As you learn and grow, you'll accumulate more and more knowledge. While each piece is important, the relationships between ideas is, at least, as important as the ideas themselves.
Knowledge does not exist in isolation. It exists at the interconnection between ideas. Recognizing this fact is the first step toward better learning and understanding.
From there, it is obvious that building a knowledge base where each idea lives alone in its corner doesn't make much sense. Such a knowledge base is dry and provides little value. But what do good knowledge bases look like? Let's find out!
A solid knowledge base is one where information is easy to Locate and to Identify (i.e., the L and I of the 12.09.01 Data Organization - The LIFT Principle ). But that's just the tip of the iceberg...
Beyond organization and good naming, another fundamental element that differentiates good and great knowledge bases is the presence of links and backlinks between the different pieces. Links transform dry knowledge bases into valuable Knowledge Graphs (KGs). It's true for personal knowledge, but also for teams and organizations. Why? For different reasons...
In addition, solid knowledge bases have a categorization system that goes beyond containers (e.g., folders) and files/notes. Surfacing knowledge is key in a knowledge base. When you categorize the information it contains, you enable others (and your future self) to explore the content in different ways. This helps in many ways. For instance, it helps to discover topics and concepts that you may not have stumbled upon otherwise. Categorization is always imperfect (and unstable!), but it is still valuable. In a similar vein, Maps of Content (e.g., indexes) provide yet additional means to explore the content.
Links and backlinks enable navigating between ideas and concepts. Links and backlinks can directly link ideas together, but they can also do so indirectly (i.e., through other pieces). Direct links make connections obvious, but indirect links can also be very useful, as they can not only connect loosely coupled ideas together but also provide additional information and/or context. This helps to build up knowledge about a specific domain and/or related ones. When you dive into a solid knowledge graph, you basically dive into a different mind, hopefully containing more information than yours has. In my previous team, I wrote more than 800 pages of documentation about various systems, their usefulness, usage, configuration, dependencies, links, and underlying concepts. There was compounding value in creating that knowledge graph. It helped newcomers and experts alike and still proves very valuable today.
Categories (e.g., tags), Maps of Content, Links and Backlinks all increase the "explorability" of the content. It's not only about increasing the connectedness of the graph through different layers of connections, but also about enabling serendipity, the "luck" factor, while exploring the ideas.
The ability to search for specific content in a knowledge base is also very useful, as it enables quickly finding what you're looking for. As a side benefit, related search results can also uncover useful and interesting content that you might not have stumbled upon otherwise.
Last but not least, the novel ability to discuss with a knowledge base (e.g., thanks to Large Language Models) brings even more power to knowledge base users. For instance, using the Smart Connections plugin for Obsidian, I am now able to discuss with all my notes, which helps me dive even deeper into the wealth of knowledge I have accumulated over the years. By the way, the same plugin also finds potential connections between notes (hence its name!) by looking at their similarity (I share an example later in this article).
In the previous section, I've described different elements that transform knowledge bases into valuable knowledge graphs. Let's visualize these elements:
In the diagram above, there are two main ideas represented by notes in a Knowledge Base: A and B. A has a direct link to B (e.g., a Markdown link from A to B). That link makes it easy to navigate from A to B. Assuming that the system has support for backlinks, the link from A to B inherently creates a backlink that you can use to navigate from B to A.
In addition to the direct link, A and B are also connected by a third note, which might provide additional information (or context) about the link between A and B. That note is also part of the whole system.
There is also a Map of Content (MoC) that provides links to various notes that are somehow related. In this example, it points to both A and B.
Both A and B also include the same Tag T, which provides yet another way to find those elements of the graph.
Moreover, all the notes are part of the same "organization system". They're all stored together (e.g., in an Obsidian vault). Thanks to this, those notes can all be found by using the search capability of the system, which can usually find elements by their name or title, their contents, their location, their tags, etc.