Zettelkasten Forum


Question, evolving ideas, contradictions: how to manage non-truth in a Zettelkasten?

Greetings,

I come to you, specialists and sages of the Zettelkasten, in a quest to understand something I cannot wrap my head around… 

I've read How to Take Smart Notes, I've practiced the method successfully for writing fiction, but there is something I cannot yet figure (or maybe need validation to implement?).

How do you manage that which you know you do not know, or do not know enough, or realize you did not know? That is:

  • How do you manage questions, which can be ongoing on life? ("How to be a good man in the 21st Century?")
  • How do you navigate and evolve forming practices? (Like: the Zettelkasten method, which you hone in through practice, and evolve insight: "this works for me, I should pay attention to that")
  • How do you walk back on ideas that do not serve you anymore, or that you realize were false? (like an early 20st Century scientist witnessing the ether being disproved, meaning all theories and constructs pertaining to it become completely irrelevant apart from scientific history)

I guess the underlying issue is "transient" content, which is a little bit at odds with the concept of permanent / evergreen note / Zettel. Now I know a Zettelkasten is personal, that truth is unattainable and everything is potentially transient – please bear with me here 🙂 I also know, in Luhmann's words, that the Zettelkasten is a "septic tank".

And yet. What would be best practices for capturing accruing, unformed ideas and questions, which would free the user by knowing they will not pollute the system down the road if they turn out to be dead ends?

Is it a simple matter of trusting that linking will isolate irrelevant content? Or should you prune links later, remove notes – but potentially destroying the path of thinking that led you to where you are now? Is it a much simpler matter of redirecting the energy of the system simply by altering the geometry of structure notes?

Any insight and especially technical approaches would be hugely appreciated. Thank you!

"A writer should write what he has to say and not speak it." - Ernest Hemingway

Zettelkasten: Bear + DEVONthink, GTD: OmniFocus, production: Scrivener / Ableton Live.

Comments

  • I guess the thing here is that everything is somewhat known as a "truth" based on a thinking and experimenting system that we humans constructed. I don't feel like there's a reason for everything in your Zettelkasten to match the truthiness truth, it just needs to make sense, even if it comes from your own head. It is your personal space for keeping your stuff contained, anyways.

    Something I do is always reference the resources that I get the information or inspiration from to write the notes. If the note doesn't contain any reference, then it means it is a personal opinion or idea that I meant to develop further.

  • @KillerWhale said:
    Greetings,
    Is it a simple matter of trusting that linking will isolate irrelevant content? Or should you prune links later, remove notes – but potentially destroying the path of thinking that led you to where you are now? Is it a much simpler matter of redirecting the energy of the system simply by altering the geometry of structure notes?

    Any insight and especially technical approaches would be hugely appreciated. Thank you!

    I wouldn't use the term "best practice" about Zettel revision since the "best practice" tends to mean personal preference. I have revised notes when that doesn't mean reassigning links. Fixing grammar and improving the wording is a good practice ("good practice" means "this is what I do").

    Occasionally, I will relink a note using a new feature in Zettlr that will update Wikilinks if you change the ID of a note--I don't recommend this unless you are simplifying unnecessarily complicated IDs, as long as the note sequence is preserved. But I am tempted to leave things as they are and let moribund threads peter out. That's how Luhmann worked. Some people revise their notes and edit out their prior understanding and struggle--forgetting history can give an overly sunny impression. It can waste time if you end up rehashing ideas abandoned for good reason.

    You did give me an idea, though. I am toying with adding a few "terminal notes" for TRUE, FALSE, MAYBE, that I don't update directly (unless I am so inclined) but that other Zettels will link to. By navigating to one of the terminal notes, say, MAYBE, the "Related Files" pane in Zettlr will show (Insidian Obsidian has a similar pane) all the Zettels that link to MAYBE. You can use hashtags for this and search for #maybe, but I find hashtags less useful than Wikilinks. I usually ignore my own hashtags.

    GitHub. Erdős #2. CC BY-SA 4.0. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein.

  • edited December 2023

    Thank you both for your answers. Doing more research, I found indeed that Luhmann never deleted any notes, which I think makes sense – a past understanding that has been refuted may become again relevant in the future, or serve as an example of a different refutation. I appreciate the idea that we should only clean the wording of Zettels – that's a clear metric, thank you. (Goes contrary to what Andy Matuschak does BTW)

    Also appreciate that the presence or absence of a reference means the idea is original or not. That's easy to implement.

    However, I remain confused as to how, in a practical way, this would work for a) questions b) accruing insight and c) refutations. How do you go about this?

    a) questions: my idea would be to have maybe different structure notes, like yearly, that would reflect an evolving understanding of crucial notions.

    b) accruing insight: like questions but on a smaller scale… 

    c) Refutations – a Zettel could point to a hypothesis: [[192312150928 If the ether exists, then it means this in physics]] and when refuted, that Zettel should point to the most current theory: [[192312150930 The ether has been disproved]].

    But in the case of questions and accruing insight, that looks messy and maintenance intensive, which is why my gut tells me there should be a better way.

    My fear is to pollute the Zettelkasten with transient content, I guess, or inarticulated thoughts, questions and ideas that get stale. But at the same time, if the Zettelkasten is a "septic tank", shouldn't these things be thrown into it in order to mature?

    I'm very curious about a technical approach to make this work without drowing the system.

    "A writer should write what he has to say and not speak it." - Ernest Hemingway

    Zettelkasten: Bear + DEVONthink, GTD: OmniFocus, production: Scrivener / Ableton Live.

  • edited December 2023

    One of the aspect I've catched in your post, maybe, is thinking that permanent mean "written on stone".
    I've made the same assumption when I first met zettelkasten, and this make difficult to start using the method at first.

    How can write "perfect notes" immediately after reading the first article about a new thing? I can't, I don't know enough at this stage about, but I can easily transform the problem in a no-issue abandoning the idea of notes as stones that I can't change or that I must consider always with the same relevance they have at the moment I wrote them.

    Permanent doesn't necessary mean immutable content, during my examination of the method in the last year I believe that "permanent" is an unfortunate term.

    I prefer the digital garden methaphor, where notes in their life intrinsically grow, can change, can evolve (or even wither, too), just like plant in a garden.

    So, the answer of your question, I think, it can be that if your ideas evolve during time, you can evolve you already taken notes. Their content,, how they are written, or their relevance too.
    In my implementation evolving, growing, refining notes, discuss them too are maybe half of the effort.
    Generating ideas contrasting already taken thoughts is a useful tool in the idea developing toolbox.

    Above all, evolving attitude is for me necessary when I don't have an initial full knowledge of a subject, I can start to take notes even when I know very little about. Over time, my knowledge increase, so I can increase the quality of the already taken notes., not only adding new notes.

    You can also maintain the history of the idea evolution, for example in a note that contains the evolving train of thought (I sometimes use an outline), or adding dates to the different versions of your thought.

  • edited December 2023

    The term "digital graveyard" might be offensive to some, but it has an element of truth: the notion of finality. I think of my Zettelkasten as a combination of garden and graveyard. Here lies idea #23922, contradicted by note #12430958 and the following references [@furd; pp 1111], [@snurd; pp 9], and [@phurd; pp 155-209]. I take a lassiez-faire attitude to my Zettelkasten. It is better to add too much than to add too little.

    Post edited by ZettelDistraction on

    GitHub. Erdős #2. CC BY-SA 4.0. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein.

  • Thank you both. Though I am… confused, but I guess I'll have to try things out.

    The "permanent" term for note is indeed misleading, but I just realized this means "permanently stored in the system", not "written in a permanent manner". Yet I find there is a difficult balance to strike between the supposedly complete state of a Zettel and the ever-changing nature of understanding and life.

    "A writer should write what he has to say and not speak it." - Ernest Hemingway

    Zettelkasten: Bear + DEVONthink, GTD: OmniFocus, production: Scrivener / Ableton Live.

  • edited December 2023

    @KillerWhale said:
    Yet I find there is a difficult balance to strike between the supposedly complete state of a Zettel and the ever-changing nature of understanding and life.

    The terminology of permanent notes is confusing. I disagree with Sönke Ahrens that a Zettel "should be written as if for publication" (Ahrens, 17–18, 43). Only Goethe could have pulled that off. Luhmann wrote telegraphic notes for himself as he got older.

    Ahrens, Sönke. 2017. How to take smart notes: one simple technique to boost writing, learning and thinking: for students, academics and nonfiction book writers. North Charleston, SC: CreateSpace.

    GitHub. Erdős #2. CC BY-SA 4.0. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein.

  • There are good comments above. I will add some thoughts even at the risk of repeating some of what was said above.

    @KillerWhale said:

    How do you manage questions, which can be ongoing on life? ("How to be a good man in the 21st Century?")

    One way is to create a question note. See the previous discussion "Do you create question notes?" (April 2020), where, among other good comments, I commented about various note schemas or taxonomies that include a "question" type, including IBIS or issue mapping, which calls question notes "issues". Questions are arguably the most important item type in the IBIS schema.

    In a more recent discussion, "Discourse graph and Zettelkasten" (February 2023), I learned that computer scientist Joel Chan recently popularized a similar discourse schema (QCE) that has a question note type too. In 2022, Chan wrote a web page titled "People naturally try to enact typed distinctions in their notes". A question note type seems to be a natural "typed distinction" to make when you start wondering about how to manage questions.

    (By the way, I find Jeff Conklin's list of question types to be helpful too. I listed them in the last paragraph of this comment in the aforementioned forum discussion.)

    How do you navigate and evolve forming practices? (Like: the Zettelkasten method, which you hone in through practice, and evolve insight: "this works for me, I should pay attention to that")

    Like anything else, I record new ideas about my note system (and related practices) in the note system itself. But I keep a separate longer document, like what Cal Newport calls a "root document", where I try to document how my note system and productivity system is supposed to work. It is like how software documentation documents how a software system works, but it's for my personal knowledge system. But I emphasize "how it is supposed to work" because it can be hard to keep the document up to date, and it can be hard to follow the principles consistently. I also have to struggle to keep the document simple, because I have a tendency to overthink/overengineer it.

    If I experienced the kind of confusion that @KillerWhale expressed in this discussion, I would note it as a question in my note system and/or as an action item about needing to improve my system. Confusion about a system is generally a symptom that my big-picture vision of the system needs to improve (original source: Enterprise Corp. Ltd. 1987).

    How do you walk back on ideas that do not serve you anymore, or that you realize were false? (like an early 20st Century scientist witnessing the ether being disproved, meaning all theories and constructs pertaining to it become completely irrelevant apart from scientific history)

    This question was previously discussed to some extent in "Processing research that's based on shaky assumptions?" (May 2021). I still agree with my comments in that discussion, which you could read as if they were a response to your question. Now I would say that what I would do with false or irrelevant information depends on how much it is integrated with other notes in the note system. If it's not well integrated, I would probably just remove it. If it's well integrated, I would need to note how it's wrong and make any other needed adjustments to the rest of the note system.

    I guess the underlying issue is "transient" content, which is a little bit at odds with the concept of permanent / evergreen note / Zettel.

    I have never used the term "permanent notes". I find it more useful to have a schema of the kind that I described in my answer to the first question above. Such a schema can help you track discourse relations like questions/issues, claims/answers/positions, reasons/arguments, etc.

    An idea that is generally related to the subject of this discussion is the property of epistemic status (see also the Urban Dictionary definition!) that some bloggers apply to their blog posts to warn people about how sure they are about their claims. I haven't schematized this in my note system as a property (i.e. as a tag), but at times I write informally about the epistemic status of a note.

  • Thank you all. I had seen those discussions but I had not taken a definite practice out of them. Time to revisit them in (profound) depth. Appreciate all the links!

    I find very interesting though that such a foundational (and important) question for a knowledge management / accruing method has no definite and ready answer, as the linked discussions show. We can clearly define a Zettel, and clearly build them when in-depth research has been made. In those other areas, there is no consensual answer. I am starting to wonder if there might be a flaw in our understanding / research of the whole thing.

    "A writer should write what he has to say and not speak it." - Ernest Hemingway

    Zettelkasten: Bear + DEVONthink, GTD: OmniFocus, production: Scrivener / Ableton Live.

  • @KillerWhale said:

    I find very interesting though that such a foundational (and important) question for a knowledge management / accruing method has no definite and ready answer, as the linked discussions show. We can clearly define a Zettel, and clearly build them when in-depth research has been made. In those other areas, there is no consensual answer. I am starting to wonder if there might be a flaw in our understanding / research of the whole thing.

    I doubt that there is even a consensual definition of a digital Zettel if we consider a large enough population of researchers, because using the term Zettel to refer to software is metaphorical, and people can and do stretch the metaphor in various ways. The proprietors of this website have their definition of a digital Zettel, tied to their software, but one could define it differently in other software systems.

    When you say "there might be a flaw in our understanding", one has to ask: whose understanding? These issues are well discussed in the published academic literature on argument technology, epistemology, and methodology. There are answers. But there are always new generations of knowledge workers entering schools or workplaces who need to figure out how to solve knowledge-organization problems, and they sometimes reinvent the wheel, as when Joel Chan in the last couple of years reinvented IBIS and called it QCE or discourse graph.

  • edited December 2023

    @KillerWhale, I noticed that recently in a comment in another discussion you revealed that you recorded a video about your fiction writing. I watched the video, and it's great. What surprised me is that you already consciously use questions in your writing process, as shown in this diagram at 2:32 in your video. Some of your questions can also be seen at 11:14 and at 13:28, where you mention federating questions into higher-level questions. I wondered: what is missing from your writing process as described in that video that would lead you to the kind of confusion that you described above?

    In the diagram at 2:32 in your video, you show a question as leading to a hypothesis/answer. The line between the question and answer is displayed as dashed/broken, not as a solid line. To me, your description in the video suggests at least two implicit principles:

    1. A question leads to one hypothesis/answer.
    2. Each question disappears when it is answered, replaced by the hypothesis/answer.

    I don't know if those two principles accurately reflect the way you use your note system. Let's pretend that they do, and let's imagine a slight change of those principles that would make your system more like IBIS, which I mentioned above:

    1. A question leads to multiple competing hypotheses/answers,1 which are then ranked in some way (the way of ranking can be different for different questions) based on the arguments and evidence for each one. The ranking can be as simple as marking one of the hypotheses/answers as the chosen one, or as complex as developing a calculus for a quantitative ranking.
    2. The question is retained in the note system with its hypotheses/answers, so that the relations between questions and hypotheses/answers (and their arguments and evidence) and other questions are an important part of the epistemic metadata of the note system. (In a previous comment, I mentioned the term epistemic status; a related term is epistemic metadata.2 I didn't notice before, but I notice now, that a good schema of note types and link types can be said to automatically provide a kind of epistemic metadata.)

    These thoughts lead me to want to give another, different answer to your question:

    How do you walk back on ideas that do not serve you anymore, or that you realize were false? (like an early 20st Century scientist witnessing the ether being disproved, meaning all theories and constructs pertaining to it become completely irrelevant apart from scientific history)

    Following the two revised principles above, "walking back" may not be the best metaphor here. "Walking back" is more congruent with the principle that a question has only one answer that replaces the question when it is answered, so if the answer turns out to be wrong then you have to "walk back" and restore the question. But if you change that principle so that a question has multiple competing hypotheses/answers, and all questions and hypotheses/answers are generally retained in the note system (i.e., in the "issue map" or "discourse graph" or whatever term you use), then you need a different metaphor that expresses that you are merely adding a new hypothesis/answer and/or changing the rankings of hypotheses/answers in some way. And this is merely the routine forward motion of working with a note system or knowledge base.

    You might object to this with what you said above:

    that looks messy and maintenance intensive

    One way to make the process less maintenance-intensive is to use fewer structure notes (which, in the video, you call "maps of content" or MOCs) and instead rely more on an IBIS-like schema of note types and link types that is automatically parsed by software. For example, if you are still using Obsidian (like in your video), you can do this using the Breadcrumbs plugin to specify link types and their relations; the Breadcrumbs sidebar will show you the notes that are related to any given note organized by kind of relation: for example, for an answer note it will show the parent question, the sibling answers, the child arguments. You can also color-code notes in graph view by note type so that you can see the logical (schematic) relations between notes in the graph (if you have a simple schema of one link type per note type). This kind of software configuration can allow you to rely less on structure notes (and spend less time restructuring structure notes) and rely instead on automatically generated views of the note system structure. (By the way, there is also specialized software that works like this in the field of government intelligence analysis.3)

    I will end with a relevant quote from an early 20th-century Zettelkasten user:

    Not once, but frequently has the general impression with regard to the causal sequence of events, with which we had started our enquiry, or which had arisen spontaneously during the examination of documents, the taking of evidence or the observation of the working of an organisation, been seriously modified, or completely reversed, when we have been simultaneously confronted by all the separate notes relating to the point at issue. On many occasions we have been compelled to break off the writing of a particular chapter, or even of a particular paragraph, in order to test, by reshuffling the whole of our notes dealing with a particular subject, a particular place, a particular organisation or a particular date, the relative validity of hypotheses as to cause and effect. I may remark, parenthetically, that we have found this "game with reality," this building up of one hypothesis and knocking it down in favour of others that had been revealed or verified by a new shuffle of the notes—especially when we severally "backed" rival hypotheses—a most stimulating recreation! In that way alone have we been able "to put our bias out of gear," and to make our order of thought correspond, not with our own prepossessions, but with the order of things discovered by our investigations. (Webb 1926, p. 418)

    Cited sources

    • Donald T. Campbell & Julian C. Stanley (1963/1966). Experimental and quasi-experimental designs for research. Chicago: Rand McNally.
    • Donald T. Campbell (1988). "Science's social system of validity-enhancing collective belief change and the problems of the social sciences". In: Methodology and epistemology for social science: selected papers (pp. 504–524). Chicago: University of Chicago Press.
    • Thomas Chrowder Chamberlin (1890/1965). "The method of multiple working hypotheses". Science, 148(3671), 754–759. https://doi.org/10.1126/science.148.3671.754
    • Frans H. van Eemeren, Bart Garssen, Erik C. W. Krabbe, A. Francisca Snoeck Henkemans, Bart Verheij, & Jean H. M. Wagemans (2014). "Arguments with structure". In: Handbook of argumentation theory (Revised edition) (pp. 633–640). New York: Springer. https://doi.org/10.1007/978-90-481-9473-5_11
    • Louis P. Elliott & Barry W. Brook (2007). "Revisiting Chamberlin: multiple working hypotheses for the 21st century". BioScience, 57(7), 608–614. https://doi.org/10.1641/B570708
    • Linda Flower, Elenore Long, & Lorraine Higgins (eds.) (2000). Learning to rival: a literate practice for intercultural inquiry. Mahwah, NJ: Lawrence Erlbaum Associates. https://doi.org/10.4324/9781410605009
    • Martin Thomas Horsch & Björn Schembera (2022). "Documentation of epistemic metadata by a mid-level ontology of cognitive processes". In: Sales, T. P., Hedblom, M. M., & Tan, H. (eds.), Proceedings of the 8th Joint Ontology Workshops (JOWO'22), August 15–19, 2022, Jönköping University, Sweden, Aachen, 2022 (pp. 1–12). Center for European Union Research. (paper, slides)
    • Christopher W. Karvetski, Kenneth C. Olson, Donald T. Gantz, & Glenn A. Cross (2013). "Structuring and analyzing competing hypotheses with Bayesian networks for intelligence analysis". EURO Journal on Decision Processes, 1(3-4), 205–231. https://doi.org/10.1007/s40070-013-0001-x
    • Pradeep K. Murukannaiah, Anup K. Kalia, Pankaj R. Telangy, & Munindar P. Singh (2015). "Resolving goal conflicts via argumentation-based analysis of competing hypotheses". In: IEEE 23rd International Requirements Engineering Conference (RE), Ottawa, ON, Canada, 24–28 August 2015: proceedings, Piscataway, NJ, 2015 (pp. 156–165). IEEE. https://doi.org/10.1109/RE.2015.7320418
    • David Rindskopf (2000). "Plausible rival hypotheses in measurement, design, and scientific theory". In: Bickman, L. (ed.), Donald Campbell's legacy (pp. 1–12). Thousand Oaks, CA: Sage.
    • Jodi Schneider, Tudor Groza, & Alexandre Passant (2013). "A review of argumentation for the social semantic web". Semantic Web, 4(2), 159–218. https://doi.org/10.3233/SW-2012-0073
    • Alice Toniolo, Timothy J. Norman, Anthony Etuk, Federico Cerutti, Robin Wentao Ouyang, Mani Srivastava, Nir Oren, Timothy Dropps, John A. Allen, & Paul Sullivan (2015). "Supporting reasoning with different types of evidence in intelligence analysis". In: AAMAS '15: Proceedings of the 2015 International Conference on Autonomous Agents and Multiagent Systems, Richland, SC, 2015 (pp. 781–789). International Foundation for Autonomous Agents and Multiagent Systems.
    • Douglas N. Walton (1998). "The inquiry" In: The new dialectic: conversational contexts of argument (pp. 69–99). Toronto: University of Toronto Press. https://doi.org/10.3138/9781442681859-004
    • Douglas N. Walton (2011). "Reasoning about knowledge using defeasible logic". Argument & Computation, 2(2-3), 131–155. https://doi.org/10.1080/19462166.2011.637641
    • Douglas N. Walton (2013). "The Carneades model of scientific discovery and inquiry". In: Methods of argumentation (pp. 181–211). Cambridge, UK; New York: Cambridge University Press. https://doi.org/10.1017/CBO9781139600187.007
    • Beatrice Webb (1926). "The art of note-taking". In: My apprenticeship (pp. 412–421). New York: Longmans, Green & Co.

    1. Discussions of the role of multiple competing/rival hypotheses in inquiry can be found in, for example: Chamberlin 1890, Webb 1926, Campbell & Stanley 1963/1966, Campbell 1988, Walton 1998, Rindskopf 2000, Flower et al. 2000, Elliott & Brook 2007, Walton 2011, Walton 2013, Schneider et al. 2013, Karvetski et al. 2013, van Eemeren et al. 2014, Toniolo et al. 2015, Murukannaiah et al. 2015. ↩︎

    2. See, for example, the explanation of epistemic metadata in Horsch & Schembera 2022 (paper, slides). ↩︎

    3. For example: Karvetski et al. 2013, Murukannaiah et al. 2015, Toniolo et al. 2015. ↩︎

  • @Andy , thank you so very much for that extremely insightful post – I really appreciate the time and thought you put into it and into sourcing everything. I also appreciate you taking the time to watch the video!

    Your naming of questions and a continuum of hypothesis as metadata really resonates with me and it helps me a lot to grasp with the problem. What I deeply appreciate with the Zettelkasten method is the way it "remembers" your path of emergence through linking, and how you can retrace that, put yourself back into that space, eliciting more remembrance and understanding that the notes themselves (which is why I think putting AI into automatic linking is absurd). This, however, clashes with the Matuschak evergreen notes and, to some extent, makes the "permanent" notes a la Ahrens confusing.

    I now think these two approaches are at odds with retracing your steps, so to speak.

    Why do I not have a problem with questions / hypothesis in writing fiction while I have one with the Zettelkasten? Because my writing notes are project notes; they are (comparatively) short-lived to my regular archive. More importantly, they are completely disposable. In Ahrens parlance, they are not permanent; they help me narrow down the project at hand, and are very often thrown away when the book is done. (Sometimes I extract value out of them, but they have no lasting power.)

    So I have no issue consigning questions in there because these questions are short-lived: they will know a definite answer (through the book being written and published), and if the answer is wrong (it happens…), there's no revisiting them anyway, and I will have moved beyond them anyway. They are project-based questions.

    Compared to this, Zettelkasten questions seem to me quite open-ended are they are potentially life-long objects of study, which I think needs a more careful approach?

    However, this continuum approach really speaks to me, thank you for that. I am starting to see a blurring of the lines between Zettelkasten and journaling, and that's an approach I have to explore — possibly with Hailey René's book, Soul Cards, which I just received – I'm wary of the marketing approach, but curious of the content. I hope this line of thought matures into something usable.

    "A writer should write what he has to say and not speak it." - Ernest Hemingway

    Zettelkasten: Bear + DEVONthink, GTD: OmniFocus, production: Scrivener / Ableton Live.

  • @KillerWhale said:
    What I deeply appreciate with the Zettelkasten method is the way it "remembers" your path of emergence through linking, and how you can retrace that, put yourself back into that space, eliciting more remembrance and understanding that the notes themselves (which is why I think putting AI into automatic linking is absurd). This, however, clashes with the Matuschak evergreen notes and, to some extent, makes the "permanent" notes a la Ahrens confusing.

    I now think these two approaches are at odds with retracing your steps, so to speak.

    Then, leave the older, obsolete notes in. I don't use the term "evergreen." I have "autumnal notes" in my digital graveyard. Luhmann's Folgezettel--a term he never used--handles contradictions. My Zettel IDs combine an initial keyword of up to five characters, Folgezettel, and timestamps in my Zettelkasten. If a note continues a thought, I increment the last alphanumeric character of the Folgezettel portion. If a note contradicts, revises, or comments on an aspect of a note, I append a letter or a number to the Folgezettel portion--letters and numbers alternate. I use another keyword if the new note neither continues, comments, or contradicts some prior note. The timestamp is always the timestamp.

    I will only revise a note to correct grammar or to improve the wording, but not to break the chain. As the comedian Jerry Seinfeld says, "Don't break the chain."

    GitHub. Erdős #2. CC BY-SA 4.0. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein.

  • I'm glad @ZettelDistraction mentioned his Zettel IDs, because it gives me an opportunity to point to a useful principle: It is not so important where you put your epistemic metadata. There is some epistemic metadata in ZettelDistraction's Zettel IDs. When I described metadata above, I was thinking of a header such as YAML frontmatter. (ZettelDistraction puts his Zettel ID in the filename and in YAML frontmatter, if I'm not mistaken.) Any format would be fine; the common principle is that there is an established schema that is consistently encoded so that not only can your future self clearly follow what you were thinking, but the metadata could even be programmatically converted into some other format or even—if you really wanted to go crazy—interpreted by some kind of semantic reasoner software.

  • edited January 3

    @Andy said:
    I'm glad @ZettelDistraction mentioned his Zettel IDs because it gives me an opportunity to point to a useful principle: It is not so important where you put your epistemic metadata.

    That's right.

    There is some epistemic metadata in ZettelDistraction's Zettel IDs. When I described metadata above, I was thinking of a header such as YAML frontmatter. (ZettelDistraction puts his Zettel ID in the filename and in YAML frontmatter, if I'm not mistaken.)

    That's precisely correct. The point of the Folgezettel is to preserve the sequence. You read the Folgezettel portion of the ID from right to left. It tells you something about the relation of one note to the next at a glance. Annotated links can do this (a good idea) but sometimes you want to see the metadata at a glance in the file pane.

    I should mention another case: when I add a Zettel with the same keyword, which isn't a continuation or comment on another Zettel. Then, I increment a section number after the keyword. For example

    LUCK.2c1a.0.24.0102 might have the successor
    LUCK.3.0.24.0102

    This is another section under "LUCK."

    How do you read this? For the record, LUCK.2c1a.0.24.0102 is a comment on LUCK.2c1.0.TIMESTAMP1, which is a comment on LUCK.2c.0.TIMESTAMP2, which is the continuation of LUCK.2b.0.TIMESTAMP3, which is a continuation of LUCK.2a.0.TIMESTAMP4, which comments on LUCK.2.0.TIMESTAMP5.

    [There are other possibilities with this ID system, but I will spare you.]

    The ID system went through several changes. You could avoid some of this complexity by opening files and looking at links. But I have a way of seeing the immediate predecessors of any note with this system.

    Post edited by ZettelDistraction on

    GitHub. Erdős #2. CC BY-SA 4.0. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein.

  • edited January 3

    Some musings perhaps better captured in a Diary.

    Luhmann had lots of "dead ends" in his Kasten; research topics that he had moved on from or were no longer relevant; sort of becomes a record of what you were thinking about in the past. Leaving old ideas in may show evolution of thought. Or could just delete them.

  • Thank you very much for all your input. You've sure given me a whole lot to think about, and I will be adhering to @ZettelDistraction 's rule of "only improved wording" on Zettels. I will try and introduce questions slowly in the system as I accept, through practice, that it's a septic tank and dead ends are accepted (as per @JasperMcFly ).

    My intent is to centralize all my explorations in the same system:

    • Quick capture (a la fleeting notes, like the Drafts app)
    • Ideation and musings (story ideas, diary entries)
    • The (more) refined Zettelkasten

    I hope to be able to have all this percolate together. A diary entry could lend itself to a Zettel, only to find that I already have written one on a given idea. Sometimes we need to re-learn the same life lessons (I know I do); I hope that this flow of diary → Zettel will improve retention in all areas of life. As I finally learn some things and make good Zettels out of them, hopefully I will not need to write again and again on the same subjects to rediscover good practices and ideas in my writing workflow.

    "A writer should write what he has to say and not speak it." - Ernest Hemingway

    Zettelkasten: Bear + DEVONthink, GTD: OmniFocus, production: Scrivener / Ableton Live.

Sign In or Register to comment.