That’s it!
I have spent many hours trying to realise a Zettelkasten modelled on Luhman’s using Devonthin, but the result is unsatisfactory.
With Tinderbox it works without any problems, especially in Outline View. Unfortunately, there is no iOS version of Tinderbox.
That’s correct. But, also, the manual sorting order is not (yet) synchronized.
What exactly did Luhman do and how did he do it? He wrote about 80,000 notes (Zettel) with cross-references between 1952 and 1996. Let us assume
On note X there is a reference to note A. How did Luhmann remember that there was a note A when he wrote note X 32 years later?
Many references are written with a different pen than the note itself. So they were added later. Where there are no references, nobody knows whether that was intentional or whether Luhmann only forgot them.
Edit: I forgot the conclusion 
What I’m actually trying to say is that any reference system is actually as good as Luhman’s. The individual pieces of information are linked … or not. Coincidence or intention?
Frank, you’re right. Luhmann was also only human and his Zettelkasten will certainly not have been perfectly organised. But it was still extremely helpful.
But I have to disagree with you when you say:
“What I’m actually trying to say is that any reference system is actually as good as Luhman’s.”
That is not the case. Luhmann’s technique is simple but the benefit is in the detail. One needs some time to understand it.
I have observed that although many people deal with Luhmann’s Zettelkasten, they unfortunately do so very superficially. I fell into the same trap: I thought I had understood the principle and didn’t recognise anything special in it. That was many years ago when I was still a student. It wasn’t until years after my studies that I came across the Zettelkasten again and only then did I understand it. Daniel Lüdecke, the author of his Java Zkn3 App, explains in this discussion how the Zettelkasten really works (Daniel understood the principle very well):
https://www.organizingcreativity.com/2013/09/luhmanns-zettelkasten-with-devonthink/
"Hi Daniel,
I think it is a little bit more difficult to adopt Luhmann’s Zettelkasten system with DevonThink, because the core principle of Luhmann’s Zettelkasten was not to use (automatic) crossreferences (tagging), but the continuation and branching of an idea/argumentation/excerpt (“innere Verweisungsfähgikeit). That’s where Luhmann’s famous numeration of card files is used for. With this technique, you can start with an idea or keyword, retrieve related cards and then start following a “path” and choose certain branches, depending on how you want to develop your argumentation (coming from “Picasso” to “museum” or to “cubism” and so on).
(…)
Best wishes
Daniel"
Daniel even describes it again in detail in German here :
(@eboehnisch: Wink mit dem Zaunpfahl!)
So the main point of the Zettelkasten is that it recognises undiscovered connections and can transfer solutions to problems from one area to another. However, in order to recognise these connections, it is not enough to link information together, you have to be able to map the train of thought. If you do not map the train of thought, you will not recognise the connection even if you have connected the bare information.
Afterwards you also understand that Devonthink - despite its performance - is currently not so well suited to recreating the Zettelkasten. The criterion of linkability is fulfilled and that is excellent. However, it lacks the ability to model thought processes on Luhman’s model. To do this, you would have to be able to subordinate one note to another - as is possible in Tinderbox. However, this is not possible; you would have to create a group and use it as a workaround. But that makes the whole thing confusing again.
I think Devonthink is only missing very very little to finally be able to map a Zettelkasten with it and that would be extremely helpful. But for some reason, the Devonthink team is not working on it and is not enabling Devonthink to do so.
subzero
Zettelkasten is not the only way to uncover new connections. Moreover, I would argue that Zettelkasten (at least in its original flavor) is not even one of the better ways, since it does not integrate well with search and automation – two things your computer really excels at, even without using DevonThink. When it comes to your observation that …
… the most obvious explanation is that Zettelkasten is simply not good enough for most people of the digital era.
Consider the possibility that you have not fully understood the Zettelkasten and therefore think it is too bad.
I think, you are overlooking important points. It is comparable to the difference between knowledge and understanding.
Quick access to information does not mean that one understands the concept behind it. In fact, quick access can even make understanding more difficult, because it makes it possible to have access to the information without understanding the context behind it. Luhmann’s Zettelkasten ensures exactly this: Access to information and simultaneous communication of the context.
This type of information management requires patience - a skill that many people no longer have these days and therefore consider to be old-fashioned and therefore bad.
Only? 
In contrast to the physical Zettelkasten, the “Zettelkasten” in Luhman’s head was a biological neural network. Connections between information arise “by themselves”. Like all exceptional talents, Luhman simply thought … all the time. The “cross-references” (or whatever you want to call them) of a Zettelkasten are not the starting point of your own thoughts. It’s rather the other way around. Information can be archived, genius cannot.
I don’t know much about such things. But I know how it works for me. I think about something over and over again. I look at pictures, read texts. Then I take a break. Then I read the same text again. Start thinking again … and sometimes … sometimes a new thought suddenly arises 
What science are you talking about? There is rarely (never?) just one concept and one context from which you can draw exactly one conclusion.
The main point of the Zettelkasten is that it facilitates discovery of connections by the human. The Zettelkasten by itself – whether analog or digital – is just an artifact.
It was extremely helpful to him. He wasn’t trying to create a universal system, he was trying to create a system that aligned with his own particular information needs and ways of thinking. Fetishizing the particular solution he came up with seems to me to entirely miss the point.
I am glad to hear you are invested in Zettelkasten and it is effective for you. But I wouldn’t characterize people who don’t like or use it as impatient or oblivious. Luhmann’s mindset isn’t universal, just as yours isn’t… or mine… or @kewms, @meowky, etc.
That is much of why we don’t try to develop DEVONthink or DEVONthink To Go into a bespoke Zettel application. But you should also take note we aren’t developing our apps specifically for Johnny Decimal, GTD, PARA, or any other particular system. Do we provide flexibility in our apps to help accommodate those things? Sure, to some degree or another. But I’d say our largest contributions in supporting such systems are our automation tech, allowing people to shape their environment in deeper ways than many apps do (or can) and the deep linking. Between those two, many things become possible. 
Of course, one is not impatient or anything like that because one don’t use the Zettelkasten. But it may be that one does not use the Zettelkasten because one is impatient. That is a difference. In any case, I didn’t understand the Zettelkasten for many years because I was too impatient and superficial with it, and I imagine I’m not alone in this.
It is understandable that your main goal is not to recreate the card box. You have to follow your own concept and not someone else’s. But as a user, I just find it unfortunate and sometimes frustrating that Devonthink can do almost everything you need to recreate the Zettelkasten (and much more) and only so little is missing, but that you therefore have to use other software like Tinderbox.
Agreed - moreover Luhmann lived in an analog paper world. Who knows what system Luhmann would have preferred or invented or used if he owned a Mac?
Bear in mind that other people also find Luhmann’s approach helpful - not just him. Accusing these people of fetishisation misses the point.
Of course.
That’s true, but it doesn’t add anything helpful to its description.
It’s not that simple. Luhmann described his Zettelkasten as a “dialogue partner”. In other words, a clever way of entering into a dialogue with oneself. Nobody can probably say what was the starting point for him here and what was not.
I didn’t claim that at all.
I use structure notes in DT to organize my Zettels in any sequence I want. Thus, a zettel can be linked to endless structure notes. When I need to put all notes together for an article, for instance, I create a markdown file and use transclusion. Managing zettels in DT is a piece of cake :),
This is my approach as well. And I wouldn’t characterize it as a Zettelkasten. That one has individual notes attached to a topical index does not automatically qualify them as followers of Luhmann.
This synthesis can help.
The shortcomings of the original Zettel method is concisely synthesized at the beginning of the link you provided.
For all intents and purposes, we often translate the German word “Folgezettel” as “sequence of notes”.
For the purpose of this argument, we may presume that the world is a web of interconnected things. It’s not necessarily possible, and surely not practical, to reduce this multi-dimensional web to a finite number of linear sequences.
My notes is a (partial) reflection of my knowledge world. I don’t accept an oversimplification of this web of ideas into a linear form. Links with context are good as they represent individual connections among nodes of this web. Sequences, numerical or in another form, on the other hand, are arbitrary. They do not represent any individual element in the web.
IMO what distinguishes the (original) Zettel method from other note-taking conventions is its liberal use of sequences. Therefore, I don’t consider my own workflow as a Zettelkasten. I’m aware there are people who do not use sequences, but nevertheless claim to be a Zettel practitioner. They are, of course, free to acquire whatever labels for themselves in this age of identity politics.
Disclaimer: I don’t know anything about Zettelkasten. But “Folgezettel” does not necessarily mean a linear sequence of Zettels. It can also indicate a temporal sequence, ie Zettels created later than the first one.
Maybe some things do get lost in translation.