# Can someone shed some more light on the meaning of "meaning"?

Since @sfast and I share a flat, we inevitably talk with each other more than both of us would like to. In a recent conversation about all the current Folgezettel and ID shenanigans, I wondered if it's our non-native-Englishness that gets in the way at times, e.g. when it comes to the use of "meaning", as in "meaningful connection".

Here's what confuses me.

In my 2015 post about Different Kinds of Ties Between Notes failed to make a fully comprehensive list, but addressed a couple of "ties" between notes. The whole post seems to be rather uncontroversial. That's part 1 of the confusion.

Here's a TL;DR of ties, sorted by strength of the tie, and slightly expanded:

1. Fetching a random note (see script); maximum serendipity, but in fact no tie at all. Your excellent brain will still try to look for sense in the results. (Not part of the post above.)
2. Word similarity. You get to this via full-text search: the results share the search string, that's it. (This point is scattered around comments and posts and not part of the post above.)
3. Juxtaposition and sort order in a list, e.g. on the file system (your file naming scheme influences this; through the sequence of notes next to each other, you probably will automatically look for patterns, as humans tend to do, and assume what's close to each other belongs together)
4. Tags/keywords; they are the result of a conscious act of grouping notes under a common term, especially to distinguish (accidental) word similarity from deliberate grouping
5. Categories are stronger than tags, because they are exclusive; a note can have and belong to many tags, but a note can only have 1 category
6. Hyperlinks are 1:1 connections between notes; they are deliberate (unless you use an automatic backlink-insertion-script)

Now you can fail at hyperlinking, like I did so many times in the past. Confer [[XYZ]] is not a helpful hint in practice, I find. There are cases where I can see from the title that the linked note is making a case for the opposite positon of whatever Zettel I'm looking at. More often than not, I'm confused what kind of relation I saw back when I added this. I'm glad I dropped this lazy way of linking notes in favor of adding a bit more context, like See [[XYZ]] for a discussion of the opposite opinion that bananas don't belong in bread.

I talked about Folgezettel in the post above, too. My argument was that you end up with a sequence where you can manually insert elements between existing elements. An automatically increasing numerical ID does not allow this; there, new stuff comes at the end. Date/time IDs, too, only go in one direction: forward. An effect of using Folgezettel, literal: note sequences, is to affect the juxtaposition of notes and the sort order of files. But since the placement is also deliberate, they fall between "Juxtaposition" and "Categories" on the scale at the top, I'd say. (Is this uncontroversial as well?)

Now part 2 of my confusion: the kinds of ties were uncontroversial, and their different strength of connection mentioned here on the forums from time to time, like "links form networks, tags only clusters, so links make a stronger connection". From my understanding, the strength of the tie is directly proportional, if not identical, with the meaning of the connection. Examples:

• A tagged cluster of notes is more meaningful than search results
• A random note finder will point you to any note, but it is ... random. The means of producing the result isn't meaningful, although our subsequent interpretation may be.

When the aim is to store connections, you won't say: "Ah, the random note finder did so well this time, I'll surely get this exact result and have the same idea later again!" -- Because by the nature of pure chance, you may never get this lucky again. So you preserve the connection. Create a "Banana Bread" note that links to both "Banana" and "Bread", for example, and then put this note in a list of tasty recipes.

So here's my request:

I fail to see how one can (1) accept different kinds of ties and a relative order of strength between them (even if one didn't accept the order presented above), but (2) not think of the kinds of ties to be ordered "by meaningfulness" in a similar way.

• Can anyone explain or elaborate what I might be missing?
• Do you think the order by strength is different than the one I presented above?
• Do you think the order by meaningfulness of the connection itself (not a post hoc interpretation/sensemaking) is different from ordering by strength?

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

«1

• Just when I get out to do other work, you suck me back in.

You are trying to understand the concept of meaningfulness of ties by relating it to strength of ties. But what do you mean by strength of ties? I actually don't understand what that phrase is supposed to point to.

In your post, "Different Kinds of Ties Between Notes" you write:

A connection or association is a stronger tie than mere juxtaposition. It’s stronger because it is voluntarily.

Does strength mean voluntariness? That seems wrong to me. To take an analogy, I might choose to spend time with a friend, but I still have stronger ties to family members I was involuntarily put in relation with. Voluntary ties might tend to be stronger in general, but I don't understand why voluntariness would equate with strength.

At one point in time, you have decided to connect A and B. Tags or keywords are the weaker kind of connection. Direct links are strongest.

Aren't tags just as voluntary as links? Both appear on a note only because I choose to add them.

It seems to me that you may be conflating a couple of different, albeit related, concepts.

• whether the connection is accidental (random note), incidental (i.e. a side-effect of individual labels on individual notes that happen to be the same label, but because of how you processed each note individually, not because you intended to label these two notes in a pair, together) or intentional (you consciously intend to connect these two notes together)
• whether the connection is more exclusive (a note-to-note link connects only two notes) or less exclusive (tags can connect large groups of notes)
• whether the connection reflects stronger or weaker similarity of content (which I actually have no idea how to measure)

To me, these are three different types of distinctions, not three aspects of the same distinction.

The way you list the connections from weaker to stronger in your post suggests that similarity of content correlates with exclusivity of link, but it's not clear to me that that has to be the case.

"links form networks, tags only clusters, so links make a stronger connection"

Why "stronger" rather than just "different"?

I feel that you are trying to rank meaningfulness as if it could be numerically measured, but I don't understand how you can determine the "amount" of meaning. Instead of "amounts" of meaning, I think about different types of meaning, which unfortunately, that leads to a much messier system because different types cannot be easily ranked or ordered.

• So if I understand you correctly, you're trying to understand when a link in a zettelkasten is "meaningful". Your understanding is that “”the strength of the tie is directly proportional, if not identical, with the *meaning of the connection.”

Firstly, Let us now consider the goal and the elements of an zettelkasten system. Understanding the goals and structure of a zettelkasten helps us understand what the purpose of interlinking notes is.

context

A zettelkasten consists of:

1. A collection of notes, each containing information on a single idea, concept or piece of knowledge. Each file also contains a reference to the original source of this information. This is the content of a zettelkasten
2. A system of interlinking these notes. Each note contains information about its relationship to other notes. There are many systems of interlinking notes, and most people use more than one. Examples of these ties are tagging, hyperlinks and structure notes are examples of these systems. This is the connectivity aspect of a zettelkasten; focusing on the relationships and context of the notes.

A zettelkasten system has three purposes:

The first goals is Knowledge Management. through learning and storing new information. This is mostly done through reading, writing and storing your interpretation of another person’s work. The active processing of the content aids understanding.

This information generates the backbone of your zettelkasten: the content layer of the zettelkasten. To do this function properly a zettelkasten needs to be organized so that the information is accessible and easily retrieved. So when it comes to this goal the connectivity of a zettelkasten is essential for supporting the efficient and effective storage, retrieval and organization of notes.

The second goal of a zettelkasten system is Generating Insight. Organizing information creates a context. This context helps us how we need to understand and use this information. When you start reading a source you don’t start as a blank slate. You start relating it to other information you already know. You automatically start relating this information to your existing knowledge. You will, for example, start to compare related concept and ideas, , gain in dept knowledge in a certain topic though the views of many different people. The process of combining ideas and concept generates new information, questions, ideas and concepts.

Here is where a zettelkasten becomes truly exceptional because it facilitates the generation of new insights. The connectivity aspect of a zettelkasten supports this goal though the encouragement of interlinking notes and ideas. you are forced to reflect on how all these notes relate to one another. .

The final goal of a zettelkasten is the application and communication of the (newly found) insight. ultimately the goal of learning and processing of information is to help solve problems though better understanding of these problems. So for insight to have any use it needs to be applied or communicated to others who will.

“Strong” links are not necessarily meaningful

When you talk about having strong a connection between notes, you seem to be mostly focused on the number of notes associated with this relationship. You seems to consider the relationship strong when a note connect to fewer other notes, and weaker hen it refers to many other notes. You also seem to suggest that strong links are desirable over weaker links.

But as you rightly point out, you can always imagine relationships between different random concepts (such as banana and bread). But these relationships do not support the goals of a zettelkasten. IAn exclusive relationship between two notes (a strong link) is not necessary effective for creating a meaningful link. They also can add more noise to the system, resulting in “meaningful” relationships between notes. Thereby hindering the purpose we are trying to achieve in the first place.

Moving on now to consider a “meaningful link” in a zettelkasten system. There is no agreed definition on what constitutes a meaningful link. But for me a meaningful link is a link that supports the three goals of zettelkasten. Therefore a good link needs to:

• Facilitate efficient and effective retrieval of information: This is often done through structure notes, indexes and tags.
• Facilitate efficient and effective storage of information.
• Add context to an idea or concept , including nuances and competing views
• Facilitate insight through linking the concept with related concepts
• Facilitate the application of communication of new information
• edited April 2020

When we talk of these things and in terms of how they are used and what they represent and usefulness and meaningfulness, sometimes we get a focus that makes sense to us. These things, others will see, use, and interpret differently. I will try and explain.

Take search, for example. It easy to see search in terms of strength and meaningfulness as greater than or less than the strength and meaningfulness of something else, but I would say that is only true in the view you have or take.

Using the Ominbar and search operators, I can build search strings and conditions and find anything with any degree of focus or broadness. I can then take that entry in the Ominbar and put it double brackets as a search link and retain it in a note or as a saved search. I can also suggest that it doesn't fit the view of strength and meaningfulness that you ascribed to search.

If we had no other tools other then the Omnibar and the search, I believe I could accomplish any and all the strength and meaningfulness you outlined. It is better to have those other useful tools, but strength and meaningfulness come in how knowledgeable and well you command your sword.

My point isn't that one view works or doesn't work or is better or more useful. They are just different. Users use tools and functionality differently either because they understand them differently, are more proficient with them, or they stumble on an approach that doesn't fit the model or view that hasn't been a ascribed to it. Who really knows for sure?

The fantastic thing about these tools and functionality when used with the Zettelkasten Method is they all work together or separately or mixed and matched or willy nilly. Everyone's mileage varies. It sounds like a cop-out of the discussion, but it isn't. It is very true. There is incredible diversity.

Thinking about strength and meaningfulness helps to increase understanding and helps to reveal best practices, but being firm in a position or point of view is another example of what I am trying to articulate. Having room and openness for the other possibilities adds to the strength and meaningfulness of the toolsets we have available.

I have another example of what I am thinking. The blog and forum here at zettelkasten.de. When I first began to lurk here, I was surprised at the openness for other tools for using the Zettelkasten Method that you and Sascha permitted and encouraged. I saw lots of competition around the internet for building and making tools and talking and teaching how to create second brains. Some of that competition may have been for-profit or approbation or whatever. You guys had an open door inviting policy that was like yeah sure you want to talk about some other software great. It was clear that you guys believe the more we think and talk about the ZK Method together, it will improve all of us. It reminded me of the Grateful Dead's policy of letting and encouraging fans to record shows and performances. Music executives were like you guys are nuts to let you stuff be readily available for pirating. Well, the Grateful Dead success speaks for itself, doesn't it?

In the beginning, I was surprised at the openness and worried it would impede the acceptance of The Archive. My point is my view initially matched a music executive type view as opposed to your and Sascha's Jerry Gracia view.

I interpreted the shenanigans experience as, "Oh no, some of our monkey friends have jumped over the fence and are loose, wild, and having crazy ass fun. Dammit, now we need to go and chase them and make sure they don't get killed."

If the grass isn't greener and the bananas aren't better, we will climb back over the fence and come home.

All in all, it is just a different view of the same data. Which is more meaningful or stronger?
Honestly, I have no idea.

If I have to make a point, it would be, changing or refocusing a view often opens other possibilities.

Thoughtfulness can be applied when something is expressed or when something is interpreted.
It is a very difficult skill to master. I would encourage you and Sascha not to second guess yourselves. I don't find your "non-native-Englishness" interfering - at all. You both apply thoughtfulness generously and often, Sascha in his unique special kind of way.

I hope we on the interpretation side do as well.

I will end similar to cobblpot.

Why "shenanigans" rather than just "monkeys just want to have fun'?

• @ctietze said:
Here's a TL;DR of ties, sorted by strength of the tie, and slightly expanded:

1. Fetching a random note (see script); maximum serendipity, but in fact no tie at all. Your excellent brain will still try to look for sense in the results. (Not part of the post above.)
2. Word similarity. You get to this via full-text search: the results share the search string, that's it. (This point is scattered around comments and posts and not part of the post above.)
3. Juxtaposition and sort order in a list, e.g. on the file system (your file naming scheme influences this; through the sequence of notes next to each other, you probably will automatically look for patterns, as humans tend to do, and assume what's close to each other belongs together)
4. Tags/keywords; they are the result of a conscious act of grouping notes under a common term, especially to distinguish (accidental) word similarity from deliberate grouping
5. Categories are stronger than tags, because they are exclusive; a note can have and belong to many tags, but a note can only have 1 category
6. Hyperlinks are 1:1 connections between notes; they are deliberate (unless you use an automatic backlink-insertion-script)

I understand your points completely. I followed along easily and agree. (I need to watch out for confirmation bias.) The meaningfulness of a 'connection' is its strength. How can it not be? Given any type of connection, a more meaningful connection has more strength and value. It is obvious to me a less meaningful connection, like some random reference, has a weaker value. These connections are how we discover answers for questions we bring to our zettelkasten.

All connections propose an answer to a question. Some answers (connections of all types) are relevant therefore meaningful and some much less so. Both meaningfulness and strength are relative and on a continuum from weak, and nonsensical to powerful eureka moments. You outline the possible types of connections and their meaningfulness is mostly right on. I'd put "Word similarity" higher because of the power of full-text search. I'd add a couple of types of connections that have more or less strength than those already listed.

2. Post-hoc Hyperlinks - connections discovered later, sometimes much later. To signify the meaning of the link I place mine in the YAML footer
3. The number of inbound and outbound hyperlinks - currently not quantities except visually. More of either would indicate meaningfulness and its strength.

There are no shortcuts., all the labeling and naming for meaning is of no help if we don't take the time to look at the notes connections to see what they contain. Shortcuts might work with a small or very focused (1-3) categories of notes, but is unwieldy when your notes number in the thousands and categories number in the dozens.

Will Simpson
“Read Poetry, Listen to Good Music, and Get Exercise”
kestrelcreek.com

• edited April 2020

Meaningfulness and strength is subjective. Change the user or the observer looking at the same data and whatever assignments of meaningfulness and strengths would/could change.
Or change the user or observer to your future self. I don't see how you could pin it down using or effectively utilize it with code or an algorithm.

They are relative as well. Depends, who is at the keyboard. Are they blind? Are they deaf? Do they have Parkinson’s Disease and can't type and use dictation and accessibility features?

I am not saying the items in the list are not useful. I am saying the more ways you have to reach and view data the more meaningfulness and strength can be achieve.

The more options and choices the user or observer has and can learn and understand and use the the more likely they achieve flow.

Some may use tags and be tag masters, some may use direct links, some may have great search skills, some may use the list you outline and succeed and find flow.

• I feel that you are trying to rank meaningfulness as if it could be numerically measured, but I don't understand how you can determine the "amount" of meaning.

Quantification is not necessary for ordering entities. In German, we call this something like "clear case with unclear boundaries". There is some similarity to ordinal data: https://en.wikipedia.org/wiki/Ordinal_data

I think we all agree that Bill Gates is moraly a better person than Adolf Hitler. But we could also agree that moral goodness is something that will never be quantifiable.

More later when don't have to fear my girl friend of finding out that I am doing something for the Zettelkasten Project on sundays..

I am a Zettler

• @sfast, I agree that that quantification is not necessary for ordering things in all cases. But criteria are often necessary for ordering things. To take your Bill Gates example, note that our ability to rank extreme cases on a spectrum does not mean that ranking on that spectrum is generally possible. For example, we feel confident saying that a person who murders 100 people is worse than a person who selflessly saves 100 people. But that doesn't tell us about most cases, which are much more complex, such as whether a person who saves 100 lives but intentionally kills five people to do so is more or less moral than a person who intentionally saves two lives but mildly injures 10,000 people in doing so.

Re: the ZK, the fact that we can agree that a completely random link is less meaningful than intentionally created link between two notes does not really tell us anything about whether a link is more meaningful than a tag or a category. Or at least it is not intuitive to me that it does, and I would need criteria that disambiguate the concept of "meaningful" to help me form an intuition about it.

• whether a link is more meaningful than a tag or a category

There is no difference between a tag and a structure note.

I can have a structure note titled "game consoles manufacturer" and link from there to some other Zettel: 'Atari', 'Sony', 'Nintendo'. I can accomplish the same with a tag #gameconsolesmanufacturer and tag those three notes with it. In an index of my tags I can have a description that declares any amount of meaning I want.

It's just a matter of how I implement those links on a technological level. (Tags even give me backlinks automatically.)

• edited April 2020

Lots of interesting replies so far! I'm curious to see how the discussion evolves; the latest reply by @Perikles strikes me as misleading, though:

There is no difference between a tag and a structure note.

The one, crucial difference is that tags don't impose a sequence. When you write a structure note, you do. You cannot not order the content of a structure note.

Most computer applications apply some kind of sort order, so when you search for a tag, you get the results in an order. But that's just an accidental display property, and not essential to the tags. With tags, you create unordered sets. You form clusters. (Like children mix playing cards on a table: lay them all out flat, then stir with the hands, then make a pile again.) If your app of choice didn't sort the results by filename, modification date, size, or whatever, you would end up with a randomized sequence for every request. That's what tags are: mere sets, umbrella terms that don't sort their elements, and not sequential lists.

It's like saying that a huge box of loose LEGO bricks is no different from LEGOLAND in Denmark.

Sure, you can forego imposing order in structure notes and regularly shuffle all lines in the file. You can demote a structure note to be a messy collection of stuff. You can drive a bulldozer through LEGO Land, and voilà, the great equalizer showed how similar the box of bricks to LEGOLAND was all the time.

But my point is that to write a note, a sequence of letters and lines, you will have to put them in order. Even if you try hard not to.

In the end, if you don't structure the content, you don't work with a structure note. If you slap a new link at the end of a file without any care for how it should fit in, maybe you're creating something akin to a buffer note: an unordered collection bucket for later processing.

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

• @Perikles said:
There is no difference between a tag and a structure note.

Just because two technological implementations have the same formal or logical structure doesn't mean that there is no difference between them, especially when considering a real person implementing a real workflow. For example, a structure note with the title "game consoles manufacturer" will show that title in search results for "Nintendo", but the user wouldn't see the tag #gamesconsolemanufacturer doing that same search. I might also think differently about interacting with that structure note vs. that tag.

• The one, crucial difference is that tags don't impose a sequence. When you write a structure note, you do. You cannot not order the content of a structure note.

That sequence of notes doesn't have to imply meaning. This is well understood, hence the HTML spec talks of an 'unordered list.'

But my point is that to write a note, a sequence of letters and lines, you will have to put them in order.

The same goes for a listing of tagged notes. The search results have to be put in sequence.

Ultimately you will always read in sequence, the point is what your mind will make out of it.

In the end, if you don't structure the content, you don't work with a structure note.

As I said in my previous post: you can clarify that structure in your index of tags, where you explain the tag.

Everything you do with structure notes, you can also implement with tags and an index of tags (or multiple thereof.)

Everything you are saying is true, but doesn't negate my point. What you are bringing into play here is the question of the HUI (human user interface), and with that questions of UID (user interface design) and UX (user experience.)

Those factors do of course influence heavily how each and every instance of a ZK (Zettelkasten) is implemented.

• A tag by itself doesn't result in sequencing, that is true. But that sequence might or can be achieved by some other means such as by convention or using the sort order feature in the View menu. Maybe someone used UID's at the beginning of the titles for the notes in that tag so they sort in chronological ascending or descending order. Or they used Luhmann numbering sequencing for that tag instead of UID's. As a result, they have an ascending or descending deliberately assigned sequencing to the view the tag gives them.

@ctietze said: In the end, if you don't structure the content, you don't work with a structure note.

This is true. If sequencing is needed or desired to provide a structure view then the tag view needs additional focusing. The structured note provides advantages and usefulness and meaningfulness and strength for someone who needs or wants or likes that view, The tag with an additional focus for sequencing does as well for someone else. Can you get more or less benefit from one or the other, yes, absolutely?

The more ways we have to affect and focus a view the more meaningfulness and strength a view can provide for different people.

There is almost always more than one way to accomplish something. More than one way to view something, say something, think about something.

These different elements, conventions, functionalities, links, tags, Omnibar, saved searches, link searches work very much like the Unix command line. We can be combined and use in so many different ways and combinations to create and provide so many different views that can be focused so many different ways it is amazingly mindblowing. I get overwhelmed by the possibilities.

Narrowing the focus of the view is very useful. But sometimes man oh man what a sight a wide-angle lens can provide. Zoom in, zoom out, mindblowing what can be done with plain text, The Archive, KBM, command line, and so much more. Mindblowingly beautiful.

Its the view. The view is everything. What view do you have in your pocket?

I guess it is all about how you look at something. Or what your focus may be.

The more things we have in the toolset and the more ways we can combine them the more possibilities can be created.

I am rambling now. I will stop.

• @Perikles said:
If you have a Zettelkasten, everything looks like a Zettel.
Or even potentially many Zettel!

Great statements! 👏

• @Perikles wrote:
There is no difference between a tag and a structure note.

So, just a small test of your claim: How do you map the structure of an argument or a model (or the structure of an argument for a structure of a model) with tags?

I am a Zettler

• @sfast: I don't know what you mean by "structure of an argument" nor the rest of it. Can you explain this, give a definition or just an example to work with?

• edited April 2020

@ctietze said:
You cannot not order the content of a structure note.

Order and structure are synonyms. Of course. My hope is that my zettelkasten will help me order/structure my knowledge in ways I currently can't predict. It will become a conversation partner. It will show me answers to the questions I put to it.

In the end, if you don't structure the content, you don't work with a structure note. If you slap a new link at the end of a file without any care for how it should fit in, maybe you're creating something akin to a buffer note: an unordered collection bucket for later processing.

Epiphany time! Deep bows to you Obi-Wan Kenobi Christian . I see the connection between creating order in a structure note and adding meaning or value to a zettel. Meaning is increased and is concretized by adding context and my own flavor to the text surrounding a link. By "flavor" I mean my own interpretation, the questions the linkage brings up, my agreements and disagreements. In other words, a structure note is not a collection of links but a description of my interaction with the material. This, like meaning and value, is a spectrum from a note containing a simple code snippet to the full digestion of a 12th-century Tibetian text. The first might only take a few seconds to record and place a link in a hub note and the second one I'm still actively working on after 2 years.

I see now where I've been a little lazy in just add links to notes when I discover relationships. In fact, I have made a place in my YAML footer to add these after note creation links. This now seems a mistake, a lazy interaction. Links want to become an integral part of the structure of the note. It doesn't really matter rather it is a formal "Structure Note" or not. By adding context and my "flavor", the link's value and meaning goes up, to the extent of my current writing abilities.

Now I want to look at forming the new habit of adding context and my "flavor" to the links I create. This too will be a spectrum, from not any "flavor" to a whole new note outlining the connection. We'll see. This seems like a best practice.

Thanks, Christian
Will

Will Simpson
“Read Poetry, Listen to Good Music, and Get Exercise”
kestrelcreek.com

@Perikles said:

But my point is that to write a note, a sequence of letters and lines, you will have to put them in order.

The same goes for a listing of tagged notes. The search results have to be put in sequence.

I think we should be clearer here: the search results are put into order by the program that produces the search results. As you said, this is a piece of the user interface. The order is not yours. It the program's.

"You have to put them in order"_ is not the same as "They are presented in an order to you".

@Perikles said:
Ultimately you will always read in sequence, the point is what your mind will make out of it.

That's right, and that's what I want to address via the "tie" of juxtaposition or sequencing. And it's the point I want to illustrate with the random note finder. You can make an ad hoc interpretation of the results; but the results were produced randomly, so there was no underlying order. Again: the computer presents a list to you, but it's not your list.

@Perikles said:

In the end, if you don't structure the content, you don't work with a structure note.

As I said in my previous post: you can clarify that structure in your index of tags, where you explain the tag.

Everything you do with structure notes, you can also implement with tags and an index of tags (or multiple thereof.)

I think you should expand these statements some more, because to me, they sound like contradictions.

@Perikles said:
There is no difference between a tag and a structure note.

Above, you say that you can "clarify that structure" in an index/in indices of tags, where you can "explain the tag". Now that's one extra step: add explanation, to make the results equal. So you now say that structure notes are equal to "explained" indices of tags. That means they are not equal to tags themselves, because tags are not identical to annotated indices of tags. You can't have both Could you address this conundrum?

In addition, I would argue that it's still a difference between (1) adding a description to an ultimately unordered, only accidentally-viewed-in-sequence set of stuff, and (2) ordering your stuff into an ordered, not unordered, list, and adding comments for up to every single reference you manually have put in, which are structure notes. Especially so once you create a structure note that references notes that don't share tags.

(For a writing project, you could invent a new #project:xyz2020 tag for just this lumping-together, but that's akin to collecting memorabilia in a shoebox for inspiration before you make a project from the collected items. At one stage, you will need an outline.)

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

• "You have to put them in order"_ is not the same as "They are presented in an order to you".

You don't have to put them in order. Physically, yes, logically, no. Hence 'unordered list.'

You can't have both

Finally somebody answers this internet meme…

It depends on what you would see as valid for holding in your head.

For instance, if I would be using the tag #TietzeOnPodcasts would you know what it means or do we have to offer an additional explanation?

Does that tag refer to:

• things Tietze said about podcasts?
• things Tietze said on a podcast show somewhere?
• appearances of Tietze on podcasts?

What about occurrences like Merlin Mann mentioning Tietze on his podcast? Does that count? Do tags change their meanings over the history of their use?

Can we do this in our heads without generating mismatches over time?

I think not.

That's why "tags" and "tags with additional explanation" are the same to me, 'cos "tags without additional explanation" are not really a thing.

Would you use tagging without an index of your tags where you explain the meaning of each giving tag?

At one stage, you will need an outline.

This should not happen inside the Zettelkasten. Writing is about much more than just a simple showing of an argument.

• @Perikles said:
@sfast: I don't know what you mean by "structure of an argument" nor the rest of it. Can you explain this, give a definition or just an example to work with?

Look. I don't want to be mean. But if you make that strong of a claim you should know the matter inside-out. You can't argue anything if you don't even know what the things are you are debating.

I am a Zettler

• Look. I don't want to be mean.

I know, don't worry about that.

You can't argue anything if you don't even know what the things are you are debating.

I agree with you.

I must have expressed myself too muddled up, let me clarify what I meant.

• structure notes
• tags.

• the structure of an argument
• a model
• (or the structure of an argument for a structure of a model).

These terms were brought up by you.

My last question to was to clarify those terms for me.

That doesn't mean that I don't have an idea, what you probably are referring to.

But I don't know for sure what you are talking about. So I ask.

Let's move on.

Say, we have these to Zettels:

1111 - Dude A is clearly awesome
2222 - Dude B is clearly awesome

Now instead of having this structure Zettel:

3333 - awesome dudes

We could have a #tag: #awesome_dudes and tag Zettels 1111 and 2222 that way.

1111 - Dude A is clearly awesome #awesome_dudes
2222 - Dude B is clearly awesome #awesome_dudes

Now let's have a more complicated structure Zettel:

4444 - awesome

1. many things can be awesome
2. dudes can be awesome
a) Dude A is clearly awesome [1111]
b) Dude B is clearly awesome [2222]

We can solve this into #tags as such:

1111 - Dude A is clearly awesome #(awesome(things)(dudes(a, b))
2222 - Dude B is clearly awesome #(awesome(things)(dudes(a, b))

and so on. LISP is very, very old.

The question is, do you want to develop your own LISP dialect for your #tags?

And what I asked @ctietze how much of that would go into a explanatory Zettel to each #tag?

You don't have to use LISP. You can use something akin to Luhman IDs or any syntax you want. It's just a matter of definitions and agreements.

(The thing is, with LISP we have the proof that it would work. No need to trust my little pea brain for that…)

• 5555 - Dude Perikles is clearly awesome #(awesome(things)(dudes(a, b, @Perikles))
I like this concept. 👏

• edited April 2020

@Perikles

Good. Then explain to me what a Stucture Zettel is, please.

I am a Zettler

• A structure Zettel is a Zettel that contains any number of nested lists of any kind. The list items are for the most part annotated links to other Zettel.

• I wrote a LISP program bout 40 years ago to find the shortest pathways for channel connected devices on IBM 360 mainframes. Never thought about the possibility of using it for finding pathways through a Zettelkasten. My only problem was that three days after I wrote the program I had no idea how it worked!

• edited April 2020

@Perikles said:
A structure Zettel is a Zettel that contains any number of nested lists of any kind. The list items are for the most part annotated links to other Zettel.

Ah, ok. I see where you are coming from. Structure Zettel are not limited to that form. They map a structure in your Zettelkasten. You identified Structure Zettel with the nested lists which can be translated to sets and subsets. That covers for the hierarchy. Tags could be used to map hierarchy. This is correct but not complete.

Sequence is not covered by tags. Or: You cannot translate a numbered list into tags. Sequence is another form of structure in the Zettelkasten.

A mathematician would perhaps say: Tagging can give you sets but not spaces. A space is a set with some added structure.

That is the reason why I asked you on how to map the structure of an argument or a model. The (mostly implicit) structure of one Zettel maps the structure of the underlying knowledge structure (model, argument, theory,...). The one thought per Zettel rule could be described by something like this: One layer of one knowledge structure per Zettel. If you don't know on the nature of knowledge structures you can't fully understand structure in the Zettelkasten. And therefore, can't really argue about the tools that are crafted to deal with those structures. After all: It is knowledge work and not zettel arrangement.

Structure Zettel map patterns that are formed by many Zettel and allows for several layers at a time. One could modify the one-thought rule for Structure Zettel: One knowledge structure about a space of knowledge structures.

I am a Zettler

• edited April 2020

I confess that I haven't read all of this rather long thread, but my cursory examination suggests that the psychological element has not been discussed much, if at all. It might therefore be worth considering this article: https://en.wikipedia.org/wiki/Meaning-making. I suppose you could crudely summarise by saying that it is impossible for a human being NOT to find meaning, or give meaning, to anything in its range of experience. In the view of psychologists of a certain stamp, meaning-making is what human beings do all the time. That is how they function in the world.

Apologies if I am derailing the discussion.

Edit: I think it is also worth having a look at Jerome Bruner on meaning-making: https://ncbi.nlm.nih.gov/pmc/articles/PMC2919784/

And to go back to the original post for a moment, a constructivist would probably say that ANY connection is meaningful, because it is the person viewing the connection who sees meaning in it, it is not the connection itself that has meaning. The same connection may have two different meanings for two different people.

• @MartinBB said:
I confess that I haven't read all of this rather long thread, but my cursory examination suggests that the psychological element has not been discussed much, if at all.

Although these links seem very interesting, it's worth noting that the original post is asking about whether links can be ranked in terms of strength/meaning, rather than just asking about how people create meaning.

Apologies if I am derailing the discussion.

At least you are self-aware and polite!

• @cobblepot

I highly disagree and moderately agree. It is a bit much for this thread to cover the whole meaning-making process in this one thread. However, it is a very important layer of the whole system that surrounds knowledge work.

Most of the time we take meaning-making for a given as an automatic process that just happens like air that we breath. But to create a meaningful connection between two Zettel you use your meaning-making apparatus which is your brain and/or mind.1

There is a psychology of interacting with the Zettelkasten and how it presents connections. It sheds some light why some techniques feel a certain way and how to exploit their value. That is, for example, a weakness of tags compared to categories: They are less tangible because you can't really use your "map-feature" of your brain. An item can be anywhere and is kind of floating in the air like a drop in a cloud. Categories on the other hand offer a place and allow for the use of the "map-feature".

This is one of the reasons why quite a number of people have the need to see. It is wish of the left brain to raise above the terrain and have a look at the territory:

Clearly we have to inhabit the world of immediate bodily experience, the actual terrain in which we live, and where our engagement with the world takes place alongside out fellow human beings, and we need to inhabit it fully. Yet at the same time we need to rise above the the landscape in which we move, so that we can see what one might call the territory. To understand the landscape we need both to go out into the felt, lived world of experience as far as possible, along what one might think of the horizontal axis, but also rise above it, on the vertical axis.2

To connect those two points before I return from me highly disagreeing and come back to moderately agreeing with you: The strength of connection is correlated with its meaning-making potential which is determined in part by the nature of braind and/or mind.

It is another reason why hierarchy of sets and subsets shouldn't be mapped via tags. You make your Zettelkasten too abstract.

But: You are right regarding the practical aspect. This thread should be concentrated on the question on how to order proximity, connection strength, meaning, asking if you can measure it, questioning those terms in general, etc.

1. @cobblepot understands this "and/or", but for the non-philosophers: There is a very difficult to understand relationship between brain and mind. Most of the time, it is just important for some ivory-jerk-off but sometimes even the domain of modern philosophy has something useful to say. ↩︎

2. Iain McGilchrist (2009): The Master and his Emissary. The Divided Brain and the Making of the Western World, Totton: Yale University Press, p.21. ↩︎

I am a Zettler

• edited April 2020

@sfast said:

Clearly we have to inhabit the world of immediate bodily experience, the actual terrain in which we live, and where our engagement with the world takes place alongside our fellow human beings, and we need to inhabit it fully. Yet at the same time we need to rise above the the landscape in which we move, so that we can see what one might call the territory. To understand the landscape we need both to go out into the felt, lived world of experience as far as possible, along what one might think of the horizontal axis, but also rise above it, on the vertical axis.

Or, to use the idiomatic English expression, sometimes you can't see the wood for the trees.

I found it useful to go back to the original post on the blog and revisit what it says there about having a conversation with one's notes, and serendipity.

https://zettelkasten.de/posts/zettelkasten-improves-thinking-writing/

It seems to me, if I have understood correctly, that what is being implied there is that you cannot predict the value of a link or connection in advance. The value is only revealed in the act of finding the note through the connection. To me, that suggests that in certain cases a tag might be as useful as a direct link to a particular note. If you want serendipity, tags have their uses, though I grant they can uncover a forest that obscures as much as it reveals.

And as to the pragmatic side of things, I am lazy, so the simpler the system is, the better. I have had experience of spending more time setting up systems and methods than actually using them for anything that contributed to the work. I've lost count of the clever schemes I have set up and abandoned. As we also say in English, sometimes the perfect is the enemy of the good.

And I hope I am not further derailing the discussion! Apologies again if people think this is not strictly relevant. To use another expression, I may be barking up the wrong tree .

Post edited by MartinBB on
• @ctietze preface: sorry if someone brought this up already and is redundant. But as I was reading this morning I made this connection.

The father of Information Theory, Claude Shannon had the concept of how surprising a piece of information is.

So in a sense when you are creating connections, for them to be meaningful means that they surprise your future self. So If I have a zettel on Cognitive Psychology, linking it to psychology is not a particularly "meaningful" connection because it isn't surprising.

On the other hand if I have a link on my Cognitive Psychology zettel to some distant field outside of psychology, say some obscure history concept, then when I click on it in the future I'll be surprised.

Hope that is valuable or surprising!