding the new experience in its branch, as the thoughts following each other. By the end of the investigation, your experience will be "categorized" (I mean, in an easy to see through form), not just a long, linear note. Since you don’t delete any notes, you can come back to any of your experiences later if you don’t remember exactly why you rejected that particular idea.
Because there is only one experience per note, you can easily link the conclusions to a summary in structure zettel, and even if you continue the research later and draw more final conclusions, you can easily improve the summary by modifying the links.
You can clearly avoid a logical short circuit because you can see if you have described the same experience before and you can keep track of where your train of thought went from there.
If I ignore the serial number ID and think of tags, it is possible to have the (datetime) ID of a zettel appear as a tag on the zettels of its own branch and that will specify the connection between them. Moreover, if I think of links, it is enough to place a link to the parent, i.e. the previously added note, or the branch opener (where a backlink to the current note is created). With this solution, the connection between them will be as visible as if we were watching their numbering, and I think that’s what @Sascha talked about.
Perhaps identification with numbers is more effective because of the chronology: you clearly know which zettel you wrote after which + makes it clearer that old notes should also be kept. The disadvantage is that the numbered connection is technically differentiated from the links, e.g. if you want to refer to a stream of thought, you can only refer to it with the phrase "see note 123/456/7 and the next in line".
@Sascha, you referred somewhere to a post where Luhmann explains what he doesn't like about folgezettek. Could you provide a link to this please?
I am not sure what you mean. But Luhmann himself complained:
I think you are trying to wrap you head around that nested tags and Folgezettel both look like hierarchy with multiple storage possibility. But both are not hierarchical.
I am a Zettler
I don’t necessarily want to categorize, but I want to make sure I find my way to every note without the full text search. A full text search assumes I know what I’m looking for, but what if I’ve already forgotten that I also have notes about material arts, for example? How will I remember searching for it if I don’t even know I have such notes? I could go through the list of tags, but if I have 500 tags, it’s almost like there’s none because I’m probably not going to go through this list. If, on the other hand, I have a main structure note with 30 "main" topics, I will notify it easier. This is the categorization of entry points, not the notes themselves, those are not fixed into one category.
That's sort of how I look at it as well. But mostly I just use tags to access my notes - it seems to me if you have 500, you have too many (unless you have thousands or tens of thousands of notes). I do use structure notes in the same manner that you indicate - simply as an entry point into what I like to call a "line of thought".
Since I don't use "keyword" tags (like #Animals), this was just an example. Instead of tags, I collect a list of relevant notes into a zettel, for me this means structure zettel.
I do use structure notes in the same manner that you indicate - simply as an entry point into what I like to call a "line of thought".
I put structure zettels into a main structure zettel ("main index") so I can see all the topics from one place. For folgezettel, this can mean a list of notes at the top level, so Note 1, Note 2, ...
If this kind of design is called categorization, then yes, I want to categorize notes.