Image of a library
Photo by Luis Tosta on Unsplash

On the Nature of Understanding

There is a lot of discussion at the moment about "personal knowledge management", with people lining up behind their favourite apps, whether that's Roam, Obsidian, Tinderbox, Craft, Notion or even more "traditional" note-taking or note-keeping apps like Evernote or Notability or GoodNotes.

The concept of personal knowledge management is essentially simple: it is the collection of processes (with associated tools) which a person uses to capture, gather, organise, store, search, share, process and use knowledge — where knowledge is awareness or understanding of someone or something. Of course, people have been doing personal knowledge management as long as they have been people. The processes of becoming aware of and understanding the people and things around us is a fundamental part of being human and involve pretty much all forms of conversation, reading, teaching, learning and exploration. The advent of digital systems has changed anything to do with information and therefore knowledge radically by allowing us to be aware of and understand far more than ever before and giving us more effective tools to do things with it.

What to make of these new personal knowledge management tools? 

It's an exciting time. People are thinking hard about how best to capture their own and others' thoughts and knowledge, in the world where so much is available to us so immediately. There are genuine innovations and older approaches being used as models for new ways to make sense of things with computers.

Inevitably, there is more heat than light. There are lofty claims and entrenched positions. For example, the team behind Roam have a White Paper in which they claim that pretty much all previous approaches to capturing and storing knowledge on computers (what they call "the filing cabinet" approach) fails because it does not match the way our minds work at a neuronal level. The idea that "associative" organisation (where discrete items of information are stored with lots of unstructured and un-organised links to all possible related items) is always better than "hierarchical" (where items of information are grouped into topics and subjects and ordered by importance within them) is commonplace in a lot of current discussion. This type of thinking is also reflected in the movement to radically separate styles and formats (e.g. headings, sub-headings, lists and tables) from information content: with a preference for plain text and "atomic notes" (where textual content is short and contains exactly one thought or concept, without any hierarchy of main points, sub-points and examples.

I don't want to criticise this, or insist that the old approaches are right. I honestly believe that we know too little about how human minds and brains work, what information is and how it is best structured and used, how concepts emerge and people share them etc. to be dogmatic about any of this field.

Making the most of new opportunities

What I am fairly certain is that dogmatism in any new field holds back its development. If you rule out or limit some aspects of a poorly understood field, at least too early, you might easily close doors to ideas and approaches that turns out to be important.

I wonder if discussion about knowledge management and note-taking apps in general, and how to use them, might be more productive if we all remember that there are different approaches and systems for a good reason. People are complicated and we are certainly not all the same. That's good for lots of reasons. We don't all learn or think in the same way and almost all of us will change our ways of thinking or learning over time as we grow and develop, and even as we encounter and adapt to different challenges, projects and tasks. There is no single way of doing anything important that will work for everyone, all the time, and even where some ways of doing things are very much better than others, new discoveries, technologies and approaches may change that.

In short, I agree with Ali Abdaal who took some time (and lots of YouTube videos) to realise that there is simply no such thing as a perfect note-taking app.

I find the distinction between knowledge architects, gardeners and librarians helpful if you don't push the analogy too far. Architects want to lay out plans and structures and then fill in the details. Typically they will make headings and folders for all the areas they think they are likely to need and put new ideas, notes and captures in the right place to construct their structure. Gardeners want to let lots of ideas grow wherever and whenever they come across them, then find ways to organise them. What emerges grows, it is not built. Librarians want to capture and store everything and then catalogue it all. In reality, architects do tend gardens and do keep libraries, gardeners have to do some weeding and reading and librarians have to plan and cultivate some chaos to let things grow. We might have habits and preferences for different approaches, but people who are good at anything tend to have wider repertoires of skills and approaches than people who are less effective.

The best tools in these emerging spaces will, I think, work pretty well for everyone and allow them to approach the capturing, learning and application of their own and their team's knowledge and information in whatever ways work well for them and allow them to be as much of an architect, gardener, librarian as is right for them at that time and for that purpose. They will also permit users to change their emphases and approaches whenever necessary, without having to abandon their system or re-learn their basic skills.


I find Craft is the best app in this space that I have used so far, because it allows me to be an architect or gardener pretty well and especially because it allows me to restructure and re-use work very flexibly and without having to worry about wasting time and effort. I have Devonthink for my inner librarian and I can link to things in there as I need.

As a general principle, I'd say that software that imposes specific requirements for how you have to work with your data (whether that is proprietary formats or a particular system for how things have to be created and stored) tends to have a relatively short shelf-life, and is often helpful at first, but is something quickly out-grown, but software that makes it easier and better to do what you want, without a huge learning curve, is going to be around longer and help more people.

Philosophically, "every tool's a hammer". Tools that can be applied in a variety of ways to a variety of problems are more likely to fit with our personal, evolving processes and the ways in which we need to change the ways we work over time.