Harnessing Your Metadata For Note-Taking and Profit

Notion, Obsidian — the tools don’t matter as much as the information retrieval design.

When you go looking for a piece of information in the world, how do you find it?

Broadly speaking, how one stores something also determines how that thing will be retrieved. Within the field of Information Science, there is a vast body of research exploring how ordinary people go about finding what they want.

This is no easy task.

What if the person who wants to know something doesn’t even know how to describe what it is that they want to know?

What if that person is you, in ten years time, trying to make sense of your own messy notes from college?

There has been a lot of excitement about tools for knowledge-work of late. Just a few of these are: Obsidian, Notion, Logseq, Bear, Roam, and Anytype. But the way these tools work for you must be understood before they have any value at all.

There are also popular systems, like the Zettelkasten Method, for storing information in such a way that the act of retrieving it creates synthesis (you connect your notes in a way that leads you to new ideas).

Both the zettelkasten method and these popular tools rely on something called representative linking.

“A representation is something that stands for something else (Tucker, 2023, p. 29). The important thing to remember about any tool or system is that they are only as good as the structure of representation you provide.

A famous sociologist named Niklas Luhmann created a zettelkasten of over 90,000 paper index cards. He credited this system for his wildly prolific scholarly output, which included some fifty books and five-hundred and fifty articles. But, it was the specific way he attached representational information onto his various cards that made his system so powerful. If we can understand that, then we can tailor bespoke knowledge systems that suit our personalities and individual needs.

Another term for something that represents something else is “metadata”.

Tools like Obsidian, Notion, or Roam function because of the metadata that we supply them with. While features like full-text search are also vital, there are numerous links that might be created by metadata that does not appear within the text itself.

“Whatever form metadata takes, it serves two basic purposes: to identify a document and to help the user find the document” (Tucker, 2023, p. 24).

I asked you to imagine yourself sorting through your notes from college. Maybe you had a system, or maybe you had none — either way, I know that improvements could have been made.

When it comes to designing your knowledge system now, as you consider working with powerful tools and systems like those I’ve mentioned, it’s vital that you start thinking about how you’ll retrieve the information you place into these containers.

This can seem alarmingly complex at the outset.

If you use the zettelkasten method, will you follow the metadata principles that Luhmann used, or a mix of those and someone else’s? Will you use hashtags, dates, or some sort of taxonomical system? A tree hierarchy?

It’s easy to get caught up in the different systems and tools that are available, but this misses the point of how such systems work. It’s not about rigidly adhering to a single system, it’s about finding ways to create overlapping representations of your information, to ensure smooth retrieval.

“How things were stored determines how you go about trying to retrieve them.” (Tucker, 2023, p.34).

You will likely find that your knowledge is best maintained by multiple strategies of storing and retrieving metadata.

Start with one or two that work best (perhaps dates and hashtags), or implement a system (like Tiago Forte’s PARA system, or Luhmann’s version of the zettelkasten) and let that lead your initial foray into knowledge organization.

You’ll doubtless develop a system that works best for you in time. And, if you keep this one principle in mind, you’ll find your future self thanking your past self quite feverishly:

Tucker, V. (2023). Information Retrieval System Design: Principles & Practice, Edition 6.2 (6.2). AcademicPub/XanEdu.