Luhmann‘s use of unique IDs

wlars

I‘m new to this very interesting approach of Niklas Luhmann, using ‚Zettelkasten‘ to process knowledge. In my research on this method I’m reading and browsing the website managed by the university of Bielefeld as well as this site. I like how Sascha and Christian let us participate in their approach to the Zettelkasten method.

However, when comparing the explanations to Luhmann’s approach on the website of the university of Bielefeld with Sascha‘s and Christian‘s approach, I come across one big difference: The usage of unique IDs.

What does the university of Bielefeld say how Luhmann used IDs?

The explanation can be found here: https://niklas-luhmann-archiv.de/nachlass/zettelkasten

Please see chapter 3.4.

They have described the approach wonderfully, so please bear with me that I do not try to translate this into English. Maybe Google translate can be of help..?

However, in my current understanding Sascha and Christian do not make use of this kind of numbering. Instead, the focus lies solely on giving each note one unique ID. In their case they use the time stamp as their ID.

What is the reasoning behind this decision? What do you see as advantages of one method over the other?

In my current point of view I think Luhmann's approach to IDs is a central part of his whole "Zettelkasten" method. Only by using this kind of approach to IDs the benefits of the whole method can be gained.

Note: For a description of the benefits please see chapter 3.3: https://niklas-luhmann-archiv.de/nachlass/zettelkasten

I‘m looking forward to your feedback!

ctietze

I recently linked to a couple on central posts on this topic here:
https://forum.zettelkasten.de/discussion/comment/3530#Comment_3530

Also see the discission "Thoughts on Folgezettel" for more.

TL;DR: Luhmann's system is super useful to organize physical sequences of notes in a sensible way. With a digital system, you can create sequences on the fly through lists of links. The benefit of having one note directly next to the other ("juxtaposition" in my post) is outweighed by clickable links.

wlars

Thanks a lot for directing me in the right direction. I can certainly connect with your focus on principles instead of techniques. That perspective helps a lot.

sigod

@wlars said:
In my current point of view I think Luhmann's approach to IDs is a central part of his whole "Zettelkasten" method. Only by using this kind of approach to IDs the benefits of the whole method can be gained.

I think this is somewhat of a technical question. Luhmann was using an analog system, he was doing everything manually. He had to devise his system in such a way that it would help him to be as fast as possible. Otherwise, Zettelkasten would've never been so successful for him.

But we're using digital systems. Computers do the work for us. And at such simple tasks computers are orders of magnitude faster than us.

Even without special data structures and algorithms, modern computers will easily work with any Zettelkasten. Even 23,000 of Luhmann's notes are nothing compared to billions of records that Google handles every second. The only way for a normal person to grow their Zettelkasten to such a size that it'll cause problems is... to become immortal.

Fun fact: Sascha and Christian use their Zettelkasten as a very simple relational database. The relational model is well established in Computer Science. So, if you ever have any issues, you can just start using any high-performance relational database. Or NoSQL, if it's your jam.

sigod

@sigod said:
Even 23,000 of Luhmann's notes

Correction: 90,000 notes.¹

---

1. Johannes F.K. Schmidt, Niklas Luhmann’s Card Index: The Fabrication of Serendipity ↩︎

masterful

@ctietze said:
I recently linked to a couple on central posts on this topic here:
https://forum.zettelkasten.de/discussion/comment/3530#Comment_3530

Also see the discission "Thoughts on Folgezettel" for more.

TL;DR: Luhmann's system is super useful to organize physical sequences of notes in a sensible way. With a digital system, you can create sequences on the fly through lists of links. The benefit of having one note directly next to the other ("juxtaposition" in my post) is outweighed by clickable links.

Hi all, this is my first post, this is a great place. I am new to Zettlekasten and I am stuck at an ID system for my file naming structure. I would still timestamp my notes, but i am not sure about having a timestamp in the filename.

What I still prefer to see in my folder is a filename structure that's easy to read. If for wat ever reason I don't have a tool I still want to be able to view my filenames in the same style as Luhnmann's id system. I like timestamps...but having filenames as dates doesn't show me the relationships in my directory system.

Example 1: file name date stamps

20210101.md
20210102.md
20210501.md
etc...

Example 2: Luhmanns ID system applied for filenames

1.md
1a.md
2.md
2a.md
2b.md
2b1.bd

etc...

plain text is future proof...but timestamped filenames is visually impossible to see the relationship's in my opinion. I am only speaking from point of view at visually glancing my filenames in my filesystem.

Maybe I am missing something?

I don't have a Mac. I am on windows trying to decide between obsidian, zettlr or anything else i can find.

Will

I'd consider filenames that were of the format - UID(Timestamp) Title.md or UID(Luhmanns ID) Title.md. Include the title in the file name. The combo of UID and Title give so much more information.

Either would work. Both show different points of view at a visual glance. I consider Luhmanns ID system harder as it takes time to slot the note correctly to get the ID, whereas timestamping is done in software. I can describe connections verbosely within a note. I don't feel I've missed anything by not using Luhmanns IDs. Remember, he didn't have a full-text search.

masterful

@Will said:
I'd consider filenames that were of the format - UID(Timestamp) Title.md or UID(Luhmanns ID) Title.md. Include the title in the file name. The combo of UID and Title give so much more information.

Either would work. Both show different points of view at a visual glance. I consider Luhmanns ID system harder as it takes time to slot the note correctly to get the ID, whereas timestamping is done in software. I can describe connections verbosely within a note. I don't feel I've missed anything by not using Luhmanns IDs. Remember, he didn't have a full-text search.

you are correct. I just want to pick a good file naming system so I don't have to restart if my system. But I am focusing on that too much rathe than the principle's

Will

Either or OR both? It probably doesn't matter.

Don't let this be the reason for procrastination. How about flipping a coin? You can change later with a simple script to rename files. Or you can intermingle both ID systems.

In the final go, I'd recommend going exclusively with UID(Timestamp) Title.md and ignoring all the esoteric discussions about the old man's ID system. He did what he did with the tools he had available. We have vastly superior tools. Lucky us!

masterful

I> @Will said:

Either or OR both? It probably doesn't matter.

Don't let this be the reason for procrastination. How about flipping a coin? You can change later with a simple script to rename files. Or you can intermingle both ID systems.

In the final go, I'd recommend going exclusively with UID(Timestamp) Title.md and ignoring all the esoteric discussions about the old man's ID system. He did what he did with the tools he had available. We have vastly superior tools. Lucky us!

Thanks Will, I'll try the timestamp.