Notes management via tags (need a 2nd opinion)

I was inspired by this image posted on discord.

I have ~600 notes on various books, but they are not made in Zettelkasten method, so I will have to refactor them based on principle of atomicity, internal links etc. Therefore those ~600 notes I will estimate to pass over 1500.

The main problem I’m confronting in the field of literature is the abundance of terminology, delimitations, opinions, definitions and various fields where a concept can be applied: dramaturgy, poetry, prose, frontiers text, ideology etc. This will result in a very fuzzy network and a very amount of time to select, process, construct a “Structured note” (or Map of Content) etc.

I want almost a bulletproof method to find my notes, but I want also to have the benefit of surprise and creativity when I’ll construct my MOC (Structured notes)

So, here is what I have in mind when I’ll refactor my notes, and here is I need your help to tell me if you spot some troubles that I’ll might encounter in the future. Tags as entry points/ doors, NOT as corridors.

Almost everything we encounter in our research has more or less the following:

  • definition / is a concept
  • characteristics, properties
  • a history
  • some boundaries
  • can go into relations with other objects, ideas etc
  • can develop a process or be part of a process
  • has a methodological approach/ procedure or can develop one
  • a structure

So I thought that these lines above should be my tags for notes. Titles of the notes should be made in order to reflect content. The content should be atomic & internal linked with other notes.

Searching through notes using tags.

Imagine we have 2 categories: fruits & vegetables

Apple’s notes:

  • A note that contains definition of an apple (#apple #definition)
  • Characteristics of an apple (#apple #characteristics)
  • Structure of an apple (#apple #structure)
  • Boundaries (#apple #boundaries)
  • Growing process of an apple (#apple #process)


Tomato’s notes

  • A note that contains definition of a tomato (#tomato #definition)
  • Characteristics of a tomato (#tomato #characteristics)
  • Structure of a tomato (#tomato #structure)
  • Boundaries (#tomato #boundaries)
  • Growing process of a tomato (#tomato #process)


When I’ll search for a particular problem (#apple #structure) I’ll get all the information I need, also I can combine for further analyses two categories: #apple process OR #tomato #process to spot the resembles of those two processes. In this way (in my opinion) I’ll restrict the tags numbers, but also I have the advantage of being surprised by the title of a note and have creative ideas when constructing a Structured notes (MOC, Hub) and also to generate future research.


Growing apples in the 19th century was restricted because of ... (#apple #boundaries)
Tomatoes can be cultivated only in a certain soil) -- (#tomato #boundaries)

I hope this post will make some sens to you.

Question for you:

  1. Do you see any limitation to this system of tagging when I will have let’s say 10,000+ notes, for example?
  2. Is it to broad?

Thanks in advance!

1 Like

Speaking only with experience of using a similar system within a single project based vault that was and still is a dense web, I can imagine that it will be unlikely to completely avoid overlap between, for example, Apple+Structure and Apple+Process. This may warrant the use of embedded content in both for easier editing and consistency.

This is not a problem for certain use cases, but when trying to limit duplicate occurrences of content in projects where you are working towards a more concrete end result, and you are adding adjacent relevant commentary on these embeds, it may become quite an arduous endeavor, yet potentially totally worth it. However, it can also feel disorienting and counterproductive when the very system that was intended to help you build out ideas creatively becomes a hindrance as your key related content becomes increasingly scattered. And, although it can be collected as one, when that is done, it is riddled with duplicate and extraneous information. This is where the note level tag filtering that I have only recently read about sounds powerful.

Anyways, in my example, perhaps this commentary content about the embedded content should be built atomically with the same tagging system and embedded besides the duplicate content. In this case, the note number surely does skyrocket, which may not be an issue.

This is all assuming a retroactive build process, which is rare to occur without sparks of new content to be captured and processed alongside simultaneously in one go. And since you really do not know what may later become duplicate content, perhaps it is safest to just keep everything unique, avoiding duplicate embeds when possible and in those cases perhaps breaking that note into smaller purpose driven headings for transclusion. Or the truly atomic option/workaround to my concern may be to simply rely on links utilizing longer more descriptive link names/aliases. This helped somewhat in my project.


1 Like

How do you see this overlap? In the example, you offered, I “see” this: I’m searching about the apple concept, and I’m interested in it’s structure & it’s process. Therefore, I can make an MOC about Apple and all the notes that are talking about structure & process.

Can you give me a hypothetical example of this duplicate? Since I’ll use so few tags, I’ll relay in a very well written title, meaning that the title should be very precise about what I’ll find in a specific note. For example:

tag: #apple #structure

Title: The apple structure is based on fussion and attachements of elements of the apple flower.

The basal portions of the petals, calyx, and stamen are fused into 
hypanthium tissue and attached to the ovary. The ovary is inferior 
(below) the petals, calyx, and stamen. “Apple type” fruit, [[which 
develop from epigynous flowers]] (<= this will be a link to a note about #apple #origin), have remnants of the flower sepals, 
petals, stigma and stamen attached at the base of the fruit, on the 
opposite side from the receptacle. 


What is this level tag filtering?

I do realize that using this kind of tagging I will have to relay on very atomic notes (I actually have in mind that my notes, even that I write them on PC, should not be to much longer than a postcard size 3x5in), but I also think that a very well written title should tell you about the content of a note without rereading that note again.

1 Like

In terms of the example overlap, I guess imagine you were an apple farmer who was dabbling in genetic modification and your vault contained hundreds if not thousands of relevant notes built intentionally as building blocks to later engineer new methods and directions for experimentation. Would it not potentially be helpful to establish a hierarchy order and stick to it when tagging?

Like, maybe apple remains a solo tag, but perhaps utilizing nesting with process/structure and structure/process tags, the MOC could stay focused and be more easily organized. To me, process/structure would be notes where the processing or processes of apples are of utmost importance but in relation to their structure, like how to squeeze my new franken-apples without damaging the extremely large and delicate mutated seeds. But structure/process may be for notes about why (in terms of processes) the seed shells (structure) to my new apple variation are so soft.

I am not saying nested tags are the answer, I am just imagining possibly further delineating regions while staying broad with tags few and far between. Bottom line is that I guess I would be concerned about broadness and consequently how long it would take to create and maintain these extensive MOCs and whether this maintenance would hold up as the organic iterative process one might hope for.

In terms of the duplicates, I am a little out of my depths with this example, but sticking to apples, perhaps the apple structure note has headings for different apple parts and those headings have a mishmash of embedded notes and headings from other relevant notes about those apple parts. By nature, blocks won’t be too helpful here, so either scrape by with links and effective naming or commit to embedding and adding a commentary embed immediately before or after. I say this because if this isn’t done atomically , the ever important drawn conclusions and connections get lost in the noise. Like, if you solved the soft seed apple squeezing problem and simply write about it beside an embed from that note within a note about your overall apple processing technique, upon later searches it might be overlooked. However, with the effective naming and focused atoms, you are probably all good, so ignore all that, ha.

In terms of the note level filtering, I just saw this post yesterday for the first time, and I have never used Roam, but the idea sounds almost like a live in-note dataview composition workspace: Filtering Everywhere! - #28 by Gaston

I like what you said about notes not being too much longer than a postcard size, but I build MOC/hybrid style mutant notes, and as an aspiring apple farmer, these notes are just as quotable when building new ideas. However, I struggle to effectively and succinctly name these meandering creations, not to mention trying to fit them on a 3 by 5. Of course, this is probably solvable using a more modular approach to MOC construction, but I am an artist, not an engineer, and I hope that I can feel way while designing my ideas, at least in the first pass.

Thanks, for the post and good questions.

1 Like

Now I see what you mean clearly. Thanks a lot for your input. As far as for the

I was referring only to notes to be 3x5. I can’t imagine this size for a MOC. My MOC’s are quite large and contains links to other MOC’s sometimes.

1 Like