Archive... or Thinking Environment?
As my zettelkasten practice has evolved, I’ve been noticing a tension in my own work(flows) and I’m curious how others here think about and relate to this.
Most PKM conversations focus on capturing, connecting, and developing ideas (note types, tools and the like). But when I reflect on the moments and situations that have lasting impact in business and life zoom out, they tend to be decisions — what to pursue, what to drop, how to respond, which direction to commit to.
So the question I'm working through and which I'd be interested to get others' perspectives on is:
When you’re facing a real, consequential decision, how (if at all) does your PKM system help you think it through (in practice)?
For example:
- Do you enter the decision into your notes and work it there — or does the thinking mostly happen elsewhere, with the notes playing a supporting role?
- Have you ever created notes that function less like reference material and more like thinking tools (prompts, checklists, decision records, assumption maps, pre/post-mortems, etc.)?
- Or do you deliberately avoid formalising decisions in your system because it adds friction or feels like over-engineering?
I’m especially interested in where this breaks down:
- moments where your system felt like it should help, but didn’t
- situations where structure helped clarity — or where it got in the way
- things you’ve tried once and quietly abandoned
I’m not assuming there’s a right answer here, instead I'm interested in how (and whether) people who care and invest time and effort into thinking, learning, and knowledge work relate to decision-making as a first-class activity and whether most people implicitly (or explicitly) treat it as something separate from knowledge systems.
Would love to hear how this shows up (or doesn’t) in your own practice.
Howdy, Stranger!

Comments
I manage bigger decisions like projects. I consider projects somewhat separate from PKMS. They live usually in a project folder where I collect research material and keep track of the process. When all is done, most of the project notes get trashed. Only a few are saved in an archive.
The archive helps to learn from earlier experiences. How did I decide back then? Why? Looking back, what part of the decision process worked and what was a mistake? What can I learn from this experience for similar situations in the future?
The archive also helps with feelings of regret. People tend to regret the consequences of decisions less, if the decisions were made for a good reason. For example: "The decision to accept this horrible job was a mistake. But it was the right decision at the time, given the circumstances and available knowledge back then." If you don't have a documentation of the decision process, there's a risk of counterfactual thinking and hindsight bias. In my experience not much documentation is needed. Sometimes a few lines in a journal is enough.
I do have notes in my PKMS, that contain reminders like these (plus some background on the psychology):
These reminders help with consequential decisions, because they remind me of successful decisions strategies, that I'm already familiar with. They help reduce stress in high-stakes situations.
I find formalized and detailed note-taking useful to a degree, but I need to balance them with other methods that activate other parts of the brain.
I use my note system for writing/thinking about life decisions. As @harr said, documentation of the decision-making process is very important. I use IBIS-like decision rationale notation, as I've said ad nauseum in this forum. I also have notes on various problem-structuring & decision-making frameworks that I can consult.
I can't think of any "moments where your system felt like it should help, but didn’t", which is either a sign that I'm doing something right or that I'm good at forgetting. I'm always trying to improve and don't consider myself a master at it.
I'm not familiar with IBIS (at least not the term - I may discover that the methodology is something I'm already aware of under a different name).
Those links are all interesting and broadly map to how I've been using the zettelkasten. In short, the zettelkasten being where my version of the 'theory' lives and it serves as a kind of tailored/me-flavoured reference library. No doubt that's of value.
The direction my thinking is going now - and this is something @Sascha advocates for in some ways - is essentially this: what could/would the zettelkasten look like/need to contain for it to actually become the environment in which thinking and decision making is done (as opposed to the place I go to refer to the frameworks that help me make decisions).
The edges between those two things are somewhat blurry since writing in the zettelkasten is thinking and decision-making. I also think there's an additional level the system could perhaps elevate to if it was to truly be the environment in which decisions are made.
@Andy perhaps a better way to phrase "moments where your system felt like it should help, but didn’t" would be: "I had a decision to make/a sticky problem to think through and I found that I needed to do something in addition to working in my zettelkasten".
For me that tends to be things like opening up a Scapple board to dump all of the loose thoughts onto a free-form canvas so I have the raw materials of the decision in front of me in a way I can reference (and explore possible connections between). Sometimes it's capturing a stream of consciousness reflection in a voice notes app. Other times it's as structured as using a decision journal template - which is something I've tended to do separate from the zettelkasten.
As I think about what this looks like for me, it tends to involve changing the format of the content (from linear notes to a mind map for example).
Yeah, this maps to how I've tended to approach decisions as well. My GTD-informed mind tends to want to maintain very explicit hard edges between the different spaces. I don't typically go as far as to create a project for the decision (unless I do, in cases such as "Move [parent] to new care home]" in which the decisions being made include "should we move?" "where should we move?" etc).
When you say you manage bigger decisions like projects do you mean that you think of it as something that needs to be managed in a different space (e.g. your project management tool or a different notes/documents location)?
This is an interesting piece of it and is something I've started managing with a running log of time/date stamped interstitial journal style notes in the inspector/sidebar of zettel (my zettelkasten is in Scrivener). I find a lot of value in being able to refer back to what I was thinking X long ago. Are you referring to a similar practice of capturing date stamped reflections in the note body in some way?
Agreed - though I've used a more formal decision journal template on occasion (e.g. for major business direction decisions), I haven't generally found that creating formal documentation was a sticky practice.
This is pretty similar to what I've got. The question I'm asking myself is whether the value in these is in having created them more than it's generally in referring to them in the moment of making a specific decision. In some sense they're the artefacts of having thought about something in the past more than they're a just-in-time tool. I'm wondering to what extent there's value in figuring out how to make notes like these just-in-time tools in some way. The shorthand for this would be to think of applying "checklist manifesto" thinking to designing these notes/versions of them.
Appreciate you taking the time to share such detailed thoughts. Very insightful.
IBIS is fascinating - thanks for pointing that out.
Do you have an example of IBIS applied in your zettelkasten you could share or point me to in another of your comments here in the forum? I'd be keen to see how you implement it.
Same here.
Yes.
I find Sönke Ahrens' terminology useful. He distinguishes in How to take smart notes (2nd ed, 2022) three types of notes (Chapter 6):
This distinction is a reason, why I wouldn't take project notes in a ZK.
Instead of expanding a ZK to include project notes, I prefer to complement the ZK with project folders.
In my Obsidian setup Zettels and projects live in the same space (vault). But this space is divided into distinct sub-spaces (folders).
They also have a clear conceptual boundary.
For example: In my project notes I love to add links to permanent notes. For me that's an important motivation to write permanent notes. They make knowledge accessible. But I do not add links to projects in my permanent notes, because they would break, when I delete the project notes.
Yes and no. I do occasionally add timestamped comments to my notes, when I want to keep track of how a thought developed. I do write a journal for personal stuff.
But I don't do interstitial journaling. It's the opposite of what my brain needs during a break. My brain needs mindless breaks, not mindful breaks.
(Keywords: default mode network, niksen, daydreaming; the original Pomodoro technique gets it right.) And I don't like mixing everything in one log.
I don't think that one is more valuable than the other:
I'm 100% with you on this - I just tend to take the hard edges distinction even further by having dedicated Scrivener projects external to the zettelkasten project.
That's likely partially a manifestation of the tools in question (my limited experience with Obsidian indicated that you could have more distinct sub-spaces there than you can in a tool like Scrivener. That said, I feel like having search end up pulling from the combined pool of data is a form of 'contamination' (strong word I know) that my mind doesn't allow space for.
As a matter of interest - what's the heuristic for deciding that a given project note is a 'delete' rather than 'file'? For me it's a case of keeping what I think of as Reference/Project support in a totally different location (a PARA-structured Evernote account) and project folders find their way to the Archive stack once the project is finished or put on ice. I've been surprised on a number of occasions by the need to go back to and extract value from an archived project folder that I was tempted to consider done and no longer worth keeping.
Yeah, I should probably just refer to these notes as more the kind of time-stamped observations that fall under "doing the work" rather than "notes I jot down during a break" - I'm totally with you re: mindless breaks.
Agreed - perhaps it's more a case of having sufficient meta-awareness to be aware of which form of value you're making use of in any given situation.
I think this question is the key to the overall topic:
The Zettelkasten speaks the language of knowledge, project-/taskmanagement the language of action. That doesn't mean that there should be no task in the Zettelkasten or no information in project-/taskmanagement.
Why would you create a checklist in your Zettelkasten? Let' me answer the question by a practical example:
Each time, I create and publish a podcast episode, I learn something new. Each episode informs the next one. The tool are checklists, workflows, and templates.
The learnings during each podcast episode creation generating during each episode feeds into the checklists, workflows, and templates which then will be the foundation of the next episode.
So, it is project -> Zettelkasten -> project -> Zettelkasten etc.
The question is best answered by intention. You might even create individual projects as use cases that you can draw on. A lot works if you are intentional.
I am a Zettler
There are many great observations above that I can identify with.
@jameslongley said:
What comes to mind for me here, and may be helpful, is the analytical framework for aspects of personal knowledge bases (PKBs) from Davies et al. (2005) in this discussion (full citation in the footnote here): 1 Data model (1.1 Structural framework; 1.2 Knowledge elements; 1.3 Schema, or formal semantics), 2 User interface/views, 3 Software architecture. When you find some aspect of your PKB/ZK too limiting, you can ask yourself whether you need a different view (2) of the same data, or need some aspect of the data model (1) to change, etc. This is a more specific way of talking about "changing the format of the content".
A Scapple board, for example, need not be something outside your ZK if you approach it with a particular structural framework or schema in mind. It could be a kind of structure note without clickable links. Clickable links are just a user-interface feature; people with analog Zettelkästen don't have clickable links!
@jameslongley asked:
I haven't shared examples because it's too personal, but a paragraph describing in general how I use it for life goals is in this comment from five years ago. In terms of David Allen's model of six levels of altitude for reviewing your life and work, that earlier comment is about the higher altitudes. IBIS or something similar can also be used at the lower altitudes, namely in specific projects like you and @harr were talking about above.
@Sascha said:
I like this, and I think it works well if we substitute "inquiry" for "knowledge" too: "The Zettelkasten speaks the language of inquiry, project-/taskmanagement the language of action. That doesn't mean that there should be no task in the Zettelkasten or no inquiry in project-/taskmanagement."
I'm reminded of something I said in the infamous discussion of Chris Rock about combining GTD with "deep inquiry":
This is a great perspective @Andy - I'll have a look at those links and reflect on this since it would be a massive time and effort saver to not feel the pull to transform the format/location of the data to as large as extent as I currently do.
It's more than a UI feature. There's a fundamental difference between text and hypertext. The clickable link is the foundation of hypertext.
Personally I enjoy hypertext, because it removes so much of paper's friction. (It adds friction in other places, though. You need access to devices and software.)
I fail to understand, what Sascha is trying to say.
Maybe I'm too influenced by GTD? Allen writes in GTD (1st edition, 2001) about "The importance of Hard Edges" (my emphasis):
Those categories are: A “Projects” list, Project support material, Calendared actions and information, “Next Actions” lists, A “Waiting For” list, Reference material, A “Someday/Maybe” list.
In my experience, this is very good advice.
I place Zettelkasten in the category "Reference Material".
@harr said:
Oh, my friend, do I have a mind-blowing reading suggestion for you: Riccardo Ridi's entry on hypertext in the Encyclopedia of Knowledge Organization! Exactly how Ridi's perspective is different from "The clickable link is the foundation of hypertext" I leave as an exercise for the reader.
Thanks for the recommendation. I enjoy deep explorations of historical developments.
I find Ridi's term hypertextuality useful to describe a set of properties that can also be found in non-digital systems on various scales, including Luhmann's Zettelkasten.
But I prefer to limit the term hypertext to digital systems, as several other authors do.
Moving from one node to another with just one click or tap is qualitatively a very different experience than looking up an article in an encyclopedia or locating an index card in a huge box or walking through a library.
The magic of hypertext (and hypermedia) is how the information appears automatically right there at your fingertips. You don't go to the information, it comes to you.
https://www.isko.org/cyclo/hypertext8.jpg
The link is where the magic happens.
@harr, suit yourself! My point in differentiating clickable links from other aspects of a PKB/ZK such as data model and software architecture is that it may help people see new possibilities for using whatever tools are available. I thought that @jameslongley would be well served by a different way of thinking about his use of Scapple boards. Another example is the blog post "TextMate as a Zettelkasten App" (2020). As @Sascha said there:
TextMate is a software system without clickable links, but does the absence of clickable links mean that a ZK edited and navigated in TextMate is not digital hypertext? Not in my view, but yes in your view.
Whichever view you take, the point is to expand your thinking about tools, breaking through any self-limiting preconceptions, so that you can make the best use of the tools you have and can "program yourself for productivity and stop searching for the ideal software" (2015).
I ask different questions:
I think so. The ability to step back, to be aware of what you are aware of, is very useful for personal knowledge management and decision-making.
Non secitur. A qualitative difference is not enough supporting reason. You have to establish that the difference is essential.
I am a Zettler
This is about a personal preference. I literally said so:
Ridi mentions several definitions of hypertext. He prefers a wider concept, because he wants to draw the attention on historical continuity. The elements of hypertextuality have been around for a while.
I prefer the narrower definition of hypertext as a digital medium, because of the qualitative difference. I don't know any non-digital tools that have that particular quality.
You could compare it to debates about definitions of the word "Zettelkasten" in the english language. Some limit the term to a Luhmannesque Zettelkasten with paper slips in a box. Others expand it to include digital workflows. Some use it as synonym for any kind of card index and even expand it to include Eminem's notes. How would you decide, who's right?
De gustibus non est disputandum.
EDIT: My concept of hypertext matches what Jeff Conklin's described 1987 in the classic article Hypertext: An Introduction and Survey:
Interestingly it also contains includes references to index cards:
I think like the people mentioned here:
Today we have a many tools available that support hypertext with plain text files, including The Archive, Obsidian, Zettlr, iA Writer, …
@harr, yeah, the analogy between the semantic range of hypertext and Zettelkasten occurred to me too before you mentioned it, and I'm familiar with Conklin's article and have cited it before in this forum. Conklin uses the term hypertext in both narrower and broader senses, as your excerpts show ("One kind of manual hypertext is...": emphasis added). Both Conklin's and Ridi's articles show the large number of aspects or features of hypertext systems that can be analyzed. In short, it's complicated.
What I would emphasize is that if you bring to your tools an analytical framework like those in Conklin, Davies et al., & Ridi, it opens up conceptual possibilities so that you're not so constrained by one way of thinking about your tools.
@harr At the moment, one support a stated opinion with arguments, evidence, or authorities the opinion ceases to be just a personal preference.
On the basis of arguments and evidence.
I am a Zettler
@Andy and @harr do I have to provide you two with a hyperlink to an online boxing ring 😅
I disagree. In this case there's no absolute truth to be discovered. There's no right or wrong. It's just a matter of preference.
The definition of "hypertext" is arbitrary. Someome made the word up (Ted Nelson around 1965). Others used the same definition. Some came up with their own. Others widened or blurred the boundaries of existing definitions.
I prefer narrower definitions, because I want to emphasize the differences between a paper Zettelkasten and digital hypertext. Wider definitions make it easier to talk about commonalities. Both approaches are valid, but they serve different purposes.
In my opinion the qualitative difference matters, when we design and use our thinking tools.
I'm writing this comment on an iPad. I'm consulting my notes in Obsidian on the same device. An 80 years old article appears on the same device when I click a link.
Why use a plain Zettelkasten and books and libraries, when I hold a Memex in my hand?
Coming back to OPs question, I'd say that a digital thinking environment is particularly useful for the research part of the decision process.
What I do on my iPad resembles more the "trails" scenario in Vannevar Bush's famous Memex article than Luhmann's process.
@harr If you are stating mere preferences, I don't know how to meaningful interact.
But if the definition of hypertext is arbitrary, I like to throw in my definition of hypertext: A cute little animal that sometimes is said to hide eggs once a year.
The above reads much more snarky than I wish I could respond. So, I'd like to temper the tone with this meta comment.
The message is: arbitrary preferences don't help and if you are justifying opinion, it is not arbitrary. By definition.
I am a Zettler
You could explain why you prefer a different approach.
I think readers would appreciate the diversity of perspectives.
I'm totally fine with your definition. In a discussion about thinking tools I'd ignore it, because it's not relevant for the topic. But I might include it as a humorous reference in a concept note about the Easter bunny.
This can be resolved by attributing definitions to persons. "X thinks of a Zettelkasten as any kind of information processing with small pieces of paper." "Y thinks of a Zettelkasten as a close imitation of Niklas Luhmann's card index, including the medium paper and the numbering system." "Z thinks of a Zettelkasten as a medium-agnostic methodology that is characterized by …"
Two articles mentioned earlier in this thread (Ridi and Conklin) are execellent examples for this kind of attribution.
EDIT: The more I think about it, the more I enjoy your definition. It's obviously absurd. But what exactly makes it so absurd, that you preemptively apologize for snarkiness?
I think it's about mistaking the word for the thing itself. (Nominal fallacy?)
I'm talking about a thing: the qualitative difference between a Memex like experience made possible by digital hypertext — and having to physically look up stuff in boxes and shelves and buildings. I content that this difference matters when you design your thinking environment for decision making.
You're talking about words.
Your example illustrates nicely what happens when the thing and the word disconnect.
I suppose the meta-question to introduce at this point, considering the introduction of the Easter Bunny and a bunch of tangential conversations about hypertext, none of which have much of anything to do with my original question, would be:
What can all of this disagreement and differing of opinion teach us about the function of the zettelkasten (I'll leave the task of defining "zettelkasten" to each of you in your own time) in thinking things through and arriving at an eventual position?
I aim to get to the bottom of things and aim for value. Increasing reliability, utility, etc. of the knowledge built in collaborative thinking.
What you should do, is to point out that I am distracting and disturbing the collaborative thinking. And even worse: I would be muddling the water with asocial idiosyncrasy. So, ignoring this behaviour is unethical.
But I might include it as a humorous reference in a concept note about the Easter bunny.
This is omitting the justification given by the people, which lowers the epistemic value in the collaborative thinking. It is also not a helpful for the context here, because it doesn't work talking/writing with each other.
No, I am not.
I pointed out that your opinion isn't ill-justified by basing it on a qualitative difference without establishing that it's essential, which is necessary to be a proper justification. You then ignored this by stating that you're just sharing your preference. Adding to that, you claimed that the definitions are arbitrary preferences, while providing non-arbitrary justification for your preference.
My example is just driving the idea of arbitrary preferences for definitions to its final conclusion.
So, you can't have both. Either we are talking about arbitrary preferences, or we reason, explore and aim for collaborative learning.
@jameslongley Jump in, if you feel we abused your thread too much.
I am a Zettler
I aim to answer OP's question: "how (if at all) does your PKM system help you think it through (in practice)?"
I'm sharing what worked for me.
You could share more of what worked for you, like the "project -> Zettelkasten -> project -> Zettelkasten" cycle.
No. I claim that definitions are arbitrary and that it's a matter of personal preference, which one you choose. I claim that personal preferences can have deeper reasons. I claim that it is useful to make those reasons explicit, if you can.
For example I dislike semantic inflation, because it makes it harder to talk about differences and commonalities. If everything is a Zettelkasten, how do you talk about historical developments of note-making technology and methodology? How do you appreciate all those inventions that have occurred, if everything is the same to you? And if everything is not the same to you, why use the same word?
False dilemma.
There's value in sharing diverse experiences and diverse points of view. Especially when OP explicitly asked for personal experiences.
There's value in trying to explain why one chooses/prefers one alternative over the other.
For example, I'm taking a mostly pragmatist approach to thinking tools. I choose and shape my thinking tools so that they solve practical problems, that I actually have. As much as I appreciate theoretical discussions, the main driver for my PKMS is to solve real-life problems.
I enjoy forum discussions, because they broaden my horizon.
For example in this thread I learned about Ridi's article. It reminded me to look a classic literature about hypertext. Which in turn helped me better understand, why I find the integrated Memex experience more useful than purist ZK methodology. I realized that not everybody intentionally switches between different modes of thinking during the decision-making process (left-brain/right-brain, focused thinking/diffuse thinking, executive function/default mode network, cognition/intuition, …). I'm grateful for the discussion, because I learned something. Maybe it also helped some readers broaden their horizons.
Your brain seems to work differently. You seem to value other things. Maybe readers would appreciate to learn from you, why you prefer what you prefer? What do you expect to find at "the bottom of things"? How do you determine the "utility" of knowledge? How do you determine, if knowledge was "built" in a forum discussion? How does this "knowledge" help answer OP's question?
Currently, you are detracting with me together from OP's initial request and topic.
If you insist on labelling whatever you are having a "preference", I will partake in this semantic inflation.
Then I modify my initial statement:
Non secitur. A qualitative difference is not enough supporting reason. You have to establish that the difference is essential.
Your preference is not well-justified. But you may have it of course.
I am a Zettler