Ask HN: How to Think Clearly?
I tend to think all over the place when it comes to problem solving. It is hard to think step by step or have a clear mental path/logic. How can I overcome this? Thx.
Write down everything, the problem, all that you are thinking, approaches, everything.
I now use a system that's more like diary. Some notes are relevant for single day, others for a week. Anything older than that, I treat mostly as read-only. I'm ok with repeating myself over and over. It sets the context for thinking and decisions, and allows me to modify and correct assumptions over time. At the moment I use Trilium Notes for this, but still exploring other options. I would love something that includes diagrams and mind maps.
Unexpected benefit is that when I work while distracted and unmotivated, writing down each step and outcome allows me to stay on track and slowly make progress. Context switching becomes easier.
It's Write-Only notes - you can only see the previous line at any one time, and editing is intentionally difficult. You write, then at the end you see what you wrote. I've found it really useful (as someone who has the same tendencies you describe.)
(I first found it from a comment on HN.)
- Read well structured essays
- Write (the difficult part). Since writing is pretty difficult, here is what I do: I write aphorisms (Nietzsche/Debord/Wittgenstein style) to force me to say the main proposition without deviating from my point. This forces me to state clearly what I want to say. At this point, the aphorism is nothing more than an assertion that needs to be proved with arguments. But my main proposition is already here and I can keep it in my mind while I develop my thoughts towards it. It's like a lighthouse.
I had a quick google on this and loved what I saw. But I couldn’t find anything on how to do this in a life/business sense.
Do you have any further tips or references on how to get better at structuring aphorisms?
For instance, clothes came out if the washing machine dripping wet.
Step two is to get a better idea of what causes the problem. That's usually a list of facts and symptoms that can help the investigation.
For example, the washing machine still makes noise, and still pumps water out. However it doesn't shake as much as it used to.
Then and only then can you start coming up with theories on what the problem could be and how to fix it.
Maybe it doesn't shake because it doesn't spin, but the electric motor still turns, so it must be disconnected from the drum. The strap must have broken.
For big projects, I have to write an outline. First big steps, then divide those into smaller and smaller ones. Again, always in writing. And somehow, paper and pencil works much better for me than electronic tools.
If you are talking about problems like physics problems, I highly recommend Reif's book: https://mitpress.mit.edu/books/applying-cognitive-science-ed... and his papers on the subject (they are under "Physics Education Research"). Even if you are talking about other problems, reading these might provide some clues.
- Problem Solving 101 - Ken Watanabe
- The Pyramid Principle - Barbara Minto
Mental Models:
- https://fs.blog/mental-models/
- The Art of Thinking Clearly - Rolf Dobelli
- Books by Nassim Taleb
Meditate:
- waking up app - Sam Harris
Here are a few portions of his method that I always come back to:
---
Understanding the problem:
- What is the unknown?
- Draw a figure. Introduce suitable notation.
- Separate the various parts. Can you write them down?
Devising a plan:
- Have you seen it before? Do you know a related problem?
---
You can see the whole method at this link: https://imgv2-1-f.scribdassets.com/img/document/68994766/ori...
I use this approach in conjunction with an A4 piece of paper, or a note program capable of "outlining" functionality (expandable nodes). The latter works great for containing and structuring lots of information, and you can selectively show and hide different parts depending on what you're focusing on. Again, helping with filtering out the "noise" factor.
Another good approach is falling back to a plain-text editor such as Notepad.exe when you want to write anything down. I got that tip from the Pragmatic Programmer book. You can easily extend this to Markdown if needed.
2. Work in short time increments (~20 min). Use a timer. Take breaks.
3. Spend time up front defining the problem you're working on. Iterating on this will help clarify your thinking along the way.
4. Spend time writing out explanations of what you don't understand. Iterate on this until you feel you understand (i.e., the Feynman technique).
5. Keep a log of some sort. Key thoughts, actions you took, observations you made, lessons you learned. Keep it brief, but keep it. When you're feeling unfocused/lost, review your log.
A physical exam should catch many of the physiological causes, so I encourage OP to explore the possibility of psychological causes.
So if you have any of those AND you have ADHD then treating ADHD will almost always treat those as well. Since ADHD is the cause for those symptoms.
Depression/anxiety are often used as misdiagnosis before someone is diagnosed properly for ADHD.
And if you believe that the disinformation is only coming from one side, you're believing lies and disinformation...
Here's my orthogonal view:
What you're low in is Personality Trait Conscientiousness.
Here's how you raise that:
https://www.health.harvard.edu/staying-healthy/raising-your-...
tl;dr - Heighten your sensitivity to disgust as applies to yourself and the environment you control (i.e. your hygiene, your room, your clothes). Get into a schedule that YOU want. Do not punish yourself. Take care of yourself as if you were someone you loved.
Your mind is your body is your environment. They all interact. Start cleaning one to start cleaning the others.
1. Problem definition
try to understand the problem (the root might be hidden behind a few whys https://en.wikipedia.org/wiki/Five_whys )
2. Information Gathering
Gather enough more or less random info surrounding the problem. I do this until I feel like there is nothing major hidden from me anymore and I have a good grasp of the topic. This is interviews with people who know more and some googling or finding a book on it. Take notes, this is the first filter for what is important.
3. Structuring the Problem Space
If you can, try to structure the random info of your notes hierarchically right away. If you can't do that right away, dump all info in to a text file, read through it and than try to come up with a structure. I often have a "quick and dirty" inbox I only later refile in to my hierarchy.
4. Identify Knowledge Gaps, fill or postpone
With a bit of structure you might be able to identify gaps in your knowledge. If they are isolated enough not to change the general outcome of my decision making I just make a note that here is a clarification-todo to handle when making a concrete plan. If they are potential deal-brakers or uncover a bigger topic that needs attention I go back to 2. with this topic.
5. Note Ideas, don't dive in
Often in parallel I already make small notes of possible solutions or solutions to sub-problems, but nothing refined yet. Just enough not too loose any sparking idea. It's important not to get lost in the details before having a clear picture of the situation. With that picture done in 1. to 4. I now try to find a solution.
6. Research Prior-Art
I like to sketch out a first solution and only than research the competition. This way I'm not destined to follow their footstep and might find a better solution. Anyways: don't reinvent the wheel, see what others did.
7. Solving big to small
Go from big to small. Can you summarize your solution in a hand full of "stages"? With that as a skeleton things become much easier and you don't get lost in details. I do the same for sub-stages.
8. Full Concept
You have a solution sketched out, it looks good. Now go in deep. I try to go through every step of my solution mentally. Like imagining getting dressed to check if you packed everything in your bag for holidays. There will be more questions and clarification-todos popping up. You can solve them now or solve just those you need for the first steps of implementing a solution. Maybe write examples, mockups etc. here. Typically this 3-folds my notes.
9. Intermediate Goals and Stepping Stones
You have a full-fledged solution specified now. See if you can cut corners, do a proof-of-concept, an MVP or Milestones. If possible try to find a first milestone that already works start to end, just misses everything non-essential in between.
10. Build, learn, adjust, iterate
Do your first version, adapt your concept with what you learned, do a 2nd version and so on.
My approach is a bit watterfall-style and only at 10. becomes agile (unless it involves experiments to fill knowledge gaps). You might want to jump from 3. (structuring the knowledge space) to experiments and RATs quicker depending on the project. I try to limit the scope of my problem in the first place and save big ideas for later so the downsides of waterfall don't manifest that much. Just keep in mind the longer your concept the more of it might become obsolete mid way.