Numerology in the UID...?

sirredcrosse

I'm just getting started, so forgive me if this is naturally part of the method or just ... impractical in the long run, but so far it's been my experience that people think a lot about their UID's, copying them out if they're analog, and just... it's a lot.
So taking a note from Scheper's file, I started out with the larger "Academic Disciplines" but I knew for a fact, if I'm going to use this for my entire career as a Humanities professor, with diverse interests in Sciences, Maths, and pretty much everything, I was going to have to add another 0 to do those disciplines justice in the first place.
So I'm starting out with: (and I'm only going to show you the parts that I've kind of filled out so far)

10000 - Humanities

• 11000 - Arts (Performing and Visual)
• • 11100 - Music
• • 11200 - Theatre
• 12000 - History
• • 12800 - English/British/UK History
• • 12900 - French History
• 13000 - Literature
• • 1381[act.scene] (For some reason... when I was reading Bussy d'Ambois by George Chapman, I placed it here in Literature instead of in Theatre... If I recall, I had a reason for this. I don't remember what it was, though. I want to say it was because I was saving "Theatre" for like... theatre theory? Anyway, to explain this UID, 13 for literature, 8 for George Chapman being English, 1 for George Chapman the author/playwright himself being the first author and 0 will be for Shakespeare. )
• 14000 - Philosophy
• • 1401[book#].[Stephanus number(paragletter)]Folgezettel# - 1 for The Republic, 0 for Plato

I have two separate systems going, apparently, in two categories under Humanities. But they both reflect something important, I feel.
I was wondering, does anyone else use their UIDs to tell them information about the card? I'm also using a date at the top, in case I can't figure out where to put it when I first make the card [YYMMDD, of course.] Like, the way i use 8 to mean "England/English" or 9 to mean "France/French" under across History and Literature. Alternatively, 0 could mean Greece and 1 for Plato in the context of Philosophy's 14018.532b1. Just thinking aloud.

JasperMcFly

Numbering is a highly personalized decision, and, as such, it is hard to both give or receive advice on how to do it.

The value of the UID is that it enables 3 things:
1) concepts can "live" at a permanent ID (concept-UID permanence)
2) linkability (can link to any Zettel)
3) Anschlussfähigkeit (can internally expand by hanging cards onto any Zettel)

Luhmann felt that he was better off creating a Zettelkasten that was thematically unrestricted and that was liberated from the mental anguish of the work to place cards in categories and also the thinking limitations (premature closure) created by them. He felt his Kasten top level cards were not categories per se (e.g. ZKI 5 "Control") rather, just "thematic entry points" that were open to the future possibility of wandering off topic.

So, at least consider creating meaningless dummy IDs, perhaps start simple with card # "1"- give a new number for each new concept and free yourself from the overhead of placing notes in categories (Luhmann felt this folderization approach invariably led to unsolvable problems of where to place something).

Relatedly, think conceptually and not factually; focus on capturing decisive concepts and reformulated encapsulations of what you read, rather than collecting facts in categories.

KillerWhale

I think the idea of preestablished categories, as with many of Scheper's (hot) takes, is a terrible idea. The joy of a Zettelkasten is to allow for emergence and roaming. Folders? BLEH.

My UIDs are just timestamps (202401161747), and even then, I only use them to visually distinguish between Zettels and other more utilitarian notes in my system.

MartinBB

@KillerWhale said:
I think the idea of preestablished categories ... is a terrible idea.

I tend to agree. But perhaps that is because I have worked on the boundaries between disciplines. I wrote a doctoral thesis on some social psychological aspects of the First World War. Was that psychology or history? Or both? For the purposes of the university it was psychology but others might see it differently.

The developer of Tinderbox (Mark Bernstein) has written about "incremental formalisation", which is basically allowing some sort of structure to emerge as you work with the data. This seems to me to be a much better way of going about things. It has something in common with Miles and Huberman's method of working with qualitative data in that it is a sort of iterative process in which one continually refines one's understanding as one finds new material.

In short, some people seem to love categories, but I do not. It seems to me that they are capable of limiting and distorting one's thinking (though they may be useful, or necessary, in some cases -- and here we perhaps have to say "thank you" to Linnaeus). To my mind a lot of interesting things can be discovered on the boundaries, where dividing lines are blurred and ideas are less set and concrete.

Andy

Like previous commenters, I think encoding a subject heading in the UID is generally inadvisable for all the reasons already stated, but I expect we will eventually hear from a dissenter to this consensus. You may also want to read a couple of blog posts by @chrisaldrich that provide more reasons for the consensus so far:

• "Thoughts on Zettelkasten numbering systems" (2022)
• "On the interdisciplinarity of Zettelkasten: card numbering, topical headings, and indices" (2023)

Here is a question that gives another way to think about the issue: How much metadata do you want to put in the UID? You can encode a subject heading and other kinds of metadata in the UID. However, it is more facilitative of what @MartinBB called "incremental formalisation" (and, more generally, changing formalisation) to put metadata in a header such as YAML frontmatter, because when you put metadata in a header you don't have to change your UIDs and update links frequently (if your links use UIDs) when you want to change the metadata that you've encoded in the UID, and because existing popular PKB/Zettelkasten software is designed to parse metadata in a header in useful ways (as tags, links-as-tags, etc., which automatically creates useful indices, attributes, etc.) but not to parse UIDs in similar ways. Creation date works well as a UID for a long-term Zettelkasten because it's usually the most unchanging of the metadata, and a header works well for other metadata for the aforementioned reason.

@JasperMcFly said:

Relatedly, think conceptually and not factually; focus on capturing decisive concepts and reformulated encapsulations of what you read, rather than collecting facts in categories.

I disagree with this advice as I understand it (but I'm not sure what @JasperMcFly means by conceptual and factual): I see no reason why a note system could not have both concepts and facts, among other thoughts, perhaps even as different note types. I don't think such advice is helpful without defining what is meant by concept and fact, which are not so univocal in meaning that a shared understanding can be assumed. If @JasperMcFly simply means that one should connect such items instead of collect them, I agree with that.

mlbrandt

I used the same system but kept to the 4 digits. I use the 1201/2.a.1 way to add notes rather than expanding it up front. Everything has pros and cons, but I have found this (so far) to be useful and useable! (I'm a physician, a theologian, and a writer... so I have LOTS of numbers I use!)

ChristophPleininger

If I were to go down the path of using Luhmann style numbering, I'd just start at 1000 with the first topic that I want to work on. Thus, 1000/1 would be the first note. If the next note doesn't connect at all with 1000/1, then make it 1001/1 and so on. If you look at Luhmann's Zettelkasten, topics usually emerged within topics. You may talk about literature in 1004/5a4c but then you link it to something related to the French Revolution, which would be 1004/5a4c1. In that case, your index entry for the French Revolution would show 1004/5a4c1. It is not a pre-defined category, it emerges naturally. I don't see the point why I should add pre-defined categories, assigning numbers to topics that I will never use.

Andy

@mlbrandt said:

I used the same system but kept to the 4 digits. I use the 1201/2.a.1 way to add notes rather than expanding it up front. Everything has pros and cons, but I have found this (so far) to be useful and useable! (I'm a physician, a theologian, and a writer... so I have LOTS of numbers I use!)

I said "I expect we will eventually hear from a dissenter", and the first dissenter appeared right after my comment! @mlbrandt mentioned that "everything has pros and cons", and the software system that @mlbrandt uses is a good example of how one's balance of pros and cons on this issue is affected by the technology one uses. I said that "existing popular PKB/Zettelkasten software is designed to parse metadata in a header in useful ways" and for me this was a reason against encoding subject metadata in the UID. If I remember correctly, @mlbrandt uses Google Docs, which does not usefully parse metadata in file headers, so putting that extra metadata in the UID (and in the filename, I would guess) makes that metadata more useable within the limitations of that software system, somewhat similar to the limitations of an analog slip box, and the pros must outweigh the cons for @mlbrandt's use case.