Zettelkasten Forum


Introduction & Teaching using ZettelKasten and Paskian entailment meshes

Hi!

I'm a Comp.Sci teacher and besides using zettelkasten as an endless source for procrastination I'm trying also to use it for preparing teaching materials. The inspiration comes from close colleagues historian father, who had this system of writing his materials on perforated index cards and then using a steel rod to 'fish out' materials for a given lecture from a giant archive drawer. Basically, he had a mechanical tagging system and could prepare a lecture in a moment's notice with it. I'd like to have that as well.

Currently, if I 'fish out' anything out of my digital slip-box (with ~500 notes) it will need significant work to be useful for lecturing. But, I got a tentative idea to fix this, which comes from Gordon Pasks Conversation Theory (of which I have only superficial understanding of). Pask arranges teaching materials into Entailment Meshes, which are build out of Topics (comparable to zettels) and arranged (mainly) into overlapping Coherences.

A coherence is set of three or more topics with the curious property that any topic in a given coherence is entailed by the remaining topics. For example, the Topics {Function, Data, Transformation} form a coherence, since concepts of Function and Data imply a Transformation (of the data), Transformation and Data imply a Function (specifying the transformation) and Transformation and Function imply Data (to be transformed).

The idea seems to boil down to this: if the student knows some of the topics, an explanation for the unknown topics is implied by the coherence.

Does anyone use any such formal structures as auxiliary organization or thinking tool in addition to a slipbox?

Comments

  • I’ve been struggling with adjusting how I once used the Zettelkasten for writing now for teaching purposes. I teach ancient history and religion.

    This is a really useful concept you are describing for capturing the problem of circularity and dependency when teaching. Is there a way that you are noting these entanglements in the zetteln or maybe in inter-linked structure notes?

    A related thinking tool that I use, I call “braids” or “hooks”. Haven’t written the concept out before.
    I developed this tool to deal with the flatness of the Zettelkasten. The strength of the Zettelkasten is that it brings a wide range of material into your field of vision all at once. This can be overwhelmingly anarchic at times. To deal with this link-overwhelm, structure notes are suggested. I see structure notes as a collection of notes (in somewhat of an order) on a given theme or topic (usually on its way to a draft but not always). I found that this was still too flat. There was still a lot of undefined space between each zetteln. Thematic proximity does not automatically produce conceptual coherence.

    I’ve used folgezetteln in a way that I track in the header of each note. Described here: https://forum.zettelkasten.de/discussion/comment/3648#Comment_3648

    I call them "folgechains" because it is not a strict use of folgezettel and one zettel can be a part of multiple folge chains.

    I also track conceptual links. These are the hooks/braids. In Zettelkasten doctrine, every note must have a link. With braiding, every note must also acquire a conceptual link (a concept note that connects two or more notes). These concepts are built from the bottom up, so to speak. I don't fashion a topic and then find zetteln that match it. I produce these conceptual connections by direct communication between zetteln.
    Now when I have a need for teaching or writing I can also move top-down.l, looking first to these conceptual links. While the zettel links and "folgechains" are flat and endless threads, these conceptual links provide hooks by which I can grasp the thread and pull it.

  • @pseudoevagrius said:
    There was still a lot of undefined space between each zetteln. Thematic proximity does not automatically produce conceptual coherence.

    @pseudoevagrius could you say more about "thematic proximity" and "conceptual coherence"? I think you are talking about finding ways to add context to a note or structure notes or a group of notes grouped in some way but not sure. "Thematic proximity" and "conceptual coherence" seem, in all the instances I can think of, to track together automatically. For example, two vastly different experiments in a field may have a low index of "conceptual coherence" but as the experiments get closer thematically the "conceptual coherence" index would increase.

    @pseudoevagrius said:
    I also track conceptual links. These are the hooks/braids. In Zettelkasten doctrine, every note must have a link. With braiding, every note must also acquire a conceptual link (a concept note that connects two or more notes). These concepts are built from the bottom up, so to speak. I don't fashion a topic and then find zetteln that match it. I produce these conceptual connections by direct communication between zetteln.

    Tell us more about how you track conceptual links. Yes, I agree, every note must have a link. On advice from @ctietze, I've evolved my workflow so a note always starts from another note or from within structure note. This causes me to search my zettelkasten for a said starting place and often leads me onto unexpected veins of notes. This way every note has at least one link. This might breakdown making impossible to start with a link when I start studying something completely new like knitting or mycology.

    Somehow I'm picturing a concept note as being the third note. Rather than linking two notes, there is a third note that connects the two. Why? (Explain thematic proximity? Explain why the two notes are linked?) I put information about a note in the note itself.

    Maybe I am completely confused.

    Aside
    @henrikenggaard over schooled us in the frequency illusion or Baader–Meinhof effect and it just happened to me. Here @pseudoevagrius talks about hooks and braids in terms of a Zettelkasten. I was in a writing workshop today and we were discussing braiding a story and story hooks. This braiding and hooks now seem important to me and I'm seeing their application everywhere.

    Will Simpson
    My zettelkasten is for my ideas, not the ideas of others. I don’t want to waste my time tinkering with my ZK; I’d rather dive into the work itself. My peak cognition is behind me. One day soon, I will read my last book, write my last note, eat my last meal, and kiss my sweetie for the last time.
    kestrelcreek.com

  • @Will said:

    Somehow I'm picturing a concept note as being the third note. Rather than linking two notes, there is a third note that connects the two. Why? (Explain thematic proximity? Explain why the two notes are linked?) I put information about a note in the note itself.

    You have my sympathies: I think of this in terms of "extracting the relationship" as well. I strongly think it's part of my vocational training in programming. In the slightly more general field of entity--relationship-theory, there's this: https://en.wikipedia.org/wiki/Associative_entity

    In practice: When you write software, e.g. a forum, there are simple 1:1 relationships. User has_a Avatar, and often the inverse is also known, Avatar belongs_to User. Then there are one-to-many, 1:N, relationships: User has_many Posts and Post belongs_to User (note the singular relation from post to user, but the plural from user to posts). And then you have many-to-many relationships, Post has_many Tags and Tags has_many Posts. The latter is sometimes expressed as "has and belongs to many", or habtm, and can only be modeled in a database by extracting a mediating entity that represents the relationship itself. I absolutely love tinkering with models and applying this kind of pattern to solve a problem, and see its utility in Zettelkasten work as well, of course, by extracting Structure Zettel from clusters of related notes to re-ify and represent the order.

    Author at Zettelkasten.de • https://christiantietze.de/

  • Not sure if this is related to these concepts, but maybe it is.

    I had been wondering if there was a way to link to a saved search and tried the following:

    • What do Shakespeare and others say about wisdom and life? [[Shakespeare OR wisdom OR life]]
    • Quotes about wisdom and life: [[202002130828 OR [202002130835]]
    • [[#Gardening AND "soil"]]
    • [["Master Gardening" AND #class-notes]]

    Both seem to work like saved searches and yield the same results when I tested them. You don't even have to have a saved search in place to use this.

    The beauty, simplicity, and power of links [[ ]] that both connect and search have many possibilities.

    Is this a way or could be a way to use conceptual links? Does this idea fit into this discussion?
    Anyone else uses links like this?

  • Thank you @MikeBraddock. I'm stealing this idea.

    Will Simpson
    My zettelkasten is for my ideas, not the ideas of others. I don’t want to waste my time tinkering with my ZK; I’d rather dive into the work itself. My peak cognition is behind me. One day soon, I will read my last book, write my last note, eat my last meal, and kiss my sweetie for the last time.
    kestrelcreek.com

  • @Will 👍. Please share your use cases. I am still trying to figure this out. B)

  • edited February 2020

    @ctietze said:
    ... be modeled in a database by extracting a mediating entity that represents the relationship itself.

    I loved reading your post, particularly the above quote sparked my old cognitive engine. A 1 to 1 relationship is formed with simple linking, 1 to many relationships are created with structure notes. habtm relationships are harder to produce in the Zettelkasten environment. I wonder what this would look like and what value could be extracted? Tags don't do it as they form 1 to many relationships.

    More fun with flow charts! Get your geek hat on!
    I'm still trying to see a use case for a concept note. Looking again at the flow diagram I see that the "concept note" would have to have an identifier thereby making it a structure note by a different name.

    @pseudoevagrius what am I missing?

    Will Simpson
    My zettelkasten is for my ideas, not the ideas of others. I don’t want to waste my time tinkering with my ZK; I’d rather dive into the work itself. My peak cognition is behind me. One day soon, I will read my last book, write my last note, eat my last meal, and kiss my sweetie for the last time.
    kestrelcreek.com

  • edited February 2020

    Im working on a more comprehensive response to these questions, but you’re already thinking along the lines I am. To put it simply, a link denotes relation. A concept note articulates that relation. This is more along the lines of a synthesis note. If I wasn’t writing this between subway stops I would link the relevant Zettelkasten blog post here.

    Structure notes for me started accruing a collector’s fallacy problem that hindered my actual drafting process. But I may be making zetteln that are less synthetic or more atomic than yours,@Will

  • I attempted an example which I might use:

    Concept Zettel:

    Zettels specific to Biblical Wisdom:

    Zettels specific to Human Wisdom

    Zettels that mention Wisdom (All Wisdom Zettels)

    I see some challenges with using search links like this and constructing search strings and formulas using multiple search clauses. It would be easy to add a Zettel that you think would match the search criteria and appear in the results and might not. With care and practice, I guess you would get better at build these search links.

    I wonder if using this would break the software agonist goal.

    I like the dynamic aspects of these search links, but not sure if it makes something simple complex, which I am often prone to do.

    @pseudoevagrius I look forward to your additional response. It may help me gain some clarity.
    @Will Your flowcharts were a big help.

  • @MikeBraddock your idea is totally awesome for me for two reasons.

    1. I didn't even know you could do this within The Archive. Great!
    2. This helps me in my own idiosyncratic use case. I read widely (cognitive science, ecological psychology, sociology, evolutionary biology, neuroscience, metaphor studies, ethics, and other philosophy) and am writing from a certain (for me logical by now) philosophical viewpoint.
      In this process of building up my Zettelkasten, filling it both with new material, but also retroactively adding stuff from already published and submitted materials or conference presentations and such, I run into the problem that I sometimes think a certain topic is already in the ZK but isn't. It is not always feasible to, then, on the spot add it -- which would allow me to finish whatever note I am working on that is now missing a part (either a link or a piece of content that I would need to look up). These SEARCH links are awesome for this because they allow you to both give some context, and also later on come back to this note and see that you still have some work to do and fill in some blanks as it were. You click the link and find out you have since added more material and can now finally link to a correct note, or integrate/migrate some content over from the newer note into this existing note. Brilliant!

    I am a Zettler, ie 'one who zettles'
    research: pragmatism, 4e cognitive science, metaphor | you can't be neutral on a moving train

  • I started running into a "collectors fallacy" adjacent problem when preparing for my comprehensive exams for my PhD. During this process I was collecting lots zetteln on, say, the Roman Empire and religion. So I had layers of structure notes embedded in each other where I was collecting zetteln I was making though reading of data on Roman emeperors, data on various religious traditions, other scholars interpretations of this data et cetera. The goal for my exam, however, was not to just list off this data. I needed to be able to synthesize it all in a meaningful way, and if possible I needed to say something novel about the information ... or at least something on the way toward a new insight. A structure note that has notes on Augustus, religious changes brought in by Augustus, some other emperors, and Constantine (and the changes brought by Constantine) is helpful, but the relationships between these separate ideas has not been thought through. They have a thematic relation, Augustus and Constantine were both emperors but I cannot assume that means the same thing for both. How they differ will be very important for my analysis. So, of course you could articulate these differences in the structure. My problem is I had layered too many structure notes so this became unwieldy. I also did not like how the relevant text to articulate the comparison between zetteln lay in a different layer that was not currently visible. When articulating a relation I want to see all the relevant text at the same time. We now have multiple windows in the Archive which we didn’t at the time.

    I landed on the concept note idea. It also gave me a GTD sense of my progress through my comprehensive exam notes. I would track this in a yaml header on each note:


    title:
    zettel id:
    tags:
    links:
    folgezettel:

    conceptual:

    The yaml allows me search for a note in various ways. I’m also free to change the title on the file and not worry about losing older links to it because I never delete what I put in the yaml. I just add to it. A note could have multiple titles or zettel ids. It usually has multiple folgezettel.

    I can talk more about folgezettel later. That is a very specific use-case for me that I'm not necessarily recommending, but I have thoughts about how that process translated to producing concepts. Anywhere I say folgezettel or folgechain, you can just replace with "structure note."

    Conceptual notes are helpful because they allow you to be more explicit with your description of the relation. They also let you link other zetteln to the relation itself, making iteratively more complex networks. The relationship itself becomes an atom to think with in the zettelkasten. Then you can abstract about relations between relations (relations between different concept notes, etc.).

    This might be better suited to what I'm trying to do with my zettelkasten. I don't want to produce a mirror of what has already been said. I want to be able to produce something new to add to the conversation. Thus I feed my zettelkasten with fragments of what has already been said, iteratively interspersed with my developing insights until I can arrive at my own references and concepts.

    In practice I've been keeping them in the same folge chain. Recently, however, I've been contemplating how these conceptual relations often represent something significant enough to branch a new chain out of the old chain. I think this will help make the zettelkasten more self referential as the structures I think and organize notes with emerge from my zettelkasten itself, instead of being imposed from outside.

    My goal is to arrive at a new statement.

    Program in Theory

    • Line -> Fold/Concept [Line = segments of folgezettel or structure note]
    • Two lines -> Many folds/concepts
    • The relation of these two lines --> Statement

    Program in Zettelkasten

    • My goal should be to make folge. These are lines.
    • Then to shrink folge to the folds.

      • Segment them (simplify the chain numbers and use sub orders) - entanglement meshes emerge
    • They should be working specifically against the abstract machines of vague categories.

    • Then explain relations between notes or sets of notes through conceptual notes (braiding)
    • Use those conceptual notes to search in the folge - what lies outside the notes? - does it still fit in the folge?
    • Once you have identified a critical relation - note and reorient. This means draw up a new folge chain (structure note) based on it.
    • if you rely too much on the concepts of others your folgechains (structure notes) can become bloated

      • Find fold (concept)
      • Make new folgezettel
      • Map that note on zettelkasten map note And relevant structure note
      • EMERGENT FOLGE = One line
      • Find another
      • these folge retain their folge status - never delete - you are always adding new relations. Never deleting.
      • When I search the folge I can see how the emergent chains relate
      • Repeat above
      • Draw a statement between these two lines
    • In this approach I am truly transitioning into making something my own oriented to my own thinking at the base layer

    
    The conceptual underpinning for this comes from my own ideas about "graph writing" that I see being discussed on Twitter. It also branches out of my inspiration from reading Deleuze and Guattari. I'm pasting below my own working out of this conceptual assemblage if that clarifies anything. Some of it includes my own personal notes which I don't want you to get hung up on but I include for as much context as possible.

    Maps

    A map is a system of two lines.

    Constraints

    • No typologies. Concepts.
    • Abstract machines produce typologies.
    • Use the least amount of words possible.

    The Mapping process is governed by the line

    • Deleuze’s philosophical process is to repeat or re-articulate the same concept of the One in various philosophers. This is the line that runs through his work.

    • Lines precede concepts

    • folds - becoming zweifalts not einfalts

    • Unfolding and disentangling:

    Virtual __________ actualization

    \~~~~<

    Possible _________ realization

    My line?

    • Another line is an equation

      • like my historical thesis equation
      • or an equation that combines (x~y>z/w)
      • like Christians + Stoics (theologians)
      • or Christians + Epicureans (ascetics) - The writing practice described in Athansius' account of the Life of Antony which could resemble the kind of parrhesiatic checking of Philodemus - you imagine a BROTHER not an ABBA looking over your shoulder - not authority top/down
    • Another line is a rivalry - Odyssey is better than the Illiad.

    Expanded lines?

    • tracking these lines - not concepts - these are kind of like the undecidable things - a line like one marking inside or outside - the things I'm not quite sure about or cannot resolve yet

      • like blood - which represents multiple lines, multiple divisions and
      • Orpheus and Eurydice is another
      • Confessing Hagiographies is another
      • what is Christian philosophy is another
      • Epicureanism and emancipation is another
      • folgechains emerge out of these themes
      • they become lines that can develop concepts

    Statements

    • Statements, however, do not occupy one system only. Statements move transversally between two systems (or more). see @deleuze1988, p. 5
    • A proposition is something that remains in one system.
    • In the transversality of statements we see the difference between a statement and a concept.
    • The statement is a function that traces a diagonal line across different corpora (unities). The statement is something we find when we break things open– unearth– archaeology. Relieve.

    • A statement represents a curve between two individual points. The statement is what actualizes (or makes possible) their relationship. The statement makes possible their proximity.

    The Curve of the Graph

    • The goal seems to me to represent the curve of the graph (that is the statement) by assembling words, phrases, and propositions in a certain way. More than my thesis, the statement is my outline. The statement is also their outline.
    • The concepts that we use to traverse the distance between two points is a subset (or part) of the statement.
    • When we break open words, phrases, or propositions we may find statements within them. Within these objects lay what makes them possible as objects to our senses– what makes them visible. We assume a metaphysical object permanence to objects, but we need to recognize how discourses make visible what is otherwise not visible.
    • The objects themselves are negatives to which we cast the light of discourses to draw outlines.
    • A concept is a word's signified. It is a variable external to the word. Statements, on the other hand, possess their own sets of concepts (which Deleuze refers to as "schemata"). Such schemata are found at the intersection of systems.

    Writing Through a Series

    • The game becomes setting the objects in a series and then tracing the distance between the objects with the least amount of words possible.

    • The novelist Roussel maps out charts or polarities of spaces between things:

      • characters and plots
      • whale to small island
      • whale bone in corset to Spartan
      • duel between two adversaries
    • I too want to make a thought machine like the zettelkasten work like this

    • Identify the notes (d and p)

      • concepts draw lines TO and between notes
  • @pseudoevagrius this is magnificent. I am saving this off for more pondering. You articulated many great ideas and thoughts. Very exciting. Thank you! 👍👏 😎❤️

    One thing you helped me to realize and synthesize is the difference between notes and zettels. I came across this last night after finding the link in one of Christian's comments here on the forum:

    The split between “Thoughts” and “Notes”, this distinction between what others have produced (inputs into your creative process) and the syntheses which follows (what you do with it), is really at the core of academic and artistic work. - from Douglas Barone - http://dougist.com/2009/08/file-system-infobase-manager/

    Seeing the distinction of

    • notes as what others say - input
    • thoughts or zettels, what you think and say - output

    And understanding and accounting for it in your workflow, processing, and Zettelkasten. I sometimes see these concepts but don't get until I get it.

    @pseudoevagrius said:
    I landed on the concept note idea. It also gave me a GTD sense of my progress through my comprehensive exam notes. I would track this in a yaml header on each note:

    title:
    zettel id:
    tags:
    links:
    folgezettel:

    The "concept note idea" turned the light bulb on for me. 💡

    I hope many others read your post. Maybe @ctietze should ask you to do a guest post. Again this was very helpful. I look forward to thinking about this more. Thank you.

  • I second the idea of a blog post about this. Brilliant!

  • Is there a way to exclude search terms when doing a boolean search in the zettelkasten?

  • Ha sure enough! Thank you! The one time I don't obsessively follow every post in this forum... This idea has really catalyzed a lot for me today. Also I'd be happy to synthesize my approach into a more legible blog post. This is such intellectually nourishing community. Would love to give back any way that I can.

  • For reference, here's a documentation of the feature: https://zettelkasten.de/the-archive/help/#can-i-use-boolean-search

    Author at Zettelkasten.de • https://christiantietze.de/

  • Hi @ctietze,

    Do you think a distinction has to be made between a concept note -- one that explains connections -- and a structure note? For instance,

    1. Top level structure note = Overview Note
    2. Structure that serves as a running outline of a new idea = Structure Note
    3. Structure that aims to explain a connection = Concept Note
    4. Content note = Zettels

    What do you think of this?

    Sincerely,

    Amos

  • When you say "concept note" I think of notes that focus on and explain a single concept. Much like Wikipedia's approach. So you seem to have something else in mind for that?

    Author at Zettelkasten.de • https://christiantietze.de/

  • Based on my definition, concept notes are higher-level insights that encompass two linked zettels.

    As an example, say I have two notes:

    • 202005150607 David's number has a real part
    • 202005150608 David's number has an imaginary part

    Then a concept note would look like:


    Title: 202005150609 David's number is complex
    Tags: #david #number #complex #mathematics

    [[202005150607]] David's number has a real part
    [[202005150608]] David's number has an imaginary part


    The note would be much smaller than a "real" structure note, although I do not know the distinction between a "real" and an "unreal" one. I hope you get what I mean.

  • edited May 2020

    @ctietze As I've been learning more about databases, both relational and graph, I'm realizing that what we're doing in essence is building databases with predominantly many-to-many relationships, which are the most cumbersome to model.

    I've looked into graph databases like Grakn and Neo4J and it's informed how about I think about my ZK. Basically we're all making graphs, and I think we can use what people have learned about knowledge modeling in graph databases to think about how we use and organize our ZKs.

    I wonder for some specific domains with known schemas, it might be worth working directly in a graph database.

  • edited May 2020

    Thanks all for the (as ever) stimulating conversation about ideas I deeply care about. This is another example of why this forum is one of my favorite corners of the internet. As I work through some of the implications of this fascinating thread (finding echoes here and there of David Clear's notion of "connecting notes"), here's my own effort to draft a summary elaboration of @pseudoevagrius' notion of hooks and braids for my own ZK.

    If I misrepresent or misconstrue any of the conceptual architecture involved here, please do let me know.

    Started ZK 4.2018. "The path is at your feet, see? Now carry on."

  • I found this thread in a web search. Here's a belated theoretical contribution to the discussion, going back to the original question of @discordian: "Does anyone use any such formal structures as auxiliary organization or thinking tool in addition to a slipbox?"

    A general term from computer science that I associate with such "formal knowledge structures" is ontology. Domain-specific ontologies are created to satisfy the need for structure in various domains. There are scholars who study this, and some familiarity with their work may be helpful when creating one's own ontologies.

    There is a recent overview article about ontologies in the Encyclopedia of Knowledge Organization: "Ontologies (as knowledge organization systems)" by Maria Teresa Biagetti. Notice Figure 1 in that article, which shows a spectrum of complexity of ontology types where "a set of text files" is low-complexity and "a set of general logical constraints" is high-complexity. In another article in the same encylopedia, "Knowledge organization system (KOS)" by Fulvio Mazzocchi, Figures 2 and 3 show a similar spectrum of types of knowledge organization. I have found these diagrams useful to keep in mind when thinking about how formal or complex my knowledge organization needs to be.

    Check out the list of other articles in the Encyclopedia of Knowledge Organization: a few of the articles are interesting and relevant, and I'm surprised that a search shows that the encyclopedia has never been mentioned in this forum before.

    In the original post above, @discordian was inspired by the entailment meshes in Gordon Pask's conversation theory. A quick web search for ontology + Pask + "conversation theory" shows that some people have spoken of the same idea as a kind of ontology: for example, the library and information scholar David Lankes said: "Memory is the agreements and the relationship of these agreements. If you want to impress your friends, you could call memory a 'knowledge representation' system like ontologies and semantic networks. If you have no friends, think of it like a map of the stuff you know." The article "Bootstrapping knowledge representations: from entailment meshes via semantic nets to learning webs" by Francis Heylighen creates an ontology based on Pask's entailment meshes: "Entailment meshes are then extended to entailment nets: directed graph representations governed by the 'bootstrapping axiom', determining which concepts are to be distinguished or merged. This allows a constant restructuring and elicitation of the conceptual network. Semantic networks and frame-like representations with inheritance can be expressed in this very general scheme by introducing a basic ontology of node and link types."

  • @Andy Thanks for the links, I didn't know about the ISKO and its IEKO -- there's a lot of potentially interesting stuff to unpack and references to follow!

    Author at Zettelkasten.de • https://christiantietze.de/

Sign In or Register to comment.