Link pruning, information entropy and obsession

(Categorical composition, highly connected Zettels (HCZ), semantics, entropy fabric, segments with graphical complexity, pragmatism)

I was hyped by the idea of making a zettelkasten with less and less graphical complexity and more emergent connections over time so I could treat my works in a more organized and concise way.

However, I didn’t know how to link things. I put to myself the goal of making “coherent recipes”(a combination of “programs that do just one thing” and “food recipes”). I had a belief that entropy reigned the world of software and the world of cooking, that irreversibility was the limit and law of all connections in programming, cooking and my zettelkasten. I didn’t know what entropy was, LOL. While connecting things, I reached the logical obstacle that there were Zettels with many connections because they served as introductions to many other things. They increased the graphical complexity of my zettelkasten because many links pointing to the same node looks pretty cluttered.

That’s when the second moment of Zett serendipity struck. I previously worked with hashtags to try to make sense of the hierarchy mess in the zettelkasten of gastronomy. I thought that if I replaced some backlinks of a highly connected Zettel with hashtags, I would diminish the graphical complexity of the zettelkasten.

But which were the right links to be pruned? I realized that the best way is to prune the vectors that point from poorly connected zettels to highly connected zettels. When you do this, every highly connected Zettel gets closer to each other. This works like a high-power axon system that lets you connect any Zettel to other pretty distant and not-so-related Zettels. The more you do the repetitive tasks of linking and pruning, the farther, more numerous and varied connections any Zettel can get

That led me to a profound question. Is everything connected in some way? That question would haunt me for a long time. It also makes me guess if the zettelkasten itself has architectural properties (because if you can link from anywhere, that means you can chose whatever decision path you want, you aren’t burdened by early decisions)

Another way of backlink pruning is by taking 2 Zettels that point to a common highly connected Zettel (HCZ). If one of them also points to the other, then I just replace the hyperlink of it that points to the HCZ with its respective hashtag instead. This way of pruning is ordinal, doesn’t rely on the number of backlinks in a Zettel. The way I prune is by doing a ponderation of the ordinal and cardinal methods. The two have their merits and inconveniences.

One thing that I also began developing at this stage was the idea of categorical composition. It would help me make every Zettel smaller and easier to understand. The idea is that I don’t need to link all the ideas that are related to a concept inside a Zettel. I can put them instead in more introductory Zettels so they don’t clutter the Zettel I’m working with .They are used only where they are mostly needed.

The idea of link pruning was what opened the floodgates of the perks and problems of graphical complexity…

Relatively recently, I read another article explaining the rules of how to make a zettelkasten. It says that Luhmann used an organized guiding mechanism to find the principal ideas in his zettelkasten, a group of lists with relevant topics in certain categories. The way I guide myself through my zettelkasten is by what results of link pruning. This leads me almost immediately to the most connected zettels and makes me gradually recognize what is good and what is wrong with my zettelkasten.

Given the features of link pruning, it’s important to remember the definition I mentioned in Part 1:

“A zettelkasten’s graphical complexity is the complexity that results from segmenting and choosing a specific part of it. It’s expressed in how the different components of the extracted segment behave together. It’s inversely correlated to the amount of “emergent” links between zettels”

Read that definition many times if you can. Let it sink inside your head. What happens in the background when I prune the links inside HCZs? The answer is that the complexity of the graph surrounding it increases!!!

The fact that more heterogeneous topics get closer and closer and you “erase” the emergent links between some topics brings pretty chaotic graphs

All this means that if you cut the global zettelkasten (e.g. by generating a local graph of an individual zettel), you will VISUALLY FIND IRREGULARITIES. This means you can know which subgraphs of the zettelkasten are poorly connected and which individual zettels are excessively connected by the forms they show (the poorly managed subgraphs will look “not circular” or more specifically “not spider-webby”). Therefore, you can “engineer” the integrity of your zettelkasten little by little instead of trying to tackle the increasing sophistication and complexity problems of the global zettelkasten. You just need to improve its isolated parts one by one.

Everything I said is cool and awesome but a new question arose: How the heck do I know the ways in which the things inside those subgraphs are related? I am left with complex graphs whose shapes are known to me but I have no idea of how to connect the contents inside them.

• One way of solving this is by applying common sense to the individual Zettels that I encounter along the path. If I’m talking about water, I connect it with the moisture zettel
• Another way is recognizing the outcome of pragmatism (the more dedication you put into solving a problem, the more you learn about it and the bigger the possibility of solving it)
• Having different global zettelkastens also helps because you can translate some insights from one to others

None of these methods satisfied my crave for trying to understand the appropriate way of finding the links between Zettels so I could minimize the waste of time searching through the graph.

I don’t remember how, but I began to realize that some pruned graphs resembled the concept of the dark forest. Close to that date, I was also reading about the meaning of information entropy. Then, eventually everything clicked to me. The new graphs that spawned as a result of link pruning were full of information entropy!

The graphs had a high uncertainty about the prior connections between things, therefore they were throwing LOTS of information to me and I didn’t realize it.

As a result of this discovery, I thought that the new information provided by the zettelkasten would eventually be in conflict with the information I already had written (I mean, by seeing the graph, I can know which things are well-organized and which aren’t. If I had the be-all end-all piece of knowledge, wouldn’t I already get a perfectly simple spider web graph? Wouldn’t I know everything about past, present and future?)

My theory was that the global zettelkasten was forcing itself into becoming the Ideal spider-web graph but given that this feature is impossible, the only way to keep diminishing graphical complexity was by modifying my mind in wild and traumatic ways. The link pruning was trying to demand a change in my thinking paradigms, forms of knowledge I guarded as basically sacred and unchangeable.

This made me rethink many things. I had to rethink everything, every assumption that I had before. It makes you think that the “more correct” way of constructing your zettelkasten doesn’t even exist. “What is change? Does it even exist or am I inside an illusion? If change exists, why do I have to struggle to achieve it?”. This really put me in existential doubt and intellectual problems.

However, the more time we dedicate to solving a problem, the more we learn about it. In reaction to everything that was happening to me I started finding ways through the problems. I discarded some of them but it remained in me this: I can’t know the future, therefore the only way of surpassing my previous obsessions and zeal is by implanting even more potent and complicated obsessions.

And that leads me to the conclusion of this part. The zettelkasten is a tool designed to spontaneously transport you to your worst obsession and making you transition by chaotic and traumatic processes into even worse obsessions in order to become tidier. This is an unending process that generates internal uncertainty. The more you try to become one with truth, the more traumatized and blind with obsession you become. Some of you might describe it as hell on earth. Maybe you’re right. I describe it as the cost of change.

From this point, the part 3 is sort of miscellaneous. You can start now practicing the zettelkasten. Nothing will surprise you more than what your mind does when left alone with the zett.

One last thing I wanted to add is that problems will inevitably strike you and you will eventually feel impotent and alone. I repeat to myself: keep going, walk towards the future, use your ingenuity…