Using Evernote as a Zettelkasten - Rough sketch on how I do it
Since @ctietze asked so nicely on Instagram for it: Here's a link to a blog post describing my current evernote-based system: Evernote als Zettelkasten.
I'm happy to answer any questions you might have. 
Howdy, Stranger!
Comments
Auf Deutsch!? Ich hab kleine deutsch. Und mein deutsch ist nicht gut.
I assume you would have liked some kind of summary in English?
Oh, not at all. I wouldn't come to a German website and presume to demand English from anyone. Really was just making fun of the limited German I still retained from highschool.
Thanks for sharing. I can’t read German, but google did a reasonably good job of translating (I think). I like Evernote quite a bit, but haven’t used it as my main notes repository in probably five years (primarily use devonthink). I am curious what you think about the new direction the company is taking? Have you tried any of the beta apps? I love how simple Evernote is and would love to give it another shot if I were confident in its future. But I have my doubts. Anyway, curious to hear your perspective on that if you don’t mind sharing.
I took the time to write up a somewhat shorter English summary of the rough sketch: Evernote as a Zettelkasten (cc: @kevin )
@cbirdsall I hadn't used Evernote for some years myself. Last year I was in dire need of a notes system that was low maintenance, could handle pictures, audio, documents, etc., was available on the phone and was still somewhat scriptable on my pc. I came from a Tiddlywiki based notes system that was very innovative (thanks to tiddly wiki's transclusion feature mostly), but it was a pain to use on the phone and took too much time to maintain. I looked at Notion and Bear and some other apps but nothing ticket all the boxes I needed from a notes app, except for Evernote. I was kinda surprised by this myself. I basically just jumped in, again. I only later found out that Evernote might have or had problems with growth and/or sustainability. But what I recently read about what they want to do, e. g. this blogpost, made me quite hopeful. It sounds all like solid and believable stuff to me. I haven't tried the beta apps that much (I mainly looked at the new web app), but am looking forward to semantic headers for example on the native client. I'm otherwise calm even in a scenario where Evernote would shut their doors since Joplin exists and the .enex-Export format is basically HTML.
@Matti the new CEO does seem good and the "behind the scenes" videos I've seen are promising. Upon reflection, I think the greatest threat to the longevity of any notes system I use my curiosity about some other system. Thanks for the translating the post!
Awesome! I’ll take a look. Thank you!
@matti
This is fabulous! Well thought out and well explained. I loved the part about using templates as prompts or nudges to intentionally work a note. The magical part where you reference Forte Labs and using an iterative process of developing notes really strikes home. Particularly the "Your rule of thumb should be: add value to a note every time you touch it."
Thanks for sharing.
Will Simpson
My zettelkasten is for my ideas, not the ideas of others. I don’t want to waste my time tinkering with my ZK; I’d rather dive into the work itself. My peak cognition is behind me. One day soon, I will read my last book, write my last note, eat my last meal, and kiss my sweetie for the last time.
kestrelcreek.com
@matti, in your blog post, you present your tag hierarchy and then say:
If you have the time would you share about adding a tag and the list of rules and practices on how to use the hierarchy? That would be wonderful.
Thank you for taking the time to do the English summary and being here with us! It is inspiring when people with deep thinking, like yourself, share their insights.
Your answer to the profile question, "How I use my Zettelkasten?" - Brillant! 👏👍
Man. Captain Disagreeable is active again: To the contrary, I have many Zettel that are truly finished. The incapsulated a specific thought that is like a platonic thought detached from any need of alteration.
To go even further: My benchmark for a very good Zettel is that I don't need to change it. I only would add to it if it is designed to be a receiving type (e.g. Evidence collector for some claims).
I am a Zettler
Jeez! So many reactions!
@Will The whole Forte Labs article is quite interesting. It might not be the Zettelkasten method, but many things are applicable nonetheless.
@MikeBraddock I'll try to write up something about my tag hierarchy soon.
@Sascha For Captain Disagreeable, I need some more words
Maybe it'd be better to say: Do not try to reach any ideal state on the first try? Anyways, the idea of iterating on zettels holds true until a Zettel is finished (or "finished" - however you conceptualize it). It's a heuristic or a good idea in general. Would you disagree with this?
That sounds like a great goal.
It sounds like you have a lot of time to write Zettel if you truly never edit them. Personally I would not be able to do my work and have a Zettelkasten, if I wouldn't work gradually on the later. The only other alternative I can think of: You keep every version of a Zettel as its own Zettel. So instead of overwriting what is there, you create a new Zettel for any improvement/change.
In theory, I'd agree with you the heuristic. And I'd agree with the statement that a finished Zettel is an ideal to strive for but not something you'd should bet yourself down with.
In practice, however, it is a very achievable goal in the majority of the use cases. And therefore, is a midterm goal of having it as a standard.
I am a Zettler
Really? In the majority of cases? Not for me. It could be a goal (we could call it 'midterm'… do you mean something like "mittelfristig"?) to have more Zettels finished from the start. But then… what are clear indicators of a finished Zettel? Won't these change with time? Why even take on the cognitive load of finishing the Zettel from start?
This still only makes sense to me, when deciding between "we have enough time at hand to finishing most Zettels at once" (defragmentation of time and attention) vs. "breaking this process up over many sessions" (iterating on what's good enough - for now).
Great post. Curious as to why you use separate notebooks? Why not throw it all together? Asking because it's been on my mind regarding my own setup.
I have found that it helps me to let things "flow". I was inspired by this post (the section about the workflow) (more the underlying ideas not the literal setup). Even though I might at some point add more processing notebooks to foster more long form writing.
Processing notebooks nudge me to touch Zettels again, stumble up on them, and so on. It's a step to optimize the note taking workflow for serendipity. I will see things again I have not seen in a while and that might spark new ideas and connections.
Imagine you are a philosopher and processing the Gettier problem. It should be very clear when you got it right.
Or, if you describe an experiment. With some practice you should get it right in the majority of cases.
Or, if you carefully analyse the thought of antifragility.
Most knowledge structures (arguments, models, evidence etc.) are not some blurry hard to grasp entities. What are the indicators that you got math problems right (back in school)? You just got it. If a thought is coherent and has not internal consistency probles it is finished.
No. The knowledge structures are the same.
In the end, it is less cognitive load because it is way less work.
The more session you plan the more warm-up time you need. Once everything is in your working memory, you have the opportunity to make it right from the beginning. The total number of Zettel will shrink in the beginnen. But the overall quality will be way better.
I think you think to pessimistic. For most people, this is quite achievable.
I am a Zettler
We're getting into interesting questions here…
The process of identifying those distinct pieces will take time. To grasp something is not immediate - or rather: it might be, but to get there takes more or less time, depending on many factors. At some point things might snap into place, but there is an investment of time needed. I think this holds true even for the best of "graspers". You might get better at grasping, but to be able to grasp costs you time - you learn a concept, you write it down, you link it, etc. Just take your buddy's (@ctietze) book processing Youtube series as an example: Does he not need a lot of time to process the book (no offense, @ctietze
)?
So my point about your point that you could reach a finished Zettel from the start in the majority of the cases has to do with the available undistracted time and attention and not with the shape of knowledge. It has to do with the practice of acquiring knowledge.
The problem, of course, is that all knowledge structures are context dependent. An historical artifact might be discovered to be fake, for example. A species might be part of a different genus. Etc.
And again: A finished Zettel might have more attributes than it just being a distilled sequential form of a thing I've grasped: metadata, aesthetics, etc.
It's also a question of what is interesting about this topic to me: I might want to drill deeper into a certain aspect of a topic, but not concern myself with other things related to the topic at hand. And this interest of mine indeed might change. I might be interested in the history of botany at first and then be interested in the history of plant geography specifically and then in the history of botanical gardens. This will lead me to write a lot of Zettels, some of them will be outlines or summary type notes, etc. - at least those will change with time. And then all the links, new information, other viewpoints on the same thing… Of course the shape of your notes will change, even if in theory knowledge structures might be stable.
An exhaustive approach ("I'll verzettel everything that has been mentioned in this one Book!") is in theory of course also possible, but see above: There is a limited amount of time I can spend.
If you are in a situation where you can plan your time like this (Student, Researcher, independent Writer, etc.), than you can chunk work in this way, of course. And that's great. As a developer working for a company, I can't (…fully, in the same way). But: warming up is easy if you are okay with the "good enough" approach, if you add value to your notes over time. If you at some point have grasped something, finally, you can trust that there will be somewhere a shitty first draft of a Zettel in your Zettelkasten which you then can add value to. And until then you at least have a semantic anchor to accumulate things around that might help you grasp the thing in time.
If by people we mean people who might be attracted to this approach of note taking, then I might agree. I think that my iterative approach is much more forgiving and therefore available not only to people who can choose session times as long as they wish, but also to others who are less privileged.
Oho, so now I have to chime in to defend my honor, good Sir! ⚔️
Ok, so the Epstein book series is fun to show how to thoroughly plow through an entire book. But of course this cannot scale. Look how much time it took to get to, what, chapter 4? 5? I also am under the impression that this book is a good pick to get started because it's easier to follow than, say, a book on Category Theory -- both for me and you all
But is it worth it? In this case, nah. I personally would pick maybe 5 or 10 Zettel worth of notes here and then discard the book. It's not a cornerstone of my study or of high relevance in my life. (That's some kind of context to keep in mind, too.) With superficiality, the quality of notes may even degrade. Or I end up taking far less notes than I'd like, because I discard drafts once I find out that I'd have to spend hours with primary sources I'm not interested in to really back up a point by Epstein.
I could write bad notes, sure. I did so, in fact. Some notes contain comments or placeholders, because I knew I couldn't wrap the stuff up in that episode, and because it got late, and this is not part of my work but instead lays dormant until the next episode, etc. -- these notes aren't yet rotten, but they aren't making me the best knowledge worker on the planet, either. I feel dirty for having kept them in such a condition. I should probably delete them because I know I don't want to go down the rabbit hole to make the note right.
Having said all that, I'm positive that there's a right way to write these crappy notes I now own, e.g. by backing up claims with the source material and sifting through all the information there. It's not easy, because I'm a noob in the field and wouldn't be able to properly process empirical studies if my life depended on it. So I'd have to learn that, first. And until I have, my notes on such studies would, very likely, be misleading or even wrong.
Is that a factor you have in mind, @matti? Like, not being proficient in the field/craft to get to useful evaluations?
Author at Zettelkasten.de • https://christiantietze.de/
Don't be so hard on yourself. The video series is helpful in ways beyond the actual "crappy notes" (as you call them) produced from processing "Range". Frankly, you could have processed "Justice for Hedgehogs" and I would have learned just as much about processing ideas into structure notes. Thanks for all your work on the videos.
Yes, as I gain more domain-specific knowledge my note creation skills should evolve as long as I don't get lazy. As I am able to devote more time to process my notes they will grow and hopefully reflect some part of reality. I view this as an iterative learning process and I'm just a noob. Long ways to go. I'm constantly reminded of the opportunity costs of exploring one tract in my Zettelkasten as opposed to some other as opposed to taking my dog for a walk and enjoying nature. None of this is easy, we are helping each other. Thanks
Will Simpson
My zettelkasten is for my ideas, not the ideas of others. I don’t want to waste my time tinkering with my ZK; I’d rather dive into the work itself. My peak cognition is behind me. One day soon, I will read my last book, write my last note, eat my last meal, and kiss my sweetie for the last time.
kestrelcreek.com
Then just create fewer Zettel. A created Zettel adds to your own critical thinking, your ability to concentrate on complex things and is a quality text.
No, they aren't. The knowledge structure is not the historical artifact but the interpretation and its asigned characteristic. The careful argument goes like the following: If the artifact is acurately described by X characteristic we can infer Y. That does not change with the fakeness of the artifact.
This connects to my point above: You'll get better results per time unit if you process carefully and thoroughly. I measure results in personal improvement and steps towards a system that improves over time. If you want to make a lot of notes than you are right, of course.
My point is quite the contrary. The iterative approach is less forgiving. It might feel better because you can add more notes to your pile. But you don't get to the threshold where the magic happens. (Which btw. is the reason why many people either don't have the endurance to stick to their system or have a wake-up to a pile of notes that is more difficult to handle over time instead of having a living Zettelkasten that get's more supple over time)
I am a Zettler
(Let me just say that I'm learning a lot by this and that I hope that I do not sound too defensive and/or aggressive towards the other participants in this thread.)
Not surprisingly we may not be that far apart after all.
I'll start with @ctietze:
Hehe. It was not my intention to imply that you didn't do go work or something like that, quite the opposite. I just took it as one example that in a practice many notes won't be finished on the first try. I think this holds true even if your great book processing series is of course a little bit of "Zettelkasten theater".
I take this to mean: Quality over Quantity. But that is not what I mean when I'm saying it's fine to start with subpar notes. (See below.)
Yes. But I would generalize even more: Even when being proficient there will be areas in your field that you haven't grasped yet. Or new things are coming up. An example from the field of software architecture: I might now a little about microservices but not so much about self-contained systems. This happens all the time. We don't need to make it sound like it only happens to noobs.
Next @Will (I think your handle is gread btw. because it makes me think "at will"…):
Absolutely! The book processing series is amazing!
Two things. As I said above, it's not only noobs. Hopefully you will process things that shake your foundations enough even after decades of knowledge work. That at least would be a good goal to have - albeit a little bit of an unrealistic one for all of us, perhaps.
The second thing: Lazyness is only a problem if you have to employ (too much) willpower. But in as much as this is possible, it should not need a lot of willpower to work in this way. It should become habitual to work with your Zettelkasten. And habit forming is more about having fun with it and good emotions about the process. I would even go so far as to say: Having to employ (too much) willpower for writing notes is recipe for staleness of any Zettelkasten. It needs to be more of a natural habit, IMHO.
And @Sascha
Absolutely! Your note taking skills will get better and better with time. And still it's a time bound thing.
The point I was trying to make here is not about being fine with a subpar end product, or throughput of books or amount of created notes. The point is: What can you do if your schedule is scattered? What can you do, if you do not yet grasp the thing but have to interrupt your session? My answer: Try to hone your workflow to 1. make it easy to continue where you have left of and 2. be okay with seeing your notes as being on their way to become mature.
Or be fine with being on your way to a great Zettel, soon! As long as you regularly come back to your Zettelkasten and you are still interested in the topic it's going to be fine to leave things only partly processed because you're going to get back to it. It doesn't matter if you have created 1 or 10 Zettels in a given time interval, either, because that's not what we're looking to optimize here.
Wow. I would like to know more!
But in any case: The fakeness of an artifact I might be researching or new Information about something or new ways of thinking about a certain concept, etc. surely will prompt me to rework certain aspects (Links for example) of my Zettel. I think this still holds true, no?
But it changes with the if clause in the beginning of the argument! Because it might not be accurately described. And how to judge this? By learning how to grasp. And how to do this? By investing time and effort. What if we get interrupted frequently? We need to get back to it. What does this mean for my Zettels? They might be in a state of inaccuracy.
I can't say that that's really the case. I feel like my approach has helped me a lot in understanding the world around me and my own ideas. The system I employ is serendipitous, which I think is a good indicator that the approach is working.
I agree however that neglect will lead to a stale Zettelkasten but this has more to do with biting off more than you can chew. This could be happening because you take on too many topics at once and/or it could happen because you try to be perfect from the start and you don't have the time for it. It however doesn't depend on doing it this way or another way.
Take a closer look at logical forms or argument forms. The form does not change, no matter what you know about the contents. Maybe maths is easier to grasp as a stepping stone? The formula
f(x) = 2xcan be instantiated with different values, sof(2) = 4andf(3) = 6. The underlying formula stays the same.With a statement of truth I imagine you have in mind, it's more like the classic
f(x) = 1/x, where you cannot pass0, because1/0is undefined. The formula is still valid, but passing inx=0does not produce a sound result.Bending the metaphor: You can still talk all day every day about the marvels of
f(x) = 1/x, not knowing that all people want to pass into it is0. Once you know what all possible values are, effectively, limited tox=0only, you may want to talk about a different function that produces more useful/interesting/... results for your listeners. That does not render your past talking aboutf(x) = 1/xobsolete: You can still plot its X/Y values, talk about its edge cases, compare it to other function, etc., even after you found that nobody will ever use it except withx=0. The only thing that'll change is you stop advertising this function as a good fit for the problem of all thex=0folks for pragmatic reasons.Author at Zettelkasten.de • https://christiantietze.de/
@ctietze I think I understand the argument, thanks for putting it so well. But I still think that that doesn't change the importance of my point.
My point is that when I say something like "all cows are blue", I can infer a lot from that, but that doesn't make it an accurate description of the coloration of cows. It is a possible description, but not an accurate one. Same with concepts: If I try to describe what Kuhn meant by "paradigm shift", my description might be simply inaccurate (however we might judge this). And I still could infer a lot of things even from inaccurate descriptions. And it is here where it is dependent on what I'm trying to describe. A decision about the accuracy of the description has to be made.
We can also say: We can infer Y even if the characteristic X is an inaccurate descriptor of the artifact, but then the inferred Y might not be something we can accurately say about the artifact.
The important part therefore is the decision about the accuracy and not what follows from X since accuracy is not a given.
Does this make sense?
@matti
The issue is that you are definitely right if it was a theoretical issue. You can always mess up and it is quite never productive to beat yourself if you fail at making it right. But in practice, it is a different story. I am doing it, it works, and I am (occassionaly) teaching it.
It is a question on what your standards are. I learned from practice that writing a Zettel that does not need any change for a long time (decades) is not only achievable but is the base standards one only need to deviate from very rarely.
You alway get what you aim for. If you are ok with mediocre you get below average results. If you aim for greatness you will achieve way more.
I have a calculation of the time and energy wasted with iterative approaches. But it is not presentable at the moment. It will come soon. Perhaps, my point will be clearer when I'll start with the public coaching.
For now, I have to shortcut the discussion. Though, it is quite important.
I am a Zettler
@Sascha I'm looking forward to your empirical argument.
But one last practical question, if you permit me: I often find the need to add links to already existing Zettels. Is this something you don't do? And if so, what do you do instead, if you stumble upon a new connection between two things in your Zettelkasten?
The discussion about the importance of context makes clear to me that the difference between @matti and @Sascha is of source material and disciplinary methodology. As I have raised in another thread, I suspect that @Sascha's methods include many principles that will work generally but some others that are specific to his field, type of writing, or type of research.
It is working in practice (I think) because it works for the type of research/writing you are doing, which is not heavily theoretical/philosophical. I guess we will see from the one-to-one whether it works on other things, depending on the topic, I guess. If you have a chance, you can also respond to that other thread (there or here).
I am worried here that you are confusing prediction with practice.
Do you even have a Zettel that is more than 10 years old? How can you possibly know what will change? I don't think it's about standards, it's more about someone not knowing what they don't know (and thus having confidence/certainty), and then learning what they don't know.
@ctietze and @Sascha are arguing, I think, that there are underlying knowledge structures that will maintain over time regardless of whether individual facts change. This isn't really the case. Only the most abstract logical structures maintain and these are useless unless you apply them to concrete aspects of the world using categories and concepts, which do change over time.
I'll give you one example. The concept of vision had an obvious meaning for hundreds of years, and we could not even conceive of the notion that we might see something but not be conscious of it. No one foresaw this apparent truism changing. Then we discovered blindsight. Even a basic idea like vision needed its Zettel revised!
Looking forward to someone correcting my overconfident claims as well.
Always and a lot.
I have to disappoint you. I do quite a lot theoretical and philosophical work.
In fact, my background is in philosophy. 
Answer is on its way.
I have Zettel that are over 20 years old.
I think I made a bad job explaining my point. It is not that a Zettel does change never ever, no matter what, come hell on earth, etc.
With a couple of years practice most Zettel do not need to change because one can learn to be quite precise.
I think we might have a different definition of knowledge structure. Knowledge structures function, as I use the concept, similar to what you refer as logical structures. Take a model for example. I have something that I call "hill model". Applied to the relationship of movement and nutrition it looks something like this:
2017-03-29-hunger-paleo.png
But I have applications to virtue, spirituality etc.
The hill model does never change. Its application might be inappropriate but nevertheless it is always in the form of "if.. then.." Perhaps, one could say, if you wrap your knowledge in a Modus Ponens you can make it eternal.
(Not true for those of don't know formal logic)
I come from a point of Philosphia Perennis and use tools I developed on the base of E.O. Wilsons "Consilience". That is a double influence: I construct knowledge in a way that it is more durable but also it is a filter for more durable knowledge. The later is a source of bias for me, of course.
I am a Zettler
Finally I caught you in an obvious lie. I have clear proof based on your videos that you are in your mid-20s at best, and thus there is no way you have 20 year old Zettels.
If all you mean by a knowledge structure is something like a abstract cognitive framework or a logical axiom, then of course you can create a Zettel describing that structure and say it's permanent. But I assume the vast majority of your Zettels are not describing such frameworks, but are instead making other kind of statements that apply concepts or structures to situations in concrete ways.
Maybe the problem here is just that this is all very abstract in the way we are discussing it. but what about my example of vision? You can say that were not really thinking of vision in a different way, it's just that we have a permanent knowledge structure and we realize the application was wrong. But, the advance in knowledge is the realization that our application was wrong, and we were thinking about the object - vision - in a misleading way. Presumably, the point of research is not to build up a network of internally consistent abstract knowledge structures, but to apply those to real world situations in order to answer questions.
I agree but this depends so much on your notion of what it means when we talk about science and knowledge, let alone truth (shiver!).
What is the goal, the task, of doing science?
Is it to give a most accurately described grid of propositional sentences mirroring the world?
Or is it to give tools to (re)construct experiences for the betterment of all mankind?
I am a Zettler, ie 'one who zettles'
research: pragmatism, 4e cognitive science, metaphor | you can't be neutral on a moving train
It's interesting; my intuition is that it doesn't actually depend that much about what we mean by science or knowledge. I think of (academic) research as a process of writing answers to research questions, which should ideally be designed to tell us things we don't know about real world situations that we care about. The notion that research is about constructing internally coherent networked grids of formal logical propositions is basically logical positivism, which has been roundly rejected. Such as goal is also antithetical to what we now know about the 4e cognitive science (embodied, etc.) way that humans interact with the world. (Based on your sig, @John , I assume you have an opinion about that!).
But maybe this was @John's polite way of saying the same thing without my rude confidence.
Or do you really think that there is a way to understanding knowledge and truth such that atomistic reduction of statements to logical structures leads to useful outcomes at a macro level?
I actually think that the ZK focus on everyone implementing the process in a way that best reflects their own knowledge structure and goals is a reflection of the importance of context in determining the meaning of claims.