Zettelkasten Forum


The Most Valuable Knowledge Management Principles

This discussion was created from comments split from: Appreciation for the Zeitgeist.

Comments

  • edited May 1

    With the current Zeitgeist (yes, I like using the word ;) ), it's got me thinking about the principles I value most in my Personal Knowledge Management.

    The Most Valuable Knowledge Management Principles

    (This is an open canvas. I'm very much looking the thoughts of others.)

    1. Notes that are Not Low-Value (to the Zetteler/Zettelnaut)

    • Ideally notes should evolve to having somewhat high value, but don't paralyze yourself thinking you have to shit gold every time you're staring at a blank page.

    2. Notes that are Transcribed Thoughts (or at least Annotated)

    • Value comes from taking a THING you've come across, and chewing on its essence, breaking down and digesting it—so you can re-articulate it with your own words and expressions.
    • Nodding to the "Collector's Fallacy", just because it feels good to paste large swaths of text into a new note doesn't mean it is good.
    • Exceptions to this rule would be pasting quotes; but the "best practice" would be to always add your commentary.
    • Temporary exceptions would be as something you don't want to forget, and that you know you'll go back to later and properly clean up. Don't rationalize your collecting though. Ask yourself: "Am I just being lazy?"

    3. Notes that are Future-Proofed and Non-Fragile

    • "Fragility" here means where your notes are easily broken or damaged. Maybe it's using Evernote and not realizing all their links are propietary until you've linked thousands of times and then don't want to use Evernote anymore for Reason XYZ, but your notes' formatting and links are relatively worthless. Your digital knowledge library suddenly feels locked and at risk—a terrible feeling. (Replace "Evernote" with the software of the month.)
    • Being Future-Proofed requires a largely software-agnostic Unique ID (UID)
    • DTID: The best Unique ID is the Date/Time ID because it is immutable, requires zero decisions in naming them (it's just timestamp of its creation) and, it is actually useful for chronological sorting.
    • Word ID's (WID) are enticing, but compelling arguments (detailed elsewhere) have me doubting the majority of their long-term value.
    • There are exceptions where WIDs are probably safe and have value. One example would be a person's name. E.g. "I met [[Ramakin Jones]] today at Starbucks..." Then in another note a few months later, "[[Ramakin Jones]] and I talked about saunas." I suppose having "Ramakin Jones" as its own note could work long-term.

    4. Notes that are easily Linkable

    • Linking between ideas is a fundamental way of thinking, very similar to our brains, and it's sort of the major draw of a zettelkasten.
    • Linking shouldn't overly sacrifice readability.
    • Bells and whistles are nice, but above all else, linking must be future-proofed.
  • Curious, can you give me an example where chronological sorting of link names is beneficial?

  • As @sfast put it in another thread, your notes should make sense to your future self, who likely won't have the same context as when you wrote a note.

  • edited May 2

    @Glade said:
    Curious, can you give me an example where chronological sorting of link names is beneficial?

    Benefits of a Chronological Sorting of Filenames

    Sorting by chronological order is sort of like breathing for me, it's inherently valuable and automatic. But let me bring conscious awareness to it and define a few benefits and drawbacks. (I'll also refer to this as "Temporal Context".)

    Three Beneficial Uses of Chronological Sorting

    When you're trying to find *that one note"
    Here's an example of an internal monologue that has happened to me countless times: "I remember having the thought when I was on the NYC subway at 2am, which was in November. (I scroll to around 201911010200)... Ah ha, there it is!"

    When you're trying to get back into the headspace you had at a previous time.
    If I scroll through 201408019999 I know it's probably about personal development and I can get back into the headspace I had during the creation of the notes around that time. It's like I'm re-entering an argument I left mid-conversation, like it's a simulation of thought frozen in time (like something out of Westworld Season 2). Sometimes I'm like, "Yeah man, totally agree."And other times I interject, "That's a good point 2014 Nick, but let's fine-tune it a bit."

    Using the Temporal Context to connect with other contexts of memory, like spatial and emotional.
    As shown in the first example, the temporal context (November 2019) is tightly associated with a spatial context (NYC subway, 2am). (As we know from prominent ancients who had to rely on their minds alone for almost all of their memory, spatial "mind palaces" played a major role in memory retrieval.)

    The temporal context immediately connects me with my emotional context: heading back to bed after a party, buzzed and excited about a few new ideas.

    One Drawback of Chronological Sorting

    An over-reliance on sorting by chronological note creation probably means the Zetteler is under-utilizing linking between notes—favoring the process of just trying to remember the temporal context. A lack of direct links between notes suggests more of a journal than a zettelkasten, and misses out on many benefits of the latter.

  • For a moment, let's make a distinction between two aspects of knowledge management that are, of course, closely linked:
    a) the process of interacting with your knowledge management system KMS, and
    b) the results of this process - to a large extent, these results are the notes and other artifacts that are part of the knowledge management system. (Arguably like many other contributors to this forum, I believe (in a prescriptive and, to a lesser extent, in a descriptive sense) in a coevolution between people and their "cognitive support systems" - so interacting with your KMS should not only change / augment its contents, but also the system's features.)

    When it comes to principles for a), I've suggested in this post the following:

    • Externalize: Find a lasting representation of your thoughts. Write them down.
    • Organize these documents.
    • Use reflection and metacognition: Find out where the obstacles are.
    • Use tools.

    The first two are self-evident for users of zettelkasten principles, the last two are perhaps a bit less common as an explicit practice.

  • Thank you @nickmilo22 for the excellent thinking on the matter. I have taken a different direction in my personal system so I'm curious what the ultimate drawbacks might be.

    In my (limited) experience, fully-encoded date tags are long [[2018032704123]], and somewhat challenging to distinguish visually, and I haven't yet been convinced that the extra length pays off in function.

  • @Glade

    Due to the difficult-to-distinguish DTIDs, I also avoided them for a while instead of using the [A-Z][0-9][A-Z][0-9][A-Z] pattern, the value was determined by the serial number of the note. Then I had to realize that a date-based ID has several benefits:

    • The ID is independent of how many notes you have already created because it does not indicate a serial number
    • Wherever they can be generated, you only have to look at the clock, so it will be a full-fledged note without modification, even if you only wrote it on paper
    • The identifiers always follow each other in the correct order due to the previous statement. For example, if you just wrote a note while walking that you only added to the system later, after creating several other notes, they will still follow each other consistently, in the order of your thoughts, rather than in the order you added them to the ZK
    • If you find more than one link in a note, you can immediately see how much time has passed between thoughts. If you use a generated ID, even though only one value is the difference between the two IDs, weeks may have passed without creating new notes, but this is not apparent from the IDs, only from the date-based ones
    • You can easily make chronological statistics from your notes. For a generated ID, you need to indicate the creation date in the file separately, and if you don't, you may not be able to retrieve it afterwards

    Negative side:

    • It is difficult to identify a specific note in a list because it is easier for the eye to distinguish letters than numbers.

    Because I can distinguish and remember letters more easily (the UID pattern mentioned above), but date-based identifiers (DTIDs) have more benefits, so I mixed the two systems: I replace the month and hour in the date with letters.

    // Now:
    2020-05-02 21:11:23
    
    // DTID:
    20200502211123
    
    // UID from DTID:
    2020E02U1123
    

    Apparently, the UID is very minimally different from the DTID, but the letters guide my eyes and so I can more easily identify the note by its name. Plus, the date is still in the ID, so if I’m curious about the creation date, I can read it out too.

  • @Glade said:
    Curious, can you give me an example where chronological sorting of link names is beneficial?

    For example: I create Zettel in batches because I mainly work in bigger sessions multiple times a week. Chrononogical sorting gives me access to those batches in an practical way.

    I am a Zettler

  • @sfast said:

    >

    For example: I create Zettel in batches because I mainly work in bigger sessions multiple times a week. Chrononogical sorting gives me access to those batches in an practical way.

    This seems like a "temporary workspace" issue.

    Once work is coalesced into permanent notes carrying their own context (including links to other relevant thoughts), the workspace in which they were created may not matter and should be discardable.

    But with date-ordered ID systems, that information persists and remains front-and-center forever whether you like it or not. I would like to learn to judge a thought by its merit rather than its age but that is hard to do when the date is the first thing you see.

  • edited May 4

    @Glade Can I ask what you use for unique ID's?

    Something @glade said really needs to be highlighted because it's a distinct and intrepid ideology (please correct me where I've assumed too much). Regardless of my over-assumptions, this is a worthy thought experiment so I will scout the terrain ahead.

    It sounds like Glade prefers to strip his zettels of their Contextual Breadcrumbs so he can judge them solely on their merits. Let's call this knowledge management ideology: Note Meritocracy.

    Note Meritocracy attempts to strip notes of their context so they can be judged solely on their content. These Context-Free Notes take the idea that "a zettelkasten should be readable by anybody" to the ultimate extreme: They make the Zetteler a stranger to his own zettelkasten!

    For example, a note with no Date/Time ID (DTID) is stripped from its Temporal Context, which weakens the chances of connecting it to others contexts Iike Spatial ("where was I then?") and Emotional ("what was I feeling then?").

    These Zettelnauts would be choosing to sacrifice the capability of using Contextual Breadcrumbs—those ornaments that adorn our memories—in favor of an austere Note Meritocracy.

    Monet was quoted to have said something about how he wished he was blind until adulthood, so he could look out upon the world with eyes free of preconceived notions—as if seeing it for the very first time.

    In the same way, looking at notes based on their content and not their context—without the colorings of past memories getting in the way—has a unique advantage.

    However for me, I would be profoundly devastated if my collection of notes were to lose the context surrounding their creation (eg: DTIDs).

    Even with the argument that contextual breadcrumbs pre-flavor my new experience of an old note, I would much rather maintain the richness of feelings that accompany that handicap, than to approach my notes with self-enforced amnesia.

  • @nickmilo22 said:
    @Glade Can I ask what you use for unique ID's?

    Certainly! [Here's an example](^B6pS^). My Atom plugin generates them on demand. Each one looks different: ^VEuC^, ^SpNf^. (But, here's a secret: these only look random. Each one has a date embedded in it which I hope to exploit in later updates.)

    It sounds like Glade prefers to strip his zettels of their Contextual Breadcrumbs so he can judge them solely on their merits. Let's call this knowledge management ideology: Note Meritocracy.

    Context matters a lot while working through temporary notes. But I find even more value when stripping that context. From page 41 of Smart Notes:

    Permanent notes [...] are written in a way that can still be understood even when you have forgotten the context they are taken from.

    It's not that my permanent notes become any less personal, but when they are rewritten to be free of immediate context they become far easier to link, shuffle, compare, reorder, and recombine. They become clean bricks I can use to make new mental structures, without all the mortar and chunks of old walls still attached.

    If a permanent note desperately depends on surrounding context, then to me that's a flag to write it more clearly and/or link it more deeply. Both activities force me to add lasting value to the permanent note, and increased value to the Zettelkasten.

    @nickmilo22 said:
    Even with the argument that contextual breadcrumbs pre-flavor my new experience of an old note, I would much rather maintain the richness of feelings that accompany that handicap, than to approach my notes with self-enforced amnesia.

    I guarantee I will not remember last Tuesday's contextual details in three years. If my notes depend on my recollection of that context I am in real trouble! My memory is not that good!

  • @Glade said:

    @sfast said:

    >

    For example: I create Zettel in batches because I mainly work in bigger sessions multiple times a week. Chrononogical sorting gives me access to those batches in an practical way.

    This seems like a "temporary workspace" issue.

    No, it isn't. It is just practical sometimes.

    Once work is coalesced into permanent notes carrying their own context (including links to other relevant thoughts), the workspace in which they were created may not matter and should be discardable.

    No. It is more brain-friendly to have many ways to access content or its surrounding structure. The more ways to access the better retrieval is (or recall when we are talking about the brain). Additionally, everything that provides any context adds to perspective.

    I am a Zettler

  • @sfast said:
    No. It is more brain-friendly to have many ways to access content or its surrounding structure. The more ways to access the better retrieval is (or recall when we are talking about the brain). Additionally, everything that provides any context adds to perspective.

    I thought that's what links and topics are for.

  • (Just want to point out to y'all that the comment aove by @bimlas was stuck in the Spam moderation queue: https://forum.zettelkasten.de/discussion/comment/5870/#Comment_5870 )

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

  • @Glade said:

    @sfast said:
    No. It is more brain-friendly to have many ways to access content or its surrounding structure. The more ways to access the better retrieval is (or recall when we are talking about the brain). Additionally, everything that provides any context adds to perspective.

    I thought that's what links and topics are for.

    There is not only no need to choose: We have episodic memory for a reason. The session batches are part of the episodic memory of the Zettelkasten.

    I am a Zettler

  • I just want to thank everyone here for the feedback and even pushback. It's very helpful.

Sign In or Register to comment.