What do you do with your new ideas? The connections and sparks that pop to mind when you’re in the shower; in a presentation; on a bus.
In these settings I’ve had all sorts of ideas, from thoughts on the past and future of content management (which I still draw on from time to time), to a business idea for hot fizzy drinks (which rapidly lost its sparkle).
There are recurring themes too – I frequently consider the links between writing, thinking, strategy, and team empowerment – ideas I hope to share sometime. And every book I read sparks connections or mental debates with previous things I’ve thought.
For a decade, I’ve kept digital notes on all this stuff. In various tools, often in Markdown, sometimes in Evernote, and for an exciting but confusing time in Workflowy’s fractal outlines.
Some of the notes sprawl over many screens, and others (such as the hot fizzy drinks), are just one line.
But I found it hard to manage this information. Notes would reference each other, without any discipline. New ones would supersede or contradict parts of older ones but with no good way to view them together. I would find myself coming up with a brilliant new idea, and months later, realize I’d had a similar idea years before and the previous iteration was rather better.
This wouldn’t matter if the notes were just a hobby or collection. But I wrote them to be useful. And they just weren’t useful enough. To put together a blog post or to use some ideas in my work involved a long process of collating information from various notes and other sources. It was intimidating, to be honest. Sometimes I’d never finish a piece of writing, because by the time I’d found and assembled my ideas, they were less relevant.
How could this happen to me? Me, a trained information modeler! Me, a budding taxonomist! Me, a visualizer of complex relationships and now a manager of a software product suite! Why couldn’t I bring a little of my professional discipline to my notes?
At the beginning of 2020, I read a tweet (no longer available) by comedy writer Graham Linehan. His creative writing and ideas had been transformed by Sonke Ahrens’s book “How to Take Smart Notes”, about chunking ideas and linking them. This information scent was like catnip to a fluffy Persian and I jumped into the book and rolled around.
Ahrens’s book describes the unusual note-taking technique of Niklas Luhmann, a sociologist who published prolifically. As Luhmann researched and thought, he recorded each individual idea on an A6 size index card (around a quarter the size of A4 or US Letter). Each card had a unique ID that placed it in a sequence of related ideas.
He navigated his thousands of cards primarily by following the references on each card to related ones. With this manual version of hypertext links, he constructed an intricate web of ideas. This web was not an archive to which he would simply add new records, but a source of surprises as he rediscovered notes that would challenge his current thinking and force him to be more creative. His wall-sized filing cabinet became almost a living being, a time-traveling dialog partner. A co-author for the 70 books and 200 articles that Luhmann wrote.
Ahrens does more than describe Luhmann’s system. He shows us how building our own “slip-box” (or “Zettelkasten” in Luhmann’s native German) could make us similarly creative and productive. He suggests ways to apply the system to our own domains, using digital note-taking tools. (There is a brief taster of Ahren’s approach in his Quora answers, and some more context on applying the method in this interview with “second brain” proponent and productivity coach Tiago Forte.)
A key concept in Ahrens’s writing and that of other Zettelkasten practitioners is writing for the “future self”. Because we can’t know what future context we will use a note in, we cannot impose a rigid hierarchy to organize it, since our classification and framing of ideas will change. What we can do is to think of notes in relation to other existing ideas, to build a ground-up, flexible network that will accommodate future projects.
I tried the method, and some tools to work with it in. At first I tried Luhmann’s numbering technique directly, threading ideas in a single sequence. Later, I found that as threads branched from other ones or even merged, the single sequence made it hard to visualize alternative threads of interest. I experimented with the note-taking tools’ tagging and link previews, settling on a structure that let me focus on any thread that a note might be part of.
Did it work? Well, kind of. I could immerse myself in my notes for an evening, emerging with some useful connections and new ideas. But it took me a while to settle in each time. It was as if I had gone to a shopping mall where the various areas were distinct enough, but each shop’s facade gave an unclear idea of what was in there, and how much content was in there. Also, most of the shops were like a certain kind of bric-a-brac shop that I remembered from my childhood on the Isle of Wight (a kind of time bubble, 20 years behind the rest of the UK). Once you walked through the door, a dangling bell tinkling behind you, the things on display had faded labels or were piled between other stuff, so you couldn’t see what they were.
Reading around the issue, I found that other Zettelkastlers too were disoriented. Partly a few weeks after linking some notes together, it might no longer be clear why they had done so. (This sparked another mini research project for me and I did come up with a useful way to “infotype” links – but more of that in a future blog post.) However, much of the disorientation came from not remembering what a particular note contained by just reading its title or even the first few lines. From having to open the note and recreate the mental pathway that had led to it. Sometimes the connection had gone.
Perhaps part of the problem was that newer note-taking tools made it very easy to create new notes and even treat fragments of notes as linkable objects in themselves. Easy is good in general, but not if it helps you create too much content without pausing to lay tracks for your future self.
I wondered what was the ideal size for a granular, networked note, and how best to title it. The same kinds of questions come up in online technical communications – help topics, KB articles, more systematic bodies of searchable reference content. These resources have moved from books or book simulacra to chopped-up pages on the web. Too often, they still feel like bits of a book rather than natives of the online world.
Some years ago, Mark Baker wrote a book on how to write more helpful, structured web pages or articles (in tech comms parlance, “topics”). He observed:
Once they discover how easy it is to cleave one piece of text from another, writers gleefully chop their content more and more finely. The result is a sea of tiny fragments that may qualify as topics – they are grammatically complete and convey one idea – but which are impossible for a reader to navigate or glean any meaning from. When you reach that point, there are three ways out: 1. Give up and go back to writing books. 2. Create a way to string fragments together into something larger. 3. Reset your idea of what a topic is and come up with guidelines for creating topics that are correctly sized and structured to be useful to the reader. — https://everypageispageone.com/the-book/
The “sea of tiny fragments” sounded a lot like modern networked notes. The purpose of tech comms is different of course – to enable a task rather than to work out an idea. And the reader is a third party, an external user of information rather than ourselves. But, time evens the latter gap. Memory dims of our past self, and we haven’t yet met our future self. The past self had better write for the future one, making the same accommodations as for a forgetful acquaintance or a busy boss. Or indeed for an impatient reader of technical communication, seeking to accomplish something rather than admire the terse prose.
Baker’s guidelines for “correctly sized” chunks of information can help us write notes that will still be useful whenever we re-read them. He elaborates seven principles for writing helpful standalone topics. Topics that just like Zettelkasten notes, have a self-contained meaning and link to many others. Three of Baker’s principles are particularly useful for note-takers.
Specific and limited purpose: An EPPO topic has a specific and well-defined purpose. This is highly related to the purpose of the person who is reading it, but it is not the same thing. One topic has to serve many readers, and is designed to serve a community, not an individual.
Here, rather than a community of individuals, we address a variety of future selves in future use contexts. But those future selves still need focused notes on which to build and extend a line of thinking.
Long-time digital Zettelkasten practitioners Sascha Fast and Christian Tietze maintain a blog. One of Sascha’s posts is “How to Write a Note That You Will Actually Understand”, which echoes some of Baker’s points. Fast does not write about purpose, but about the form of notes. However, to write to this form will inevitably focus one’s notes. About titles, he writes:
If the title doesn’t tell you what is in the Zettel, it is useless. It should inform you and not just grab your attention. – A bad example: Being and cognition – A better example: No being without cognition – BETTER: Cognition is the process of constructing being — https://zettelkasten.de/posts/how-to-write-notes-you-can-understand/
Simply writing more focused titles has forced me to rethink the purpose of each note. If I were a philosopher, a note called “Being and cognition” would encourage me to catalog all the various different aspects of this subject, which would not create a very helpful note. A full sentence title that takes a stand is far easier to work with and link usefully to other notes. If in the future I found reasons to doubt that “cognition is the process of constructing being”, I could argue against the previous note. Arguing against a note called “being and cognition” would be like weightlifting with a barbell made of polystyrene. It wouldn’t accomplish much.
Baker’s second principle is that topics should:
Conform to type: It turns out that, unlike book length content, Every Page is Page One topics seem to naturally conform to fairly well-defined types, often the result of a community process that develops the best way to treat a particular kind of subject. The type of a topic is based on its purpose: the type defines the information necessary to serve its purpose.
This principle works for Fast too, who has:
… different types of Zettels. Sometimes a Zettel describes an entity. For example a hormone. Every Zettel about hormones has the same form: – Header – One sentence to describe its basic character – When it is up-regulated – When it is down-regulated – The effects of its change in concentration Every hormone has its own Zettel.
Later in his book, Baker explains how types of topics need their own structures. Without knowing it (as far as I’m aware), Baker and Fast have developed remarkably similar principles for granular, structured writing.
The third principle of Baker’s that echoes Fast is:
Establish context: Readers can come to an Every Page is Page One topic from anywhere. An EPPO topic must establish its context in the real world so readers knows where they are and what to expect.
Of course, Fast starts to prepare context for the future reader with a well-written, precise title. But the description of a note is important too:
The first sentence of a Zettel should be a short description to give you an idea what this Zettel is all about. You should give your future self an abstract. It will thank you because it can decide early if the note is important.
So Mark Baker and Sascha Fast agree on how to write chunks of useful information.
I believe that Baker’s work has moved modern tech comms forward. I see fewer examples now of the short book fragments that one has to keep clicking forwards on to form a picture of the subject. I see more medium length, article-like writing that works on its own, and doesn’t just describe an isolated feature but helps us understand it and other features in the context of our task. I don’t think all this comes from Baker, but he has helped to create the circumstances where writers feel empowered to write this way. (I believe he would feel that the field has not moved enough, particularly regarding templated structure, but anyway I think he’s made a difference!)
How about us busy note-takers? Do Baker and Fast’s guidelines help us when we come up with ideas on a plane, in a presentation, in the bath?
They do, but only when we take them as the latter part of a two-phase process.
First we need to get our ideas down quickly in whatever way is easiest. Paper notes can be great for this, as the physical action seems to cement thoughts into memory better. (Except in the bath, where it tends to cement them to the side of the tub.)
Soon afterwards, while they’re fresh in our minds, we need to write up our rough notes to file for the future. There’s a balance to strike. If we polish a note too much – if we try too hard to accommodate the future use contexts that we can imagine – it becomes less useful for those contexts that we can’t anticipate. It is like overfitting in machine learning.
Yes, this is more effort than the purveyors of some notes apps, and their fanatical devotees, would let us assume. If the apps were subject to medical regulations, they would have to be advertised with a voice rapidly reciting the small print: “can only enhance creativity and productivity as part of a balanced process of quick jottings and thoughtful write-ups. Side effects may include being subject to odd looks when you casually inject obscure references and German words into conversation.”
Does all this effort pay off? It certainly has done so far. I’ve only been writing properly networked notes for half a year, and some of that was trying different structures. But already it has:
helped me write blog posts at a greater than glacial speed (big win for me)
let me quickly locate relevant resources to send to colleagues in discussions
facilitated some proper strategic thinking about a couple of work things
So it comes down to some new ideas plus that dreaded descendant of the Protestant work ethic, “productivity”.
I don’t think this technique is for everyone, nor do I think people should take any such technique too seriously. A handy tip is that if you are spending more time rearranging your notes, playing with new tools, or writing about note-taking techniques, than actually getting some value out of the system, then you may have things out of proportion. (Unless this field is your actual job.)
This post was originally at joepairman.com, where Mark Baker commented:
“Without knowing it (as far as I’m aware), Baker and Fast have developed remarkably similar principles for granular, structured writing.” Without knowing it on my end, certainly. But I have always regarded EPPO as an observation, not an invention, so I would hope and expect it to be similar to other approaches that are founded in reading and comprehension, as opposed to those designed to solve technical content management problems. You are not wrong that I don’t think things have gone far enough. It grieves me that the tool space, in particular, is still dominated by methods that solve content management problems rather than writing and comprehension problems, and that support for rich linking remains so poor. But things have moved substantially nonetheless, and it is gratifying to have played some part in encouraging that movement.