The Money Is in the Hubs: Johannes Schmidt on Luhmann’s Zettelkasten

edited March 2018

The Money Is in the Hubs: Johannes Schmidt on Luhmann’s Zettelkasten

The Zettelkasten note-taking method has made book writing and writing scientific papers easy for hundreds of years already.

• Do you have any links to Schmidt's work (German is fine)?

There is an ongoing research project by Johannes Schmidt on Luhmann's Zettelkasten. Some articles are available publicly and in English.

See the above link for a list of publications on the topic by Johannes Schmidt, Guardian of the Godfather's Zettelkasten.

I am a Zettler

• Just my 2 cents about the use of hubs.
I had been using hashtags quite rigorously, having shortened my list to 135. What I found is that when one searches for a specific hashtag one can end up with a long list of notes that have that hashtag. Not very satisfactory. @Sascha has mentioned somewhere that using hashtags is a waste of time for precisely this reason, and I fully agree.

Hubs, OTOH, are different. A hub note refers to a specific topic, and it has links to all the relevant notes. In other words, when a note is created one has to decide which hub needs to show a link to it. Each note is linked to by at least 1 hub, sometimes even by more than 1 hub. I call the hub notes spokes, like the spokes of a bicycle wheel.

1. since each spoke has at least 1 hub linking to it, it is not alone, it is not isolated, and it cannot in principle be orphaned unless a hub would be deleted accidentally;
2. a hub can serve as an entry point to other notes; and
3. the hub system, in a way, is like a tag: a spoke can be linked to by more than 1 hub,
except that tags are not linked so you don’t know beforehand which notes go with a certain tag, whereas looking at notes listed in a hub you do know which spokes are there.

Using hubs is a good step forward in setting up a Zettelkasten, if I am not mistaken.

• It is about the scaling dynamics. Tags get worse the more you use them, Structure Zettel better. It is all about being antifragile.

I think you are spot on.

I am a Zettler

• I gave this a bit more thought, and believe tagging does not need to be an either/or choice: either tags or hub notes.

At 1st one may think that a hub note is the same as a folder, but that is not true: a hub note can link to any note, whereas a folder is a container for notes but it cannot link to anything. In other words, if one has a hierarchical organisation of folders with notes, each note can only be in 1 specific folder, whereas a hub note can have links to files in different folders at the same time.

Thus, the combination of tagging and filtering is powerful because it allows a more precise search filter.

If one were to filter only on tag1, the result can be a long filtered list of notes. If one only filters on HN1 (= hub note) one gets a shorter list, but even then not all the notes linked to from HN1 may be relevant.

If one filters on HN1 AND tag1, one gets a shorter list. The number of potentially interesting notes linked to from HN1 is thus narrowed down.

Filtering on HN1 AND tag1 AND tag2 gives an even further narrowed down list.

1 aspect to keep in mind is that, like tag names, hub note titles need to be unique or else the search filter will give false positives.

• edited March 2020

@Peter I would offer as a suggestion/reminder/hint that all the operators can help with tags when we run into scaling issues.

   HN1 AND tag1 AND tag2 NOT tag8
("Rome" OR "Gaul") AND Tag1 AND Tag2 NOT Tag8 (HN1 and HN3)


Also, these search strings can be search links:

   [[HN1 AND tag1 AND tag2 NOT tag8]]

[[HN1 AND tag1 AND tag2 !tag8]]


Mastering the search operator, search strings, and the Omnibar reduces the impact of scaling issues.

Very useful for narrowing your focus.

https://zettelkasten.de/the-archive/help/#can-i-use-boolean-search

@Peter I would offer as a suggestion/reminder/hint that all the operators can help with tags when we run into scaling issues.

HN1 AND tag1 AND tag2 NOT tag8
("Rome" OR "Gaul") AND Tag1 AND Tag2 NOT Tag8 (HN1 and HN3)

Thanks for that further addition, Mike, I fully agree with that. I made this point, though, because in my comment yesterday I concurred with Sascha about his premise that tags in a Zettelkasten with many Zettel is a waste of time.

Some deeper thought about this change my point of view: using both tags and hub notes is useful if one uses the various and right search operators, as you point out.

• The link to research is broken, also would like to get some research in English, I am struggling to find any.

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

• I've used hubs for a while and find them very useful. Their proliferation fostered the creation of a hub-of-hubs, which is nothing more than an alphabetized index into all of my hubs within the Zettelkasten. Over time, the initial hub-of-hubs grew to a point that I now have two hub-of-hubs, one A-->L, and the other M --> Z. As soon as the size of a hub-of-hub extends beyond a generic screen size, it's time for the birth of a new hub-of-hubs. All my hubs are alphabetized by hand, which I have found forces me to stay just a little bit closer to all my zettels. Synchronistic events are always welcomed.

• @ctietze said:

It is the link to Schmidt‘s publications:
https://uni-bielefeld.de/soz/luhmann-archiv/publikationen.html

• Unfortunately many (most?) links were broken when Bielefeld University ‘updated’ their website two years ago. The page you were looking for is now most likely:

https://niklas-luhmann-archiv.de/projekt/publikationen