DYNOMIGHT INTERNET NEWSLETTER

The DYNOMIGHT INTERNET NEWSLETTER probes a wide array of subjects from scientific skepticism, the interaction between class and taste, the rise of technology, to existential and philosophical questions. It intertwines critical analysis, historical perspectives, and personal insights to explore contemporary societal norms, technological impacts, and the nuances of human behavior.

Science and Skepticism Class and Society Technology and Its Impacts Existential and Philosophical Inquiry Personal Development and Behavior Historical Perspectives Education and Teaching Healthcare and Well-being Writing and Communication Evolutionary Biology

The hottest Substack posts of DYNOMIGHT INTERNET NEWSLETTER

And their main takeaways

Links for July

281 implied HN points • 08 Jul 25
🔬 Science Biology Physics Technology Health Research
  1. Goats have unique eyeballs that rotate to keep their vision horizontal, allowing them to see better while munching or looking around. This is pretty cool and shows how animals adapt physically to their environment.
  2. There's a way to stimulate specific cone cells in our eyes to potentially see new colors beyond what we can normally perceive. It suggests our brains can interpret unusual signals in fascinating ways.
  3. Recent findings indicate that only a small percentage of light-sensitive cells in our eyes control melatonin production, which affects our sleep. It's important to focus on dim lighting at night, rather than just blocking blue light.

Do blue-blocking glasses improve sleep?

625 implied HN points • 03 Jul 25
🏥 Health & Wellness Sleep Nutrition Mental health Wellbeing Lifestyle
  1. Blue-blocking glasses may help increase melatonin levels, especially when you're exposed to bright light at night. This means they could help you feel sleepier.
  2. The evidence on whether blue-blocking glasses improve overall sleep quality is mixed. Some studies show a small improvement, while others show little or no effect.
  3. Instead of wearing blue-blocking glasses, it might be better to reduce blue light in your environment, like using dim, warm-colored lights at night.

Scribble-based forecasting and AI 2027

562 implied HN points • 30 Jun 25
🕹 Technology AI Forecasting Data Analysis Machine Learning Simulation
  1. Both math and intuition can be used for forecasting, but they serve different purposes. Sometimes, using intuition can be more practical when creating predictions about complex situations.
  2. Math-based forecasts are best when the rules of a situation are well understood and complex. For simpler scenarios, basic predictions may be just as effective.
  3. Creating simple visual predictions, like drawing lines, can help clarify your thoughts. It's a great exercise to explore different potential outcomes and express predictions clearly.

Thoughts on the AI 2027 discourse

828 implied HN points • 23 Jun 25
🕹 Technology AI Discourse Peer Review Critique Methodology
  1. The discourse around AI 2027 shows both excitement for its predictions and criticism regarding its methods. This mix of reactions indicates a deep interest and concern about the future of AI.
  2. Peer review in academic work has flaws and can often delay important findings. This can sometimes result in long and complicated processes that may not effectively ensure the accuracy of research.
  3. An open and collaborative approach to discussing and critiquing ideas, like what's happening with AI 2027, could lead to better outcomes. When people engage directly and constructively, it helps improve the ideas being presented.

The AI safety problem is wanting

531 implied HN points • 26 Jun 25
🕹 Technology AI safety Machine Learning Ethics Computing Innovation
  1. AI safety is a big concern, and the main challenge is to make AI systems want to be nice to us. If they don't want to, they won't care about what we want.
  2. Trying to impose restrictions on AI won't work because a smarter AI can always find a way around them. Instead, we need to align AI with our values so it chooses to act positively.
  3. If we can ensure that AI genuinely wants to do what's best for us, the rest of the alignment problems become easier to manage. It's all about making sure AI understands and respects our values.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

Moral puzzles: Man vs. machine

562 implied HN points • 19 Jun 25
🕹 Technology AI Ethics Moral Philosophy Human Values Machine Learning Future Technology
  1. Current AI can understand human values to some extent, but it may not cover all complex situations. It's crucial to keep testing AI's responses on moral questions.
  2. People's opinions on moral dilemmas can vary significantly, especially on more unusual scenarios. This highlights the complexity of human ethics.
  3. Readers recognized that their views might differ from the general population, showing self-awareness in moral reasoning. It's good to be mindful of how diverse perspectives can be.

Please take my weird moral puzzles quiz

468 implied HN points • 17 Jun 25
📖 Philosophy Ethics Morality Logic Existentialism Metaphysics
  1. A survey is created to gauge people's moral beliefs on unusual scenarios that challenge common ethics. It involves questions about difficult decisions we might face.
  2. The scenarios include serious topics like abortion and euthanasia, but also quirky ideas involving advanced AIs and future societies. This mix aims to provoke thought about morality.
  3. Participants are encouraged to respond honestly and intuitively, with the creator assuring no tricks are involved in the survey.

Futarchy's fundamental flaw

546 implied HN points • 12 Jun 25
🇺🇸 U.S. Politics Political theory Decision-making Market Analysis Economic Theory Statistical Methods
  1. Futarchy uses prediction markets to make decisions, but it has a basic flaw. It relies on conditional probabilities that don't show cause and effect, which means it can't actually provide reliable information for making choices.
  2. Just because high vitamin D levels are linked to lower death rates, it doesn't mean vitamin D causes those lower rates. Other factors like being healthier or wealthier often play a role. The same idea applies to Futarchy: just seeing a correlation isn’t enough to make decisions.
  3. While conditional prediction markets can give useful insights, they should be treated like any other statistic—helpful but not the only factor to consider. We need to recognize their limits and look for better methods to understand cause and effect.

Car trouble

1328 implied HN points • 06 Feb 25
😂 Humor Personal Philosophical
  1. Sometimes we spend too much time trying to figure out problems instead of actually looking at them closely. Instead of thinking too much, it might be better to observe what's happening directly.
  2. It's easy to get caught up in analyzing complex issues without taking action. Getting stuck in thought can prevent us from finding simple solutions.
  3. In life, not everything needs a deep understanding. Sometimes, just gathering more information and looking at the situation can lead to better outcomes.

The first RCT for GLP-1 drugs and alcoholism isn't what we hoped

1109 implied HN points • 20 Feb 25
🏥 Health Politics Addiction Medications Clinical Trials Public Health Research
  1. GLP-1 drugs like semaglutide help with diabetes and obesity, but new research shows they might not be effective for alcohol addiction as previously thought. The first clinical trial results were disappointing.
  2. The trial had only 48 participants and mixed results, with no clear evidence that semaglutide reduced alcohol consumption. Some data suggested it might help in a lab setting, but overall findings were weak.
  3. Despite the poor results, there are still many personal stories claiming these drugs help with addiction. More research is needed, as the search for effective treatments for addiction continues.

Do you need permission from the government to do independent research?

1015 implied HN points • 13 Feb 25
🇺🇸 U.S. Politics Government Regulation Research Ethics Law Public Policy Human Rights
  1. You don't always need government approval to do research on people. Many people believe IRB approval is required for all human research, but that's not true.
  2. The rules about what kind of research needs approval are complicated and can vary by state or institution. It's not always clear where the lines are drawn.
  3. Many journals require IRB approval to publish research, which can make it harder for independent researchers to share their findings without going through the process.

Algorithmic ranking is unfairly maligned

1156 implied HN points • 23 Jan 25
🕹 Technology Algorithms Social media Digital Culture User Experience Data Ethics
  1. Not all algorithmic ranking is bad. Some algorithms can be useful if they align with what you want to see and achieve.
  2. A lot of current algorithms are designed to keep you engaged and make money for the companies, not necessarily to help you find what you like.
  3. We need better control over these algorithms to ensure they serve our interests, possibly through new technology or structures that prevent companies from taking that control away.

I am offering mentoring

1062 implied HN points • 16 Jan 25
🚌 Education Mentoring Self-Improvement Academia Personal Development Statistics
  1. A mentoring program is being offered to help people achieve their goals. It's open to anyone, regardless of age or career stage.
  2. The mentor will have three 30-minute video sessions with chosen participants, focused on helping them with their specific challenges.
  3. This program is free and based on an application process to select three people that the mentor believes they can assist most effectively.

Underrated reasons to be thankful IV

1437 implied HN points • 28 Nov 24
📖 Philosophy Gratitude Well-being Science Human experience Mindfulness
  1. Meditation can help you feel really good, almost like drugs or other intense pleasures, and it’s cool that you can explore new ways to relax without getting addicted.
  2. Your brain can automate tasks as you practice, which makes it easier to do complex things without thinking too hard about every detail.
  3. Even though there are worries about pollution, some studies suggest that common plastics might not harm your health much, which is a relief.

Something weird is happening with LLMs and chess

1515 implied HN points • 14 Nov 24
🕹 Technology AI LLMs Chess Gaming Research
  1. Large language models (LLMs) can somewhat play chess, but they struggle after the opening moves. They were not specifically designed for chess, yet they can manage to play using their text training.
  2. The performance of different language models varies significantly when playing chess. Some models like 'gpt-3.5-turbo-instruct' excel at it, while others perform very poorly.
  3. It seems that focusing on instruction tuning can make LLMs worse at chess, suggesting that training style impacts their ability to play games effectively.

How minimum prices for ride-sharing can be counterproductive

937 implied HN points • 02 Jan 25
🇺🇸 U.S. Politics Economic Policy Labor Rights Government intervention
  1. When minimum prices are set for things like rides, it can actually make drivers wait around more without getting paid, which ends up lowering how often they drive.
  2. Even if the pay seems higher, drivers may not earn more overall because they spend less time actually working when demand drops due to higher prices.
  3. In the end, customers often pay more and get fewer rides, which doesn't help anyone in the long run; it's like paying drivers to do nothing while traffic increases.

Bayes is not a phase

453 implied HN points • 27 Feb 25
📖 Philosophy Epistemology Logic Decision Theory Rationality
  1. Bayesian reasoning is something we all use, even if we don't realize it. It's more about how we naturally think than some complex math.
  2. There are two types of uncertainty: aleatoric (random) and epistemic (based on knowledge). Mixing them helps us make better decisions.
  3. Arguing over which type of probability is 'real' is silly. It's better to recognize that life involves many messy decisions where formal reasoning can help, but is often complicated.

Things to argue about over the holidays instead of politics III

937 implied HN points • 19 Dec 24
🎭️ Culture Social Behavior Cultural Commentary
  1. It's fun to argue about small, everyday topics instead of serious issues like politics around the holidays. These lighthearted debates can bring laughter and connection.
  2. Questions about cultural differences, like how we enjoy music or food, can spark interesting conversations. Exploring these topics helps us understand each other better.
  3. Discussions about abstract ideas, such as consciousness or the nature of reality, can make for deep and thought-provoking talks. It's a great way to share perspectives and learn something new.

OK, I can partly explain the LLM chess weirdness now

796 implied HN points • 21 Nov 24
🕹 Technology AI LLMs Machine Learning Data science Chess
  1. LLMs like `gpt-3.5-turbo-instruct` can play chess well, but most other models struggle. Using specific prompts can improve their performance.
  2. Providing legal moves to LLMs can actually confuse them. Instead, repeating the game before making a move helps them make better decisions.
  3. Fine-tuning and giving examples both improve chess performance for LLMs, but combining them may not always yield the best results.

Links for January

515 implied HN points • 09 Jan 25
🎭️ Culture Politics Science Music Food Psychology
  1. Jimmy Carter once told a funny story about a rabbit that scared him while fishing, and it became a big deal in the news, even though he only shared it privately.
  2. There's interesting talk about how we classify mental disorders nowadays and if labels like 'ADHD' will still be relevant in the future.
  3. Recent studies show that drinking alcohol can lead to a higher risk of cancer, and it's great when reports present this information clearly and let people draw their own conclusions.

Attitudes one can take towards people who have behaved badly

875 implied HN points • 07 Nov 24
🎭️ Culture Social norms Ethics Human behavior Communication Philosophy
  1. Social punishment is a natural part of how humans interact, helping to enforce a social contract. It can feel too harsh or unjust at times, but it still serves a purpose.
  2. There's a tricky balance in discussing the ideas of people who have done bad things. We need to find ways to remember their contributions without overlooking their wrongdoings.
  3. It's important to correct the record if someone is proven innocent after being publicly condemned. Reputations can suffer permanently, and it’s crucial to acknowledge when we've gotten things wrong.

Sloth was not the right answer

625 implied HN points • 31 Oct 24
🇺🇸 U.S. Politics Elections Political Behavior Political Discourse Public Opinion Immigration
  1. People often let politics affect their emotions too much, which can lead to unhappiness and frustration. It's better to focus on deeper and more meaningful conversations.
  2. Many view slanted information in politics as normal, but this approach can prevent good ideas from spreading effectively. It's important to look for objective facts instead of just picking sides.
  3. The upcoming US presidential election will be influenced by factors such as immigration, inflation, and candidate popularity. Understanding these can help make sense of the election dynamics.

Tactical parenting

437 implied HN points • 05 Dec 24
  1. Tactical parenting involves using practical strategies to deal with common parenting challenges. It's about finding effective solutions that work for your family's needs.
  2. It's important to stay flexible and adapt your parenting techniques as your child grows. What works at one age may not work for them later.
  3. Communication with your child is key. Keeping an open dialogue can help build trust and understanding between you and your child.

Grug on diet soda and autism

1234 implied HN points • 05 Oct 23
🏥 Health & Wellness Nutrition Research Science News Journalism
  1. Grug organizes ideas about diet soda and autism from news sources to consolidate important information.
  2. Correlation is easy to see but doesn't prove causation, and it's crucial to consider carefully in scientific studies.
  3. The news often emphasizes alarming claims based on research findings, even when the researchers themselves are more cautious.

Shorts for January

671 implied HN points • 04 Jan 24
💾 History Military Ancient Greece Inventions Archaeology Medieval
  1. In Finland, support for joining NATO increased over the years, leading to eventual membership in 2023.
  2. Winston Churchill's father expressed disappointment and criticism towards him, revealing a strained relationship.
  3. The student protests in Paris in 1968 reflected a rejection of societal norms and an influence of Situationism.

Historical analogies for large language models

890 implied HN points • 08 Dec 22
  1. Large language models can start off underwhelming, but eventually surpass human performance in writing.
  2. The relationship between large language models and human writers can be one of competition and collaboration.
  3. The impact of large language models on writing and other fields may lead to shifts in job roles and consumption patterns.

The midwit home

416 HN points • 12 Oct 23
🕹 Technology Sensors Automation
  1. Smart homes can be complicated and require a lot of technical knowledge to set up and maintain.
  2. Simple, midwit solutions like remote-controlled outlets, motion-sensing bulbs, and mechanical outlet timers can make daily tasks easier and more convenient.
  3. Using basic power control for devices can offer a surprisingly effective and user-friendly home automation experience.

What if you just bought everyone air purifiers?

296 implied HN points • 25 Jan 24
🔬 Science Environmental health Philanthropy Public Health Cost Analysis
  1. Buying air purifiers for everyone could extend life expectancy, but at a high cost compared to other health interventions.
  2. The cost of buying air purifiers for random Americans to extend life expectancy is $32,899 per DALY, which is not as efficient as other healthcare interventions.
  3. Air purifiers may be a good health investment for individuals, but philanthropically, they are not as cost-effective as other interventions like malaria drugs.

Your friend the language model

437 implied HN points • 03 Mar 23
🕹 Technology AI Language Models Training Data Fine-tuning
  1. Large language models are trained using advanced techniques, powerful hardware, and huge datasets.
  2. These models can generate text by predicting likely words and are trained on internet data, books, and Wikipedia.
  3. Language models can be specialized through fine-tuning and prompt engineering for specific tasks like answering questions or generating code.

Valid arguments with invalid conclusions

515 implied HN points • 01 Dec 22
  1. Mispronouncing a word can change the whole meaning of a sentence.
  2. Technical jargon doesn't always belong in everyday conversations.
  3. Questioning assumptions in statistics can lead to deeper insights.

The anxiety of the moderate

578 implied HN points • 29 Sep 22
  1. Moderation allows for better understanding of different perspectives from the right and left.
  2. The world is complex and nuanced, requiring a thoughtful and incremental approach to issues.
  3. Moderation can be a strategic tool for influencing change and addressing non-polarized but important issues.

Underrated reasons to be thankful II

186 HN points • 24 Nov 22
  1. Cyanobacteria 2 billion years ago led to the rise of eukaryotes that benefit us today.
  2. Refrigeration reduced stomach cancer deaths by 93% without intentional effort.
  3. Different materials like concrete and steel can work together to create durable structures.

Creative nonfiction training exercises

234 implied HN points • 04 Apr 22
  1. Focus on describing rather than arguing
  2. Embrace naked honesty over cleverness or creativity
  3. Say 'obvious' things because your reactions are unique

Teaching is a slow process of becoming everything you hate

203 implied HN points • 11 Mar 22
  1. Assignments with precise instructions can stem from misinterpretations by a few students, leading to excessive clarifications by teachers.
  2. Grade boundaries create dilemmas for teachers, potentially leading to lack of flexibility in rounding grades.
  3. Policing cheating can be challenging for teachers due to diffuse negative impacts and may require using arcane methods to detect dishonesty.

You, your parents, and the hotness of who you marry

187 implied HN points • 21 Mar 22
  1. When looking for a partner to marry, people prioritize different qualities like intelligence, health, kindness, and attractiveness.
  2. There may be a difference in priorities between parents and children due to evolutionary reasons.
  3. The conflict between older and younger generations about mate selection could be due to small differences in incentives or wisdom that lead to significant disagreements.