Conundrum

Speedrunning my Computer Science degree [100%] (2/6)

Mon, 15 Dec 2025 00:00:00 +0000

Clickbait Title: Lazy evaluation.

Background

Interim

As previously said, this entire post is mostly motivated by spite. Now, sadly, this can be interpreted as some kind of exceptionalism or superiority complex which I hope to thoroughly avoid. The spite is instead rooted in being misrepresented to a wider audience of people, specifically ones which may be interested in studying Computer Science.

I think the tone is hard to judge here, because speedrunning tends to convey some level of effortlessness and the jokes make it feel as if the narrator feels above the material. Neither is true and I hope indeed that by sheer effort that goes into writing so much, I prove that I do this because I love the subject matter.

Speedrunning my Computer Science degree [100%] (1/6)

Sat, 13 Dec 2025 00:00:00 +0000

Clickbait Title: Brevity is the soul of cowards.

No, it’s not “so over” for you if you did something differently here. No, this is not accurate for everyone, even my peers. Yes, some details have been changed for reasons. I will forget things and say wrong things, I’m not pulling up my old notes for accuracy’s sake. There will be jokes until morale improves.

Background

After avoiding this video for months, I finally relented and spent the precious 3 minutes of my life to watch it. I was disappointed and almost shocked. Surely you learned more than that in 3 years? So I’m not going to name my article after this video, but rather the much better one with the same premise.

The Death of Problem-Solving Anticipation

Sun, 30 Nov 2025 00:00:00 +0000

Clickbait Title: The Birth of AI Gambling.

We joke around a lot about programmers being awake at impossible times. Not sleeping for a night for a competition or a personal project is casually mentioned. Why is that? Sure sometimes you’ll be fixing up-time or only have 24 hours in a CTF, but come on, we’ve all worked into the night when we could’ve just gone to bed and worked on it the next day. I find that remarkable.

Writing 24000 words in 30 days

Wed, 19 Nov 2025 00:00:00 +0000

Clickbait Title: My greatest achievement in months.

I’ve been writing a bunch of articles. I’m going to talk about what I wrote and how it felt.

I’m obviously glad I kept myself more or less to writing an article a day. That is an achievement by itself. Making this something I needed to do greatly reduced me caring about what I wrote and more just that I got something out. This is good, it means I did not scare myself out of writing until I have the perfect idea.

Ponds

Tue, 18 Nov 2025 00:00:00 +0000

Clickbait Title: Gifted Kid Anthem (Remix)

As much as I try to fight it, I tend to strive for comparison between myself and my peers. It gives me motivation to not want to fall behind, to not only continue to understand what they are talking about but to be potentially useful (i.e. better) in at least some aspects.

On the other hand, it is very advisable to always search out new people who you can learn things from. People have sung a lot of praise about being the “stupidest person in the room”. I tended to do this (unconsciously) by befriending people in higher semesters than me. They are taking very interesting courses and I wanted a piece of that, early.

Close Encounter with a Heron

Mon, 17 Nov 2025 00:00:00 +0000

About a year ago, I took a path through some agricultural pasture. I was exhausted from work, longingly missing my bed and fridge. My 40 minute commute was threatening to turn into a 60 minute one, if I would not hurry to my bus. And as I was passing by a free-standing garage, I saw a bird standing proudly on top.

It was a heron, towering over the landscape on its unnaturally elevated position. You normally see them in grasslands and wading through ponds. This one had chosen this garage as a vantage point. Maybe it was still young, the feathers were ruffled, as if it had grown faster than its plumage.

Rhetorical Double-Jumps

Sun, 16 Nov 2025 00:00:00 +0000

Clickbait title: Being Wrong Twice is more effective than being Wrong Once.

So here’s a pattern in bad-faith discussions I’ve seen several times.

A: I believe that bananas are fruits.
B: No, bananas are not fruits.
A: Why not?
B: All fruits come from a tree and bananas don’t come from trees.

B is obviously wrong here and we can even explain two different ways in which they are wrong:

Music as the most effectively vague art

Sat, 15 Nov 2025 00:00:00 +0000

Clickbait title: My OCs are having crazy fights to the best soundtracks.

God forbid I know what art actually is, but I would describe myself as a person who has had little artistic training. I have neither learned how to draw, write or play an instrument beyond standard education, in which I was mainly trying to survive art classes.

In stark contrast, I harbour impossible respect for the arts, maybe even jealousy. I love to read, look at and listen to art. Additionally I am one of the people who gets stung by the project bug. I need to be on the active end of creating something new for the world at all times. And so I have tried writing (somewhat successfully?) and drawing (unsuccessfully), but I have found music to be completely impenetrable.

On my style

Thu, 13 Nov 2025 00:00:00 +0000

Clickbait title: Why do I write like that.

This article will lean a bit hard on meta-references and self-analysing. Reader discretion is advised.

Having now written more, I think it’s a good time to look back and see what kind of stylistic choices I make consistently when pushing myself to write articles.

Looking at the choice of topics, I would argue that I tend to write most about things that are currently bothering me to the point I want to dissect them publicly. I choose issues which are general enough I can feel like I am actually contributing knowledge (and don’t reveal too much about my personal life), whilst still clearly being the kinds of issues that are very me to think about.

How to change your mind

Wed, 12 Nov 2025 00:00:00 +0000

Clickbait title: Don’t argue on the internet.

There has a lot of ink been spilled about how to change other people’s minds, some people call it debate, others calls it campaigning, others call it propaganda. Changing other people’s mind is desirable for many, because it is seen as giving you power over them. The natural reaction is of course that you my dear reader should not be vulnerable to having your mind changed. You’re clearly smarter than some evil outside influences.

Good Media is remembered fondly

Tue, 11 Nov 2025 00:00:00 +0000

Clickbait title: Finally someone solved the the age-old question about what is good media and what isn’t? What do you mean the answer is subjective, how am I supposed to feel superior in my taste now.

Now obviously, quality is kind of a stupid concept if you look into it. It also obviously exists, but trying to measure or debate it only leads to misery. We still do of course, and this is how aggregated rating systems work etc. And these are of course never flawed (don’t look up why Youtube got rid of the star system).

Conversation in a Cave

Mon, 10 Nov 2025 00:00:00 +0000

I like to go exploring sometime. Today I stumbled on a cave in my garden. There’s a rough stone-cut circular staircase leading down a cylindrical well. At the bottom there’s a large bonfire. Curled around it, a large dragon, watching the shadow of the flames play on the wall.

Obviously surprised to find a dragon so close to me, I hesitate. Since when has it been here? Why have I never seen it outside? Is it aggressive?

Why I don't participate in note-taking culture

Sun, 09 Nov 2025 00:00:00 +0000

Clickbait title: It’s just goddamn text files, how the fuck did you mess up text files.

Usual disclaimers

Obviously I take notes to remember things. I have sung the praises of Obsidian before and have since not been using it much, although many articles of mine are written in it.

Look this article will have immense overlap with the best video on note-taking culture.

What do I mean by note-taking culture? Really, I am including an incredibly wide range of productivity tools and “tips” surrounding text. The idea that spending time creating custom organisational systems are worth it. I will neither give a “takedown” of this culture, nor will I try to sell you on my way of doing things, but I will try to convince myself I was correct in never wasting my time on this stuff.

Celebrating Undazzling boardgames

Sat, 08 Nov 2025 00:00:00 +0000

Clickbait title: The SECRET connection between boardgames and theatre that videogames and cinema will never achieve.

There’s a lot to be said about how easily we can now find entertainment at the press of a button on our phones. The economics of digital media are truly staggering to the point we can have a somewhat sustainable system of low-cost subscriptions that give us everything we could ever watch/do etc.

Digital screens are also incredibly powerful in their ability to render basically anything we want, as long as we can tell a computer which colours the pixels should be. This has given us anything from the concept of editing to CGI.

My meta-saturated rot

Fri, 07 Nov 2025 00:00:00 +0000

Clickbait title: Tertiary source complains.

One of the things which lingers in the background of every social media diet is that the main topics of social media tend to be things not happening on social media. You may talk about your graduation on it, but the graduation is not happening on social media (I hope), it’s happening in real life. You are probably also not going to watch a full movie on social media (unless it was artsy or old enough to be on there). But what you can watch easily are reviews, discussions and commentary.

Is 4x video speedup a sign of the end-times?

Thu, 06 Nov 2025 00:00:00 +0000

Clickbait title: Hey when I get inevitably popular for my great writing ability and someone reads out this article on a live-stream and the live-stream gets distilled into a YouTube video, be sure to watch the video at 4x speed.

It is no secret that I have chosen YouTube to be a major part of my media diet. Only speaking about video content, I’d say 95% of all I’ve consumed in the last 5 years has been YouTube videos. There’s a lot of reasons for this. YouTube has/had a lot of videos I was interested in. It also has addictive design. But I think I can blame the speedup feature in great part.

Lean for learning Mathematics

Wed, 05 Nov 2025 00:00:00 +0000

Clickbait title: Is Lean about to do the impossible?

Some techno-optimism for a change. Disclaimer for people who know more about this than me: I am not a contributor to Lean or MetaRocq and am stepping into these programs as a Computer Science student and primarily a user with some interest in understanding their internals.

Depends on what sphere of computer science you are in, but if you are staunchly not interested in Machine Learning, then I think there are few things as interesting as the grassroots movement of Lean4 and Mathlib.

Suspiciously targeted effort

Tue, 04 Nov 2025 00:00:00 +0000

Clickbait title: Not to judge, but why are you spending all your time complaining specifically about this?

I think we are all aware of the concept of lying by omission. You make a statement that is not technically wrong but is clearly disingenuous in the way it does not mention something quite important. Well, I think the internet has given us a new way to do it.

On the internet you tend to exist as an anonymous person which pops in and out of existence for others. If you are not currently livestreaming yourself from 5 angles, then people are not seeing all you’re doing. And if you stay between different accounts, it’s going to become increasingly hard to track what you actually spend your time with. This is generally good, I find it good that the internet gives us this asynchronous way to communicate.

Remembering Despite

Mon, 03 Nov 2025 00:00:00 +0000

Clickbait title: A simple world with no contradictions.

It is no secret I believe the desire for a simple explainable world is responsible for many ills from conspiracy theories to abuse of power. I also believe that the heightened competition in the attention economy is partly to blame for this. Nowadays you need to sell people a good story that instantly fits a comfortable framing. And of the most comfortable framings we have is one of causality and “deserving”.

Dragons as self-empowerment

Sun, 02 Nov 2025 00:00:00 +0000

Clickbait title: Individualistic fantasy in fantasy.

Dragon is a strong word. It is certainly not meaningless, but it is very flexible in what it can be applied to. I do not think we can find some platonic ideal of a dragon, but I want to discuss the complex emotions the concept gives me.

Dragons

Firstly, dragons are powerful, certainly physically and sometimes also magically-gifted. In basically every story a fully-grown dragon has the agency to do whatever it wants. This means that in most stories there has to be a good reason why the dragon does not intervene, often its boredom, self-interest or some cosmological balance.

Games where you lie

Sat, 01 Nov 2025 00:00:00 +0000

Clickbait title: Games where you lie to your friends.

I have been and will continue to be interested in boardgames. I think they are a really fascinating discipline of game design, having completely different limitations that videogames. I think computer scientists tend to be used to stuff like infinite dices you can spawn or complex calculations you can run, whilst being starved for the magic that being in the same space can do for social interaction.

Sharing, Recommendations and Utility

Fri, 31 Oct 2025 00:00:00 +0000

Clickbait title: Gifts. Again.

I’ve found myself critical of gift-giving on this blog before, but just as then (and with most things) what I’m really seeking is to justify emotional reactions. But introspecting to that has me thinking about this in wider concept.

I think I could write this as an introduction or a conclusion: It’s good that people interact with each other. No matter what else, there’s some seriously positive things to do anything with others. It will be inevitable that social problems appear, but the correct solution isn’t to then avoid all social interaction. The people around you might suck, but being with people, in general, doesn’t. We have an entire tradition of storytelling of outsiders who finally find “their people” and open up.

Puzzles vs. Problems

Thu, 30 Oct 2025 00:00:00 +0000

Clickbait title: Not everyone can make good puzzles.

I don’t know how common this experience is for other’s who have “always been interested in maths”: All my life, people who wanted to gift me something have defaulted to puzzles. Sudokus were probably the most popular. I don’t think I ever solved a single Sudoku in my life (maybe as homework at age 7?).

I do not want to sound ungrateful for any gifts I received (despite my generally low opinion of gift giving as a mandatory tradition), people put in as much as effort as they could bear to give me something. And for a lot people I think mathematics looks very similar to “logical thinking puzzles”. But my interest has always focused on the former, whilst disregarding the latter.

Mfw when rust

Wed, 29 Oct 2025 00:00:00 +0000

Clickbait title: Local programmer cannot believe how good Rust is.

I’ve taken a crack at learning C++ (again). I’m approaching this the same way someone would approach reading Shakespeare: It’s a classic, you gotta know it. The most I can say about it is that I’m learning a lot about templating and preprocessors.

Having stayed optimistic with C++, I’ve tried starting projects in it. I tried, and I went back to Rust. This is a me problem. There’s a joke going around that you should learn Haskell (or the obviously superior OCaml) first before you’re learning Rust. And it is kind of true. Not because the knowledge from those languages is essential (Result Monad is very neat tho), but rather that if you’ve ever used stuff like typed unions (= Haskell datatypes, Rust enums), you will miss them so much.

Striving for something

Tue, 28 Oct 2025 00:00:00 +0000

Clickbait title: Getting what you want is not what you need.

I consider myself to be a very satiated person. I’d say as soon as I was no longer limited on screen-time by my parents, I did not have much to wish for (massive privilege!!). Importantly this does not mean I am happy all the time. And a lot of considerations of adulthood are still sneaking up on me. Not to speak of the state of the world.

A lukewarm defense of gatekeeping in technical spaces

Mon, 27 Oct 2025 00:00:00 +0000

Clickbait title: That one YouTuber who keeps copying from ChatGPT and watching tutorial videos instead of just reading the Arch wiki.

This is a discussion mainly informed by online coding forums, communities and open-source projects. The author is not a prolific contributor.

I think the stereotype of the elitist tech user does not even need to be mentioned at this point. They think they’re better than you or are at least projecting that online. It’s no fun interacting with them. I am not going to defend this behaviour, but I will try to motivate the reasoning behind it.

Chronological Feeds are recessive

Sun, 26 Oct 2025 00:00:00 +0000

Clickbait title: Algorithms are hostile.

If you estimate my age or read other articles of mine, you are aware that I have grown up during the awkward transition period between chronological feeds and algorithmic recommendation systems. I know why people bump a thread, but only learned about RSS as a thing old people use.

As my free time has decreased and my standards increased, I have come to appreciate chronological feeds for a variety of reasons. They allow me to easily stop and leave when I have reached posts I already know, defeating the slot machine effect seen from algorithmic feeds you can refresh.

I loooove Defaults

Sat, 25 Oct 2025 00:00:00 +0000

Clickbait title: Why I use KDE.

If you have the pleasure of being around Computer Scientists, you will find that they have peculiar ways to go about things. Now some of these things actually make sense, in the logical abstract sense, and everyone agrees on them. This is a myth. Name one such thing and I will find a Computer Scientist that calls it the worst thing.

But there’s two other reasons why Computer Scientists insist on doing something. The easiest and worst one is “We always did it this way”. It is also the most dominant one, because inertia is the most powerful force in business. We call this “Windows”. Maybe even “Windows XP”.

The active act of reading

Fri, 24 Oct 2025 00:00:00 +0000

Clickbait title: Why nothing will ever come close to text.

Is it cynical to praise your own medium or is it just practising what you preach? Everything which I will write here, will already have been said by experts in a more concise and constructive way.

The first mass medium

There are some clear material reasons why books became the first true mass medium (and for the same reasons architecture was the first mass medium before it). What I find fascinating is just how long they’ve stuck around. When we pick up a piece of paper, read a tagged wall or shuffle through pages, we are basically mimicking the consumption of like historical people, real historical people.

Phones and Slow Tigers

Thu, 23 Oct 2025 00:00:00 +0000

Clickbait title: Call your trusted parental figure and tell them what you’ve been doing all day.

Warning: Autobiographical

If previous articles on personal feelings weren’t enough, I’m now writing about a revelation that comes straight out of the self-help section of the book-shop of life.

This video is obviously currently going apocalypticly viral. For people who don’t trust my nebulous links (safe internet browsing!), here’s the summary which incidentally reveals more about me than the video:

Becoming an expert is being annoying

Wed, 22 Oct 2025 00:00:00 +0000

Clickbait title: Why YouTube does not feel fun anymore.

Warning: Autobiographical

My dirty secret

One of the articles for this blog which has been in the draft-graveyard for years is one which attempts to enumerate what easily accessible online resources have influenced me.

Although said article will probably never see the light of day (for boring reasons), it has helped me organise my media consumption in an interesting way. Especially because it’s all goddamn Youtubers.

The cuboid shape of rules

Tue, 21 Oct 2025 00:00:00 +0000

Clickbait title: Death by a thousand cuts exists for a reason.

Before I can make my argument, I will need to warn of the dangerous confluence of geometry and anecdotal evidence underlying this.

In normal words: I’ve noticed, that… and it reminds me of…

Rules

Compared to informal friendgroups, as soon as one creates some sort of formal setting, often for people who are strangers to each other, one defines some rules. Why have rules? For many reasons of course. Mainly as a way to create shared understanding of what the goal of the group is.

A taxonomy of AI criticism

Sat, 02 Aug 2025 00:00:00 +0000

Clickbait title: A Modern Luddite Guidebook to AI Discourse

I do not specialise in AI in my research.

So what is AI anyway?

Artificial Intelligence is an old and very general term. That being said, most people engaged with the internet in the last few years know that the term has taken on a more accelerated meaning.

It was arguably birthed from the more measured hype around Machine Learning ever since maybe AlexNet. What we consider AI now are mainly diffusion image models and generative transformers for text, with some audio and video generation thrown in for the mix. This is how I will also be using the term.

For God's sake dont trust GPT for uni

Sun, 08 Sep 2024 00:00:00 +0000

Clickbait title: I will fucking piledrive you if you insist on an answer given by ChatGPT again

I do not specialise in AI in my research.

In high school I wrote a paper on the philosophy of Wittgenstein and how it relates to the concept of a programming language. One central question was if computers would ever be capable of understanding natural language. I cited the GPT-2 paper as state of the art. Leading up to my time in university then, I was primed to GPT-3 being hyped up by select people have gotten access to it. This Tom Scott video is a good time capsule. ChatGPT was released in the middle of my first university year, and it took less than a month for someone to copy-paste entire university chats into it and letting it answer. They were figured out after the third message and claimed it to be a social experiment. A social experiment with only 3 data points is lousy, so I wonder how much faith they had in it passing the Turing Test.

Understanding the CRS Matrix format

Tue, 26 Dec 2023 00:00:00 +0000

The compressed row storage (CRS) format is a matrix storage format for sparse matrices, which is most useful for matrices which have a linear number of non-zero elements (compared to dense matrices having a quadratic number of non-zero elements).

Storage

The format takes a $n \times m$ matrix $A$ and turns it into the following arrays:

The val array stores all the values of the non-zero elements of the matrix, therefore it has length $\textrm{nnz}(A)$ (the number of non-zero elements in $A$)
The col_ind array stores for each non-zero element of the matrix the column index it is in, therefore it has length $\textrm{nnz}(A)$
The row_ptr array stores for each row when that row starts in the other two arrays, plus it always ends with the sentinel value, therefore it has length of $\textrm{rows}(A) + 1 = n + 1$

Therefore, the number of values we need to store a $n \times m$ matrix is $$2 \cdot \textrm{nnz}(A) + n + 1$$

Reliability At Scale

Sun, 02 Jul 2023 00:00:00 +0000

A framing device for the role of Informatics in our shared world.

This is phrased as if it is an objective theory which has found truth in the world. It is not. It is written in that style because that style lends itself to being a heavily abstracted theory. All mistakes are my own.

Theory

Informatics is principally the study of manipulation of information. This manipulation is typically done along a given goal, which consists of the description of the input and the desired output. A so-called computation of this output can be called reliable if from the described input the output is produced according to the specification.

A Brief Guide to Partial Fractions

Mon, 13 Feb 2023 00:00:00 +0000

Motivation

Given a complicated fraction, it is often favorable to split it into two separate fractions as such: $$\frac{1}{6x^2 - 5x + 1} = \frac{2}{2x-1} + \frac{-3}{3x-1}$$ But how to do that can be very obtuse, so here is a general guide for the most reliable (but also most time-consuming) method which uses systems of linear equations.

Step 1: Factoring the bottom part

This is self-explanatory, try to find any way you can split up the bottom part of your fraction into two parts. In the given example: $$6x^2 - 5x + 1 = (2x-1)(3x-1)$$ This may be done by finding the roots of the polynomial (via the midnight formula) or just trying a few values out.

The Cache Formula

Mon, 13 Feb 2023 00:00:00 +0000

The formula

$$ \textrm{Set}(x) = \textrm{Set}(y) \iff \left\lfloor \frac{x}{B} \right\rfloor \equiv_{S} \left\lfloor \frac{y}{B} \right\rfloor $$ where $$B = \textrm{Blocksize}$$ $$S = \frac{\textrm{Cache Size}}{\textrm{Associativity} \cdot \textrm{Blocksize}}$$

Motivation

Learning about how caches work can be a very disorienting process to say the least. Worse even, learning theory may not exactly be correlated with being able to actually determine if a cache access is a hit or a miss, as one can be caught up in the hardware. Here I present a formula which abstracts most of this thinking away in favour of a very mechanical process.

The Extended Euclidean Algorithm in Finite Fields

Mon, 13 Feb 2023 00:00:00 +0000

This article has been adapted from an earlier PDF I wrote.

Motivation

Given that several operations in discrete mathematics require one to find the inverse of integers or polynomials in finite fields, it is important to learn an efficient algorithm to do so quickly. The Extended Euclidean Algorithm is the most primitive of these algorithms and essential for students.

In this article, I will explain use this algorithm on a few example problems, hopefully giving some intuition to future students.

The Intuition behind Error Correcting Codes

Mon, 13 Feb 2023 00:00:00 +0000

This article may require some additional technical knowledge.

This article has been adapted from an earlier PDF I wrote.

Motivation

Alex wants to send Bernd a secret code $m$ explaining their plan to take over the world. The only problem is that local Iris-agents are jamming their communication, and therefore Alex cannot guarantee that the entire message will arrive unchanged. Is there some way that Alex and Bernd can develop a protocol that ensures that Bernd is guaranteed to receive the correct message from Alex?

About

Mon, 01 Jan 0001 00:00:00 +0000

About Me

I’m a Computer Science Student who is interested in Mathematics, Formal Methods, Philosophy, Proof Assistants, Dependent Type Theory, Compilers, Boardgames, Programming Language Design, Education, Genre Fiction, Combinatorial Game Theory, Game Design and whatever I am up to since the last edit of this page.

This vast selection should not highlight expertise but rather a certain lack of focus when it comes to my interests. Transitively this lack of focus will most likely also appear in my work on this blog.