Monatomic Nobel Notes
How many basic elements does it take to create the world we live in? How many truly fundamental ideas have we invented? How many do you think you have? How many would be left after compression?
Simple ideas can often be the most profound. Smaller organisms have significant survival advantages over their more complex counterparts. Serverless microservice computing infrastructures provide better reactivity and adaptability than traditional monolithic systems, although this may involve different overheads. The pivotal Unix philosophy1 states that tools should be narrowly scoped and excel at single tasks, allowing them to be chained into information processing pipelines of unlimited sophistication. Productivity frameworks typically require breaking broad and complex goals into manageable, actionable steps. Short sentences stand out. Small things can indeed be lovely.
…
Schools teach us lasting mental images of molecular structures and atomic orbits. When we revisit these diagrams mentally and zoom out, we observe how they combine into complex structures through composition and rearrangement. These complex structures serve as components of larger systems. Composition is a powerful and ubiquitous phenomenon that enables almost everything to behave like mesmerising fractals.
For example, Western music consists of seven notes arranged into chords, chord progressions, and lengthy pieces with a certain degree of controlled repetition. Brad Frost’s “Atomic Design”2 discusses crafting user interfaces using blocks of the same type, such as buttons, checkboxes, and text fields. These blocks compose widgets and pages that, in turn, form websites. Component-driven design applies this same paradigm to software and hardware engineering on a larger scale. Letters of the alphabet create syllables, words, sentences, paragraphs, and more.
Atomic composition is a powerful, indisputable fact because our universe, including ourselves, resembles a giant Lego structure made of only 118 blocks. 99.1% of that structure consists of two such blocks: hydrogen and helium. This number is not large, even by the most minimalistic standards. It’s sobering to realise there are probably no more core novel ideas than entries in the periodic table of elements. There is likely just a micro-subset of seminal postulates accounting for the overwhelming portion of all concepts. They might be given different names by various people at other times to make old concepts seem new again or simply because they were unaware of their existence. In retrospect, we haven’t invented much.
Zooming back in, we see only a handful of monatomic elements, which can exist in a stable form while consisting of just one atom: helium (He), neon (Ne), argon (Ar), krypton (Kr), xenon (Xe), radon (Rn), and oganesson (Og). Their minimalistic nature has earned them the coolest nicknames: noble gases.
Don’t Bore Us, Get to the Chorus
Never read a book that can be adequately summarised.
—NASSIM NICHOLAS TALEB
To a less restrictive degree, this is how I feel about fundamental ideas. Just like noble gases, they aren’t numerous—perhaps even fewer than I like to believe. If I were to come up with only one genuinely original idea, I’d consider my mission on Earth accomplished. If we could distil every core idea to its most indivisible atomic form, we could refer to it as if they were elements of the periodic table in chemistry or substitutes in replacement tables used in compression algorithms.
This last analogy is particularly relevant at the time of writing, as Large Language Models find their way into every aspect of our lives. Now that the technology has reached its lifecycle phase, where the projections of what it is and what kind of adjacent possibilities it can unlock are more apparent, many have started noticing that AI might be the tool that could finally make knowledge engineers’ dreams come true: to ingest all knowledge, eliminate unnecessary noise, and provide us with executive summaries in their purest form. Just as bloated emails are summarised in a handful of bullet points, long-form information would be dissected into atomic ideas. From then on, creating valuable outputs would resemble how Niklas Luhmann sequenced his index cards into books3. In an infallible Unix philosophy way, valuable solutions could be constructed by sequencing monatomic ideas into constructs of varied sophistication without repetition. Goodbye to one-idea books.
According to Sam Altman, the leader of the pioneering company OpenAI, compression could solve our knowledge management inefficiencies. Many of the Extract, Transform, and Load (ETL) pipelines I build professionally merely lift and shift redundant data from point A to point B, reshaping and cleaning it to create a tidy, gold-standard data lake. However, this compression level is nowhere near what we could achieve with tools that drastically reduce information overload at scale.
I vividly remember one of my first writing assignments in school. Although it felt pretty open-ended, I couldn’t write more than a paragraph on the topic. In our household, my mother was the person who read the most and had the best command of our mother tongue, so I asked her for help. She read my draft and burst into laughter. Then, she offered me advice that might be the source of the “technical debt” we must repay. She said my problem was that I went straight to the point. “Add more water,” she suggested. Several decades later, we’re spending unorthodox energy dehydrating the Internet.
PKM is not about Marie Kondo-ing4 stuff you’ve hoarded; it’s about clarifying your thinking. Therefore, even if the knowledge-compression tools mentioned above become everyday household items, I will still take the time to crystallise my thoughts into atomic standalone units. The idea-distilling process cannot be outsourced to achieve the effect of a complementary cognitive artefact.
In the episode “The Man Behind ChatGPT” on the “How I Write” podcast, Sam Altman admits to experiencing a “What have we done?” moment while observing a student generating a homework essay from a prompt. It’s certainly what I would have done instead of calling my mom for help had I had such a cheat code back then. Yet Sam and his Y Combinator fellow Paul Graham agree5 that forecasting the death of writing as we know it because of the rise of AI is pretty shortsighted. Although it will undoubtedly lead to more pronounced polarisation between writers and write-nots, people will continue using writing as a tool for thought, just as we still go to the gym for reasons other than a lack of machinery to help us lift heavy stuff.
I write daily because I must. It helps me understand the topics I write about, and most importantly, it allows me to turn chaotic, fleeting ideas into coherent and articulate thoughts.
As I edit my scribbles, I realise how much noise I introduce because the first draft resembles how I talk. I read many writers’ advice books that sing in unison, stating that one should write like one speaks to make a book sound relatable. However, in doing so, I introduce useless junk. I don’t want to write how I speak. I want to talk like I edit. There’s only one way to get there that I know of.
Sometimes Small is Too Small
Everything should be made as simple as possible, but not simpler.
—ALBERT EINSTEIN
Removing unnecessary stone chips may sound appealing, but it is not easy.
Some ideas can be tightly bundled into a note’s yarn ball, requiring complete rewriting or reconsideration. Distillation reduces entropy and, therefore, requires energy, sometimes a significant amount.
Sometimes, specific thoughts may seem atomic today. However, with new information, you might notice a composition to refactor tomorrow, almost as if atoms, previously considered fundamental by less-informed scientists, now reveal their makeup of subatomic particles.
Refactoring is a term used in software engineering to describe improvements to a functioning program's source code. Developers refer to various types of enhancements as refactoring. In this context, we are particularly interested in the kind that helps us adhere to the Don’t Repeat Yourself (DRY) principle.
When the same logic can be reused throughout different parts of the program, creating a self-contained function, class, or file that includes only this logic and nothing else, this can then be reused as needed by anyone requiring it, aligning with the Unix philosophy mentioned earlier. Such functions make writing, reading, maintaining, and sharing code easier. An experienced developer who recognises patterns better than a beginner often writes their code in an atomic manner, even if a separate standalone function is used only once during the initial coding. The more you code and refactor, the better you become at writing pre-refactored code. This process is more of a craft than a science, but it can yield substantial benefits when executed effectively by a professional.
I noticed a similar phenomenon with note-taking over time. Repetitions, inefficiencies, and new ideas are naturally introduced throughout the entire body of knowledge. Therefore, pruning your knowledge is highly beneficial.
Although refactoring is an excellent way to organise your thoughts, it can become a bottomless pit, much like programming. You should set boundaries if you don’t want to spend your life refactoring. The precise location of this boundary is too context-dependent for me to advise you. Instead, I suggest progressing from coarse to fine in small steps. Unless you are an extreme “architect” note-taker, spending time creating one or two-sentence notes and constantly referring to them is counterproductive. It’s like a compressed file that has become larger than the original.
Whether you speak an alphabet-based or glyph-based language, you can divide complex constructs into smaller parts: words and components. They convey meanings when used alone. You can take it further by splitting words into letters or components into individual strokes, but these do not convey semantic meaning. Should you refactor this much? What is the necessary compression rate? Only you can draw that line. If you are a calligrapher, your atomic level is the stroke. If you are a language learner, you should not go beyond the words you can meaningfully use in conversation. Anything beyond that may lead to degradation.
What kinds of books will we write and read when consuming information in its most compressed form becomes the de facto standard? Will they be 50-page-long, straight-to-the-point books? Or do we still want stories? Are we part of the team in and out of a fast-food joint, or do we seek the grandeur of King James that atomic notes can never provide?
Nassim Nicholas Taleb uses stories as a medium for simple, atomic ideas. He states, “Ideas come and go, but stories stay.”6 He wrote his book, The Black Swan, as a narrative instead of as brief interconnected notes, emphasising this because he understands that writing and reading are fractal experiences. One must constantly zoom in and out, much like analysing scientific diagrams from school.
A Few Suggestions
Like all simple ideas, atomic notes are not easy to implement. This doesn’t mean we shouldn’t try; it simply means we shouldn’t overthink or obsess over them. Refactoring something as non-verbose as software source code can go awry if overdone, let alone something as freeform as human language notes. However, here are a few characteristics to use as a yardstick for your notes to determine whether you’re on the right path.
Self-Sufficient
A note should be self-explanatory. Pull one from the deck; you should understand the topic without needing another note to complete the picture. Your notes can be bite-sized or verbose, depending on how thorough you want to be. Some claim that condensing notes makes this exercise valuable, while others say their future selves can’t decipher cryptic, short notes once taken. The more you trust that your brain will reconstruct the complete picture in the future, the more compressed you can be. I struggle to fit something profound on a typical index card, but it might work for you.
Fits Into a Single Pigeonhole
If you feel your note belongs in several folders or has many tags attached, it strongly indicates its scope has become bloated. It likely contains more than one core idea.
When you come across notes like these, split them into the appropriate number of atomic notes and connect them accordingly. Today’s digital note-taking tools simulate the refactoring functionalities of typical Integrated Development Environments (IDEs), further solidifying the analogy between the engineering approach and everyday life.
You can select a part of your note, ask the app to extract it into a separate file, name it accordingly, and link the parent and child notes with backlinks. This is useful, but as mentioned before, as you gain note-taking experience, the need for such manipulations should diminish. This is especially true for analogue note-takers, who will increase their upfront idea segregation efforts because future refactoring is resource-intensive.
Contains Pros and Cons
Consider the following atomic analogy:
Your atom should have the nucleus as the core idea, positively charged protons as arguments supporting it, and negatively charged electrons as arguments against it.
For this monatomic noble gas equivalent to remain stable, the number of protons should equal the number of electrons—an atomic yin and yang.
Darwin spent years proactively poking holes in his revolutionary atomic idea of evolutionism. He knew this idea could have seismic effects, leading to criticism. He considered every possible pushback and more. The rest is history. A robust and well-articulated idea on a single note should have as many supporting arguments as counterarguments.
At some point, you may notice that parts of one note contradict the claims of another note. This might indicate that both notes carry more “weight” than they should. It may also mean that you’ve used Cal Newport’s QEC approach but have lumped questions, evidence, or conclusions together, while each should reside in its note and link back to the main topic.
Refactor.
…
These are a few indicators to help you get started. Avoid doing too much, or you’ll lose motivation, and the value of your notes vault will decrease because you might accidentally split words into letters.
I extract notes from my PKM system for spaced repetition and reformulation daily. This helps me stay current with my evolving knowledge and avoid “knowledge rot,” similar to “code rot.” Knowledge rot refers to the effects of entropy on software source code (a collection of notes in this case) over time.
TL;DR
What follows is this essay’s “Too Long; Didn’t Read” summary, courtesy of AI.
“Extract core ideas from this text and compress them into straightforward, standalone concepts.”
…
Importance of Size: Smaller entities, whether organisms or ideas, often have significant advantages and can be more impactful.
Survival Advantages: Simplicity and smaller scales in nature and technology generally lead to better adaptability and efficiency.
Unix Philosophy: Tools should be narrowly focused and excel in specific tasks to form complex systems.
Atomic Design: Just as in music, letters, or molecular structures, simple building blocks can be combined to create more complex systems in design and engineering.
Composition and Complexity: The composition of simple elements can lead to intricate structures and systems, resembling fractals.
Limited Core Ideas: There are likely very few original ideas, akin to the limited number of elements on the periodic table.
Monatomic Elements: Noble gases represent minimalistic entities, serving as an analogy for essential ideas that are few in number.
Core Idea Distillation: Original ideas can be distilled to their most basic forms, similar to the periodic table in chemistry or elements in data compression.
AI and Knowledge Compression: With advancements in AI, we can more efficiently compress and distil large amounts of information into actionable insights.
Efficiency in Knowledge Management: Compression technologies can help eliminate information overload and improve data handling processes.
Writing Simplification: The ability to summarize and distil information is essential for clarity and communication.
These concepts encapsulate the essence of the original text, focusing on the themes of simplicity, composition, and the interplay between smaller components and larger systems.
https://en.wikipedia.org/wiki/Unix_philosophy
Frost, B. (2016). Atomic Design.
Cevolini, A. (2016). Forgetting machines: Knowledge management evolution in early modern Europe. Brill.
Kondō, M. (2014). The Life-changing Magic of Tidying Up: The Japanese Art of Decluttering and Organizing. Vermilion.
https://paulgraham.com/writes.html
Taleb, N. N. (2007). The Black Swan: The Impact of the Highly Improbable. Random House.