There are such kinds of tools which, when you interact with them, an organization emerges that modulates the interaction. That organization takes on a life of its own. The design and affordances of some tools for thought enable the emergence, persistence and adaptability of such an organization.
This is the fourth out of five posts of the series on Roam. The first part was about what is Roam like. The second was about the powerful concept of distinction, and the fourth built on that to explain self-reference in Roam and why it is important. The current one is on organization. The series started in April 2020 with the intention to be published within a couple of months. That wasn’t to be as my workload tripled1pun not intended but is welcome once I realized it: it wasn’t just the volume that was three times more than before but also it was mostly work with triples by that time. And although the productivity boost coming from Roam made it possible, cutting things was unavoidable and writing blog posts was one of the first things to go. Not go forever though, just wait patiently in the backlog.
Three things happened in the meantime. One, I gathered more experience working with Roam and more observations on how that structural coupling works. Second, my book on what makes organizations work was published in November last year. This means the current essay will be shorter than intended and the curious readers are invited to dive into more supporting arguments and details there. And third, the ecosystem changed. Now, apart from Roam, there are personal knowledge graph tools like Logseq, Athens, Obsidian, Kanopi, Codex, Foam, and Dendron, among others. They all have, in both similar and different ways, systemic properties that enable the interaction with the tool to be like with a living being.
Roam is a tool for organizing. You can use it to organize your day, project, research, knowledge, life. But there is also another organization, emerging from the interaction it2That doesn’t apply only to Roam, but also to some other PKG tools as those mentioned above.. This organization behaves as if it cares about its viability. It belongs to a class self-sustained systems, where the most typical organizations such as companies, government agencies, and NGOs belong to. Not surprisingly families, clubs, music bands, teams are also self-sustained systems of this kind. But it may seem strange to you if I put habits3Egbert, M., & Cañamero, L. (2014). Habit-based Regulation of Essential Variables. Artificial Life Conference Proceedings 14. MIT Press. and emotions4Colombetti, G. (2014). The feeling body: Affective science meets the enactive mind. The MIT Press. in the same class. There are good reasons to do so. Some say hurricanes are something like that too. A friend of mine also sees some dishes as self-sustained systems.
Self-sustained organizations emerge also in simple situations when we are trying to pass each other in a narrow corridor, but that lasts only seconds. Or during meetings. We tend to see meetings only as events. But they are social species, having a mind of their own, emerging from the complex dynamic of intentions, preferences, agendas, personalities, emotions, egos, and tribes, acting within the constraints of norms, rituals, and rules. This interplay between enabling conditing and constraints is always present in self-sustained organizations.
Do you need to interact with a software application for personal knowledge management to enable a self-sustained organization to emerge? No. Luhmann saw his Zettelkasten – not a software, just indexed cards in boxes – as his thinking partner:
As a result of extensive work with this technique, a kind of secondary memory will arise, an alter ego with who we can constantly communicate. It proves to be similar to our own memory in that it does not have a thoroughly constructed order of its entirety, not hierarchy, and most certainly no linear structure like a book. Just because of this, it gets its own life, independent of its author.
Does it happen only when you interact with tools for thought like Roam? Certainly not. It will be a matter of future research to find out when this happens, how, and when interacting with what sort of applications. A type of application that most likely belong to this class are videogames.
[G]ameplay is argued as being the achievement of dyadic and reciprocal coupling between a player and the game. In this reciprocity, gameplay arises as autonomous organization that is both self-sustaining and precarious. Coordination and exploration are offered as constitutive principles of videogame gameplay.
The viability of all these autonomous organizations, habits, emotions, teams, companies, and the one emerging when we interact with interconnected notes, depends on three essential balances.
The Essential Balances
Self-sustained organizations work when they are able to maintain three essential balances. If one or more of the balances are not maintained, these organizations don’t function well or simply cease to exist.
This webinar will give an idea about the balances.
In Roam, you write on a page, on pages linked to it, on blocks reached from their references and also on the query results. This autonomy brings a variety of ways to access, modify and link ideas and thoughts. But it’s only effective because of the cohesion brought by the persistent identity of the blocks, by the linking mechanisms for both blocks and pages, and by the daily notes pages, working as a spine. The daily notes pages bring cohesion not only in its structuring role (more on that later) but also by the fact that it’s the one type of page you can’t change the title of. A very important constraint.
You are free to create new “page references” in Roam in the context of a paragraph (block) or outside the context from the search bar. But this autonomy is balanced by the cohesion of the search itself, which regardless of the method, brings the labels of nodes in the graph, containing the sequence of letters you’ve typed. So, you can make another link, in a new context, increasing the cohesion of the graph, or, if what you mean is not available, create a new node but already in context (or not, if you do it through the research bar), creating a new relationship. Basically, you don’t need to decide what to do, create or reuse, until you see what you can reuse. And importantly, this happens in the context prompting the need for that act and decision. Compare that experience to tools having a button to create a new note.
Roam is somehow out of the way. This gives you space to freely create or capture new ideas (autonomy). But these ideas are stimulated by both obvious and surprising associations and increase their value through new relationships (cohesion). You can get an idea of how Roam brings cohesion for ideas from Joel Chan’s grant proposal entry point example.
The daily notes pages (DNPs) bring overall cohesion. And does so in several ways. You don’t need to think about where to write something. You have that choice but having a default place comes in handy. There is already a reliable universal relation by the simple fact that no matter what kind of content you want to add, that decision, idea or step in a workflow happens on a date. You don’t need to choose and write or create a new page and write but just write directly, making a useful and reliable relation when you do so. In a way, it’s also paradoxical: by talking out freedom, it gives more freedom.
The interaction with DNPs shows well also the second balance, that between stability and diversity. The balance is maintained at different levels. First, you have the stability of always landing on one and the same home page, your daily page, when you open the tool. But that’s balanced dynamically by the “home page” changing every day. It meets you with things you planned in past days and on non-DNP pages, by referring to this particular daily note page. It’s like getting messages from your past self. Second, as mentioned earlier, it is a type of page, the title of which you cannot change (stability). Yet, you tend to use that type of page the most. It works as a time-space hub to tame the wildness of the other pages (diversity). You have full control over them and so many easy ways to create, that they grow quickly and would get easily out of hand if there wasn’t for DNPs to bring stability and cohesion. But it’s not only the DNPs that bring stability and cohesion. There a plenty of other ways, for example, when using templates and pre-defined workflows (that part mostly comes through extensions such as Roam42). One such thing is attributes5It’s an unfortunate choice of a name, most likely inherited from Datalog’s attributes. I hope at some point they are renamed as properties to be in line with their conceptual role and with the knowledge graph conventions.. They work like directed edges that can have as values not only text but entities – blocks and pages reference, the attributes themselves working like page references. When attributes are used in an organized way, with stable semantics, they make querying and filtering even more useful. More on that in the next instalment. Another benefit comes from the fact that attributes are instantiated in blocks. So the same attribute type ID corresponds to one or many block IDs, where it is used. This makes reification, making statements about statements, very easy.
In Roam, almost every action is both exploration and exploitation (that’s the third balance). The possibility to create a workspace of different facets of your graph using the sidebar and narrow it through filters or reach one degree further by opening the linked references, increase the exploitation of the available explicit knowledge for creating new knowledge. Importantly, what the supporting resources bring to your area of focus is reciprocal as you can update these old resources with what you’ve learned in the meantime and what comes to mind at the current creative process.
This exploitation is balanced with exploration. New associations are created by bringing contexts of the existing links, and often trigger insights, new ideas, new relationships and open new paths for exploration. Exploitation and exploration are not just balanced but happen simultaneously. You explore while exploiting6Simulatious exploration and exploitation tends to accelerate the evolution of ecosystems. It is typical for example for mobile apps markets. By making it easy for the app writers to release new versions and simple for the app users to install and update, such ecosystems are different from the traditional software world. When releasing new apps and features, app writers explore the market while at the same time exploiting it. Each app and each feature work as an almost unbiased market survey. At the same time, they are actual businesses, actual exploitation, with revenue being generated either by ads or by selling the app.. You look for something that you brought in or noted. But in the meantime it got connected with other things, some simply forgotten, others not seen associated in this particular way. Serendipity, to find something when looking for something else, occurs rarely in life but with these kinds of tools, it can happen many times a week, and sometimes many times a day. You can also go into pure exploration mode by bringing random blocks or in purposeful exploration adding rules to narrow the pool of blocks.
There is a lot more to say but I’ll stop here. I believe that understanding how organizations work through the lens of the three balances can help designers of tools for thought create great applications and can help their users get more value out of them.
- 1pun not intended but is welcome once I realized it: it wasn’t just the volume that was three times more than before but also it was mostly work with triples
- 2That doesn’t apply only to Roam, but also to some other PKG tools as those mentioned above.
- 3Egbert, M., & Cañamero, L. (2014). Habit-based Regulation of Essential Variables. Artificial Life Conference Proceedings 14. MIT Press.
- 4Colombetti, G. (2014). The feeling body: Affective science meets the enactive mind. The MIT Press.
- 5It’s an unfortunate choice of a name, most likely inherited from Datalog’s attributes. I hope at some point they are renamed as properties to be in line with their conceptual role and with the knowledge graph conventions.
- 6Simulatious exploration and exploitation tends to accelerate the evolution of ecosystems. It is typical for example for mobile apps markets. By making it easy for the app writers to release new versions and simple for the app users to install and update, such ecosystems are different from the traditional software world. When releasing new apps and features, app writers explore the market while at the same time exploiting it. Each app and each feature work as an almost unbiased market survey. At the same time, they are actual businesses, actual exploitation, with revenue being generated either by ads or by selling the app.