At school and university, I never really bothered taking notes. Sure—I scribbled things in a notebook as I was reading, and I defaced texts with highlights and marginalia, but I never took actual proper *notes*. I certainly never went back to any of my scribbles when it came to revising, or writing papers. Instead, I relied on the power of improvisation and working memory. It worked—for me. I got good grades and my teachers (usually) said nice things about my work. But it was such an inefficient way of working. And I regret not learning to take proper notes before leaving university. I regret it because I feel like a hypocrite. When I tell my students to take notes, part of me knows that I am telling them to do something I never found value in myself. For a long time, it felt dishonest to preach a practice that, deep down, I had written off as a waste of time. My perspective changed over the Christmas holidays (2025/26) when I got lost in a rabbit hole reading about reading, writing, and note-taking. I emerged with a new conviction: [[Nowhere to Hide|writing is thinking]]. As a result, I now argue that note-taking isn't just filler homework; it has the potential to be an engine of learning both now and in the future as a revision-ready resource. Don't bother taking notes—unless you do it well. ## Don't write while I'm talking You're bad at multitasking. I am too. Humans can only perform one task competently at a time. Listening to and processing the information you're hearing is hard enough without also worrying about the mechanics of writing notes. This relates directly to an educational concept teachers (should) think about a lot: **cognitive load**. This is the idea that your brain can only process so much new information at once, and processing new information is cognitively expensive. It's why watching TV while also texting your friend is a terrible idea: your attention is split so you fail at both. Writing notes while your teacher speaks splits your attention in the same way. There is an additional problem with writing while your teacher speaks: speed. Humans speak much faster than they write—whether by hand or on a keyboard. To write properly, you need to grapple with the idea you want to write and figure out the best way to express it on the page. This takes time. By the time you've figured out the phrasing for slide 1, your teacher has moved on to slide 10, and you've missed everything in between. You could forgo the rephrasing and copy down exactly what your teacher says, but that falls foul of another rule of effective note-taking: [[#Use your own words|use your own words]]. Listening and processing are one stage of building understanding; writing is another. They are mutually exclusive. Put the pen down until the talking stops. ## Use your own words [[Nowhere to Hide|Writing is thinking]]. If you don't engage in the craft of writing in your own words, no thinking will happen at all and you will have produced an artefact of no value to anyone. You already have a textbook; you don't need another. Take the time to grapple with the ideas you want to write—and reconsider the point of taking notes in the first place. Your notes are not primarily a record of what you know, they are a test of whether you know at all. If you're unable to write something down clearly and in your own words, you do not understand it. One of the main points of note-taking is to expose gaps like this in your understanding. When you get mid-sentence and pause because you don't know how to word what comes next, this is a sign that you need to go away and think harder. Use the act of writing to hold yourself accountable and be honest about your ignorance. To be absolutely clear: there is no value whatsoever in copying notes from a textbook, website, or friend, or asking an AI chatbot to take notes for you. All of these shortcuts are academically dishonest and, more importantly, cheat you of cognition. Don't waste your time with them. ## Ask questions Writing exposes gaps in your understanding. What should you do when you discover such a gap? Turn it into a question. When you're mid-sentence and struggling with what to write next, or your fingers hover over the keyboard in anticipation of inspiration, write as much as you can to reveal the gaps in your knowledge, then stop. Think hard about the information you need to finish the sentence. Do you know it? Great, write it down. Do you not? Even better. Write down the questions you need to answer and then answer them. Take this example of a student who wrote as much as they could and, when unsure, wrote an ellipsis rather than guessing: > [!example] Write as much as you can > The job of the CPU is to perform the fetch-decode-execute cycle. > In the fetch stage, the … fetches … from … using … They've succeeded at writing as they've exposed gaps in their understanding. They now need to turn their gaps into questions to be even more explicit about identifying things they don't know. This is an example of **metacognition**—thinking about your thinking—which is an extremely powerful tool for learning. Their questions might look like this: > [!example] Write questions to fill in the gaps > - What component of the CPU fetches things? > - What does it actually fetch? > - Where does it fetch stuff from? > - How does it fetch things? The problem now is answering the questions. They could return to the original source and read it again, or ask their teacher, or ask AI. This student looked up the answers in their textbook: > [!example] Answer your questions > - What component of the CPU fetches things? > - The control unit > - What does it actually fetch? > - Instructions and data (but in the FDE cycle it's getting instructions) > - Where does it fetch stuff from? > - Memory > - How does it fetch things? > - I realised I don't need to know this at GCSE Not every question you ask will have an answer that you need to know. It's important to be explicit about this as it saves you time and effort. The additional benefit to writing and answering questions like this is that it works towards our secondary goal of [[#Create a revision-ready resource|creating a revision resource]]. When they've written answers to their questions [[#Use your own words|in their own words]], they return to the original note and fill in the blanks: > [!example] Fill in the gaps > The job of the CPU is to perform the fetch-decode-execute cycle. > In the fetch stage, the control unit fetches instructions from memory. By thinking about your thinking and naming your ignorance, then doing the work to address it, you turn note-taking into an active learning activity with the potential to embed your learning for a good while to come. Slowing down and taking the time to write a question rather than writing a half-arsed paraphrase keeps you honest and ensures accuracy in your notes and understanding. ## Link to prior knowledge Knowledge is relational. When we learn new things, we learn them best when our brains create an explicit link to things we knew before. We call this a **schema**: a network of knowledge, both old and new. One of the most powerful sentences you can write in your notes is "this reminds me of…" as finishing this sentence forces you to find connections with prior knowledge, thus folding the new knowledge into your schema. Some schema-building sentences include: - This reminds me of… - This seems like *x* but is actually different because *y* - This is an example of… - This happens because… The connections you find don't have to be from the same topics in your exam specification. It's perfectly legitimate to find connections between disparate domains. Networking and data structures are an excellent example of disparate connections as they have graphs in common: networking uses graphs to draw network diagrams; data structures includes all the graph algorithms. We can grow our schema further by comparing graphs in Computer Science to graphs in Maths: "graphs in Computing sound like graphs in Maths, but are completely different because they are not plotted on an axis or used to represent data—instead they are made of edges and nodes and represent relationships." The wider you cast your net in the search for connections, the stronger your understanding of everything—old and new—becomes. ## Go beyond facts While it's true that at GCSE and A-level, much of what you need to know is *fact*, facts are the domain of textbooks. You already have a textbook; you don't need another. This is not to say you should ignore facts—facts are foundational to understanding. Rather, you must go beyond. Your notes are an opportunity for you to explore the richness of [[#Link to prior knowledge|relational knowledge]] and the real world. One of the most important [[#Ask questions|questions to ask]] when learning something is "Why?" Answer this question to find purpose for the facts and grow your schema. Let's take the example of this student who is writing notes about embedded systems. They start by writing the facts: > [!example] Start with the facts > An embedded system is a computer system with a dedicated function in a larger mechanical or electrical system. Which are true, but uninteresting—a necessary starting point with little stickiness to embed itself in their schema. They then explore *why*: > [!example] Explore why > We need embedded systems because it would be absurdly inefficient (and impractical) to put a full-on desktop PC in a microwave just to count 3 minutes and turn on a heating element. This sentence takes the factual foundation and applies it to the real world. It's much stickier and it implies the definition from before. A few more *why* sentences like this, and this student will have some excellent notes on embedded systems. Other ways in which you can go beyond the factual include using diagrams and analogies. Diagrams are excellent at representing processes, and can be annotated to explain *why*. Analogies are useful for building your schema—but should be handled with care. Analogies are only useful when they work; a forced analogy is like a dentist giving you a root canal on a cruise ship while dolphins eat your mum's sandwich—ultimately unhelpful. A good analogy requires no explanation. ## Create a revision-ready resource If your notes are not an actionable revision resource, you're leaving half their value on the table. The job of turning notes into a revision resource is easier than ever with AI chatbots, but there are a couple of things you can do to make the conversion process easier. We have already discussed the most important one at some length: [[#Ask questions|ask questions]]. By including the questions you asked while writing your notes, you have exactly the information you—or an LLM—need to make flash cards in the future. Organising your notes with headings can be useful too. Clearly labelling things like "Definition", "Why", "Links to prior knowledge" sets you up for AI success and makes your notes more readable for you in the future. You should, however, be wary of spending too much time on formatting if it comes at the expense of thinking. I like to write first and refactor later, which gets my thinking out of the way and gives me a second pass over my writing to check for relevance and accuracy. That said, I don't want to be too prescriptive about how you should structure your notes, as I think it's mostly personal choice (and your teachers may have their own requirements). If you follow all my recommendations above, you should be able to copy/paste your notes into an AI chatbot and have it produce some useful revision material. The difference between using AI at this stage compared to earlier is that by now you've taken ownership of the cognition. Creating a revision resource is just a matter of reformatting, which LLMs excel at. Consider these prompts for inspiration: - Take the explicit questions and answers I wrote in my notes and format them into a table for Quizlet, keeping my exact wording. - Hide the answers to my questions. Ask me them one by one, and grade my responses based on my notes. - Look at these notes. I am going to have a go at explaining \[Topic x\] and I want you to prompt me with questions when I make a mistake. Don't correct me, but scaffold me towards a correct answer. Remember that, even with the best prompts, AI can still hallucinate or prod you towards knowledge beyond what you need. To help mitigate this, you can upload a copy of your exam specification and explicitly tell it the course you're studying. Don't trust AI blindly; remember that LLM responses usually skew towards university-level rather than GCSE or A-level. ## Summary The point of note-taking is to facilitate thinking in the here and now and set future you up with a good revision resource. Failing to do either of these things leaves half the value of note-taking on the table. In summary: - Note-taking is cognitively expensive, so don't multitask. - Take ownership of your thinking by writing in your own words. - Interrogate your ignorance by explicitly asking questions. - Build schemas by linking new to old knowledge. - Build better schemas by asking *why*. - Don't let the written word die on the page. Use it for revision. ## Recommended reading - [[How to Read a Book]] - [[A System for Writing]] - [[How to Take Smart Notes]]