What is this post about?

To form more accurate beliefs in the future, I kept track of beliefs I changed my mind about. I have already posted a similar article some time ago and won’t repeat its content here. If you are interested in a deeper motivation or the things I have previously changed my mind about just check out the other article.

This post not only includes opinions I explicitly changed but also things I have learned and want to share.

If you like the post and know other people who you expect to like it, please share it with them. If you didn’t like it or disagree please tell me why.


Entropy-based decision-making

In one of his lectures my PhD supervisor, Philipp Hennig, introduced the concept of entropy (information content) in the context of experimental design. You should, for example, choose participants in an experiment such that you maximize information gain rather than choosing them at random, e.g. if you are a political pollster it makes sense to specifically ask the group you currently know least about. I found this framework very insightful and try to extend this to everyday high- and low-level decision-making. In more simple terms, I try to ask questions such that the answers provide the most valuable information. In conversations, I think this framework makes it easier to force yourself to understand what the disagreement is than to e.g. fall back on your prior beliefs or fall victim to your own biases. I usually try to ask “What does the other person believe?”, “Where do we disagree?” and “Do I find their position more plausible than my current one?”.

This mindset is applicable in lots of places, e.g. when reading books or blog posts. Most books are not good enough that it is worth reading every page. Thus, if you want to maximize information gain, it makes sense to skim or skip parts that you expect to know already and very carefully read the parts that you don’t know. My blog posts are also usually designed such that people with this mindset can gain information as efficiently as possible, i.e. my captions are summaries of the section’s content and people can decide themselves whether they want to read or skip it.

I think it also changed the way I view news. Most everyday news give a feeling of urgency or importance. But in the grand scheme of things they are usually pretty irrelevant. Thus I have shifted away from consuming daily or weekly news to mostly consuming in-depth analysis on blogs, podcasts, or traditional media and then sometimes read summaries of current news.

Clearly, the metric I’m applying is not only information gain because it would imply learning a lot of irrelevant stuff, e.g. memorize a phone book. Thus what I’m actually maximizing is something like a relevance-weighted information gain, e.g. I might try to learn as much as possible w.r.t. a particular goal. Overall this entropy-based decision-making is something I found conceptually helpful but haven’t developed a numerical model for it.

High-level prioritization > low-level hussle

Previously, I thought that just working hard would be the best strategy to achieve a certain goal. Now I think it is more important to make good high-level decisions, e.g. what to work on and especially what NOT to work on, than how hard you try. Somebody could for example work all day to transport dirt by hand while someone else could use a wheelbarrow and be done much quicker. Using the wheelbarrow is clearly more efficient, but it might feel less productive because it didn’t feel complicated. To prevent striving for a perception of productivity that focuses on how I feel about the work, I try to identify how much closer it got me to my final goal. Spending five minutes to ask the right person often gets you much closer than spending hours on your own even though it feels less productive. In practice, while I have gotten better at it, I still say yes to too many projects that are only tangentially related to my overarching goals. After making a high-level prioritization there are still a lot of tasks that “just have to be done” even though they are boring. But by drastically prioritizing them beforehand, I realized that many tasks are either not necessary or have low value and thus can just be skipped.

My intuition on this issue was partly shaped by the decision-making of CEOs. Some of them are able to create businesses worth multiple hundreds of billions of dollars and they are obviously not working a million times as hard as the CEO of a smaller company. Thus they must apply better strategies to their decision-making.

To apply this strategy, I have tried to make it a habit to first ask what my high-level goal is, then look for the best way to achieve it, and finally to follow the most promising plan.

A framework that I found helpful for this prioritization is trying to focus 80% of your energy on the most important 20% of your projects and the other way around (see Pareto Principle). In the context of my academic career, this means, for example, to focus most of my energy on the project at hand and actively trying to keep myself from starting interesting sounding side projects when they don’t fit into the overall trajectory of my Ph.D.

Another habit I found useful is doing back-of-the-envelope calculations for everything that you can reasonably assign a number to. If you, for example, don’t know whether it is worth buying a new TV, just estimate how much you expect to use it and how much cost-per-hour this translates to. If you are uncertain, just create a best- and worst-case scenario. With the hourly values, I find it much easier to estimate whether the purchase is worth it or not.

I have written an extensive article on productivity if you are interested in more than just this short summary.

I like money more now

For some time I thought that you didn’t really need a lot of money and that following your passion is much more important than trying to get rich. While it is certainly true that you can lead a decent life without having a lot of money, I value having it much more by now. This is mostly because money can save you a lot of time and annoyances and thus increase your freedom. If your phone is old and semi-broken, money can buy you a new one. If you want to master a skill quickly, money can buy you a good teacher. If there is a task that takes up a lot of time, you can employ someone to do the task for you. Money might not buy you happiness but it can certainly remove inconveniences.

I’m probably still spending less money than the average person in Germany, but I have become much more willing to spend money to make my life easier. My past self, for example, would have never been willing to pay for Roam Research or Whoop. I have even recently decided to buy a MacBook because I was fed up with Linux problems. So far, I’m pretty happy about all of the three purchasing decisions.

Nature vs. Nurture vs. Skill vs. Luck

There is this never-ending debate about whether people’s success can be attributed to their own decisions, their genes, or their circumstances. I can’t offer any theoretical solutions regarding this question but I have clarified my own thoughts and found some practical answers for me.

First of all, I have updated towards higher importance of genetics. Partly this comes from understanding how specific the genetic setup of professional cyclists is (and other sports as well) and partly from interacting with very smart people who solved problems on the fly seemingly without even trying while having little specific expertise. Secondly, I have updated towards a much more stochastic world. A lot of great inventions were pretty random, many careers happen because of random conversations or other seemingly random details and life in general just seems much more random than I initially assumed. Thirdly, I now understand better what baffles me about the American Dream debate, i.e. whether it is possible to get into any position in life you want if you just try hard enough. On the one hand, we make decision all the time and thus it seems plausible that continuously making better than average decisions (e.g. to work an extra hour) would eventually accumulate to very large differences in outcomes compared to making average decisions. Most people, for example, don’t prepare as well as they could for an exam or pursue their goals with clearly inefficient strategies. Thus, “going the extra mile”, should accumulate to better results over time. So in this sense, I think it is theoretically possible to move towards better outcomes to a large extent (constraint by genes and luck). On the other hand, the statistics show that social mobility is very limited even when people try hard. This seems to implicate to me that either the circumstances disable certain people from making good decisions (e.g. by not being able to afford education) or people have a hard time to differentiate between good and bad decisions in some structural way and outcomes are therefore mostly determined by where you started.

From this follows three practical consequences for me:

  1. I try to have two modes, one for personal and one for societal questions. In the personal mode, I try to improve my own decision-making, treat success and failure as results of my actions, try to improve them, and not blame others. This does not imply that I take full credit for my success (I know that I’m privileged), it is much more about dealing with failure. In the societal mode, I try to assume that people are to a large extent not responsible for their life outcomes and try to design systems to account for this fact. In an election, for example, I usually vote for parties that try to improve systems rather than focus on individual responsibility, even if my personal mode is more in accordance with the message of the latter.
  2. Maximizing Luck surface area. As described before, I think that there is large stochasticity in life. Thus a practical consequence is to diversify your bets and maximize luck surface area rather than focusing all of your efforts on one particular scenario. For my career choice, this could, for example, look like having an explicit plan A, B, and C and as a more general rule it implies that I try not to tie my own self-worth too much to one particular goal or achievement because the probability of failure is larger than I intuitively assume. This belief is compatible with the prioritization argument from above. I try to focus most of my energy on one project but am willing to shift to plan B when it fails.
  3. Find out what your predispositions are good for. From a societal perspective, it is optimal when people specialize in what they are best at but also from a personal perspective it is more rewarding. Since I currently think that genetic and social predispositions have a large impact on your life outcomes, it makes sense to spend more time figuring out what you are good at and specializing in that. At this point, I’m not sure what that thing is for me. I seem to be decent at many things but not excel at any of them.

Follow the argument (and act on it)

In a trade-off between logic vs. intuition or logic vs. societal conventions, I have updated to choosing logic in the vast majority of cases. There were a lot of cases in the recent past where the logic turned out to be correct even when my intuition was originally not aligned. When the news of the Covid19 outbreak in China went public in January 2020 I understood the mathematical models of pandemics and yet I didn’t adapt my expectations or life accordingly. I probably should have bought some masks, shorted some stocks, and invested in Amazon. But I didn’t follow the argument to its logical conclusion. When I first heard of the topic of wild animal suffering my intuition was that it might be true but kinda weird and then mostly ignored it. Now, a couple of years down the line, I find wild animal suffering very important. But to get there, I really had to put my intuition and societal expectations away and just look at the arguments and numbers. I had a very similar experience with AI-safety which I brushed aside when I first heard about it but now think is the most important problem of humanity. By now, my intuition has proven to be wrong or at least laggy so often, that my default mode has become to first figure out why my intuition and the logic of the argument are not aligned. Then, if this process doesn’t yield any strong defense of my intuition, I try to override my intuition and act upon the logic.

On the flip side, defying social expectations or going against my intuition in favor of a logical explanation has turned out to be positive for me most of the time. Most people, for example, don’t invest their spare money in ETFs and I was thus intuitively uncertain about it. I did it anyway because it made sense to me and it turned out to be a good decision.

Choose your identity, don’t let your identity choose you

I have observed that people are often proud of objectively bad traits and often even built their identity around them. Some people are “the person who is always late”, “the person who drinks a lot of beer” or “the person who never shows mercy”. At some point, I realized that I also show this kind of behavior. In school, for example, I was “the guy who never prepared for an exam” or “the guy who always wears shorts even if it snows”. Both of these are clearly bad because my grades were worse than they could have been and my knees might suffer some long-term damage from my stupid behavior. Some of my examples can also be explained through positive traits, e.g. someone who drinks a lot might be invited to every party and thus have high social status. But I still think that all of them clearly have a negative component and better alternatives.

I’m not quite sure what the psychological explanation for this trend is but I see it all the time. My best guess would be some mixture of a) a psychological explanation, e.g. people like attention and standing out for something and the short-term fix of “being the person that does X” is more rewarding than the long-term harm of the behavior, and b) an evolutional explanation, e.g. that being proud of bad behavior signals that you are alive and ready to procreate despite your bad trait and thus more worthy of mating than all the plebs without this baggage.

Building your identity around a bad habit is clearly not optimal. Thus, I try to adopt two strategies.

  1. When I realize, that something is part of my identity or I’m proud of something about me, I try to evaluate whether this is a good trait. If, for example, I realize that I’m “the guy who defies a social expectation”, I think about whether I would want the people around me to have this trait or not.
  2. Choose your identity. It feels like identity building has been a rather passive process for most of my life, i.e. mostly determined by my surroundings, habits, or random events. Basically, my externalities rolled the dice, it lands on 3, and I then pretend that I’m a person that should be a 3 and it was written in the stars that I should have always been a 3. I think that the process should be active not passive, i.e. you think about which identity you want to have and then work towards it. If I realize, for example, that I really like a specific trait in a person, e.g. Julia Galef’s intellectually honesty, then I try to build my identity around that trait. It also helps to remember that there is no duty to keep true to your past self and there is no actual harm in leaving bad traits behind even if you have been proud of them at some point. This strategy obviously only applies to the parts of your identity you can control and not e.g. sexuality, ethnicity or gender.

Personal purity

It is clear that personal choices have consequences, being vegetarian or vegan prevents animal suffering and using less plastic or flying less is good for the environment. However, there is a large discrepancy between the real effect of different personal choices and how important people perceive them to be. For example, choosing not to eat meat, especially beef, is already the largest step to reduce your climate impact from food choices (see Our World in Data). Everything else, like cutting on dairy products or eating fewer eggs still decreases your carbon footprint but its effect is marginal compared to meat. Also, many people try to limit the use of plastic bags due to environmental reasons. The production of one kg of plastic produces around 6Kg of CO2e and creates around 180 plastic bags. A flight from Berlin to Mallorca and back produces 609 kg CO2e. Thus one holiday trip to Mallorca is worth around 18270 plastic bags in terms of climate impact. And yet in the social discourse (maybe this is just Tübingen though), many personal actions are treated as if they didn’t have huge discrepancies in real-world impact. Rather, they are treated in more binary terms, i.e. flying is evil but using a plastic bag is evil too. A violation of either principle is seen as nearly equally bad. The social norm around personal actions seems to be focused much more on purity than on defining the moral weight of an action proportional to its consequences.

I think this is a bad norm for many reasons. Just to list a few: a) people are usually not as pure as they pretend and thus sometimes violate their principles. If actions are not seen proportional to their impacts it is easier for people to e.g. justify a flight. Furthermore, humans tend to apply some moral discounting, e.g. if we rarely use plastic bags we might feel pure and can thus justify eating meat more easily. b) If people don’t have an accurate understanding of the action-to-impact relation they vote for ineffective policies, e.g. a ban on plastic straws. c) Every decision we ever make has an influence on another person or the environment. Treating them as if all of them were nearly equally important means you overemphasize the ineffective ones but underprioritize the important ones. d) If someone else commits a minor violation of the purity norm and you vilify them for it, you drive them away from your overarching goal of e.g. reducing carbon emissions.

My personal consequence from this is that I try to focus most of my energy on getting the important questions right and don’t think long about the small considerations. Concretely this means, I will try to choose a job within the Effective Altruism space because it’s probably the most impactful decision of my life. I try to fly only when necessary and don’t eat meat. Whenever it doesn’t create additional effort I focus on small stuff like going completely vegan or using less paper. Whenever it creates annoyance I just don’t do it, e.g. when there is no vegan diet in a restaurant I just eat vegetarian, and whenever I want to write something on paper rather than a computer I just use paper. When other people have a strong preference to eat meat from time to time, I don’t mind but if they consider taking unnecessary flights I try to convince them not to take it or to at least compensate their climate impact.

Look for the boring stuff

I have observed that the paths associated with high status are often not those that maximize money or social good. Often complicated things lead to high status while much more mundane or boring options are those where small changes could create better outcomes quite easily. This seems to be very clear in the context of Effective Altruism where one of the most effective interventions is to just give people money (GiveDirectly). It is so simple and obvious that you cannot signal your intelligence by suggesting it and thus many talented individuals do something much more complex. I also realized this, for example, when talking to medical students. Very often, the most talented people in medicine want to become neurosurgeons or neuroscientists because - you know - it’s complicated, impressive, and “looking for the true meaning of human intelligence”. However, Neuroscience is pretty hard, progress is rather slow and currently, most neurological diseases cannot be treated effectively. From my conversations with neuroscientists, it also doesn’t feel like big leaps can be expected in the near future. So if your goal was to either make more money or help more people in an efficient manner, it might make sense to search in the medical spaces that give less status but have high prevalence, e.g. Malaria, cardiovascular diseases, or chronic pain. Another area where I found a similar phenomenon is within academia. There seems to be a bias against writing overviews, benchmark papers, or tutorials because they are seen as boring and people would rather want to be associated with something new and shiny. However, the boring papers often get many more citations, help out newcomers or generate ideas for new research directions. Similarly, most academic papers are way more complicated than they need to be because writing in a simple style is associated with lower effort or intelligence. Thus I intend to write lots of boring and simple papers in my academic future because it is the maximum value option if you care less about short-term status.

Furthermore, I will try to explore paths that other people find boring when there is no logical explanation for its low status. If, for example, I were to found a start-up right now, it would probably involve applying my expertise, e.g. Machine Learning, to a conventionally boring area with a large reach such as utilities.

Sleep more

Previously I have tried to sleep less to have more time and now I try to sleep longer to have more productive time. For a long version of my findings see my blog post on sleep. The short version is that good sleep has a much larger effect on my mood and productivity than I thought and I have thus tried to improve it through various means including a better bed and very regular sleeping times. Additionally, I have bought a Whoop strap which tracks certain measurements as a proxy of sleep quality. An interesting finding was how many disruptions there are during an average sleep that I didn’t notice afterward. When I sleep from 11 pm to 7 am, for example, I am awake for over 30 minutes per night split into 10-15 different short intervals. So if my goal is to get eight hours of sleep I should aim for 8:30 instead.

Research is even harder than expected

When I started my Ph.D. I expected research to be hard but it turned out to be even harder. An analogy might capture this discrepancy best. The Bachelor’s program is like a highway - everything is already established, there are lots of resources online and one can generally move fast. The Master’s is more like a country road - content gets harder, you can still find stuff online but the speed is still decent. I expected the Ph.D. to be like riding a bike on a cobbled road but now it feels more like slashing through the jungle with a machete. I’m not sure whether this is due to the nature of my project, the situation with covid, that I’m just not suited for a Ph.D. or just a general feature of research but I found that other people tend to share my perspective.

On the other hand, I feel like research and academia could be much more efficient. Many PhDs (including me) work too much on their own when working in groups might be much more effective. Furthermore, aspects of academia are set up in inefficient ways. This might be especially true for Machine Learning but reviewers are often not experts in the subfield they are reviewing and thus the noise for reviews is much higher. Therefore researchers often have to do things that don’t make any sense to please their reviewers or have to reroll the dice until they get expert reviewers. This process costs a lot of time and it feels very suboptimal. This is not a complaint about my personal rejections. My paper has been rejected by experts for good reasons and their feedback genuinely improved it. But the overall structure of academia just feels like a big pile of coordination problems vaguely held together by underpaid researchers who sacrifice their free time to prevent the system from collapsing.

Social Interaction

I’m not sure the classical divide between introversion and extraversion fits well to describe me. A common heuristic to identify someone’s position on the spectrum is to check whether they seek or avoid social contact when they are in a low-energy state, e.g. whether you hang out with friends or prefer to relax alone after an exhausting day. The reason I feel like this is an inaccurate heuristic for me is that my preference really just depends on the kind of social interaction I could have. I find a party, for example, exhausting and boring if most conversations are short or contain little information even when my day was relaxed. On the other hand, I find listening to or participating in long and in-depth discussions refreshing and they improve my mood even after an exhausting day. I also feel like other markers of introversion or extroversion, such as whether you have problems approaching people, don’t fit well. I would say I have comparatively little problems approaching someone regardless of their social status. For example, I usually don’t have a feeling of discomfort when asking my professor a question even when the entire class is listening where many of my friends say that it makes them very uncomfortable. So this trait would categorize me more towards the extrovert end of the spectrum. But many other traits commonly associated with extroversion are mostly absent. I have no problem just listening to an interesting conversation without the urge to present my own views, I have no problem being alone for long amounts of time (as long as I have the internet) and I tend to have a less social view of friendship than most of my peers. All of these would put me towards the introvert end of the spectrum.

I think the last part is mostly due to a different perception of friendship than is common. For example, it happened multiple times that someone who I considered a close friend was not sure about whether I considered them a good friend at all. In those cases my perception was: “I enjoy the time we spend together, I feel like we have a good connection, I expect this relationship to hold for long and thus they fall into the category of close friend” but that didn’t mean that I wrote them everyday-things about my life or regularly followed up on theirs. Their perception was: “I also enjoy the time together and feel like we have a good connection BUT Marius doesn’t remember my birthday, rarely writes about his everyday life, and doesn’t seem to care a lot about mine. So maybe I’m not in his circle of close friends?”. I think that the difference in perception came from a different expectation about what a close friendship consists of. I valued the discussions and connection very highly but rarely signaled this through acts that would typically show such an affection (e.g. birthday messages). They, on the other hand, noticed the absence of these signals and concluded that I don’t consider them close friends. Realizing this discrepancy hit me hard in the beginning because it made me suddenly realize that some of my best friends might not know that I consider them good friends. Whenever I realized this, I tried to explicitly talk about my and their perception of the friendship and if either of us wanted to change it in some way or another. Talking about the different expectations very explicitly has improved the friendship in all cases and - even though it might feel weird in the beginning - is something I would generally recommend. Even if you realize that one person doesn’t want to be part of that friendship anymore, you have saved yourself the weird phase of a friendship slowly fading away. I guess this directness would put me towards extroversion but my atypical views on friendships towards introversion which just leaves me more confused.

If you know me and have any opinions on whether I’m an intro- or extrovert, I would be genuinely interested in your opinion.

Moving in with my Girlfriend was better than expected

I expected that moving in with my GF would be just fine - that there would be advantages, e.g. of being able to spend more time together, but also disadvantages, e.g. that we would annoy each other or start fighting over small stuff (~ rating of 4/5). It turned out that I overestimated the downsides and underestimated the upsides (~ rating of 4.9/5). For example, I had underestimated how much I value having someone to talk or cuddle whenever we feel like it. We also didn’t really get annoyed with each other even though we spent the entire pandemic in the same flat with minimal other social contacts. Then there is a lot of small stuff that I underestimated such as being able to split up household duties depending on who busy we are or which things we like more.


I used to think that Apple products were overpriced because people were willing to pay extra for a status symbol. Recently, I needed a new phone and laptop and no matter which comparison I looked at the Apple products were very price-efficient. I decided against an iPhone but in favor of a MacBook (with the M1 chip) and it was a great decision. Apple products seem to cause fewer problems than e.g. a Linux alternative and I can therefore spend more time working toward my goals rather than fixing problems. Therefore, I think Apple’s core products, i.e. iPhone and MacBook, are correctly priced rather than overpriced and they are underpriced if you value time highly. However, the additional gadgets such as adapters, more RAM or chargers are massively overpriced and I feel like being extorted whenever I buy them.


I really like Joe Biden

During the primaries, I already was more of a Joe Biden fan than most of my progressive friends. Our disagreement mostly came from how likely we estimated a more progressive candidate to win the overall election and how conservative Joe Biden would be in comparison to e.g. Sanders or Warren. But now after the election is over and we have a better idea of Biden’s politics, I have become an even bigger Biden fanboy. First of all, Biden didn’t get baited during the primaries to commit to incredibly unpopular policies to please very progressive members, e.g. decriminalizing border crossings or reparations for black people. I’m not saying these are necessarily bad policies, just that they are very unpopular and could have cost the candidate the overall win. Secondly, his cabinet seems to be very well-chosen. In most positions, it combines expertise with an overall progressive agenda and identity, e.g. more women than previous cabinets. Thirdly, he shows an actual effort to win over Republican lawmakers and work with them but also uses his majority in the Senate to pass important legislation if republicans decide not to cooperate. I find this very refreshing because I can’t stand the Democrats who either want to never cooperate with Republicans or only want to pass something when Republicans agree as well. Fourthly, I just like most of his policies. His administration is willing to tackle important problems even if they are boring and don’t earn him lot’s of airtime or attention. I guess it just feels good to have someone in power who is willing to fix infrastructure and sort out the pandemic without being distracted by the daily clown show of American politics. I don’t like some of his policies though. I think his stance on immigration is much too restrictive and I don’t like his isolationist stance on economics.

Neoliberalism & YIMBYism

I have written a separate blog post describing my economic positions. In short, a) I have become more market-friendly but think that the rewards should be distributed, e.g. fast technological growth is desirable but should be combined with steep progressive taxation. b) I’m a radical pragmatic, i.e. I believe in what works, independent of its ideological connotation. c) I would prioritize economic growth over social equality most of the time. My opinion was informed by the Our World in Data charts on economic growth and extreme poverty. But the base case is just: Economic growth is the best way to reduce poverty and poverty is worse than inequality, at least in longer time frames. A person with median quality of life today is much better off than someone in the top 10% a century ago on basically all metrics (e.g. life expectancy, freedom, personal finances, etc.).

Neoliberalism also has a stance on the housing debate and supports building more houses and deregulating zoning laws. The more I learned about housing policy the more I got convinced of the importance of available housing. When rents are lower people have more money which increases their agency and leads to more economic growth. On the other hand, the absence of housing has effects on social justice because poorer people, especially from minority backgrounds, are more likely to rent rather than own a home. Thus a strongly regulated housing market is effectively a redistribution from poor to rich. It also has effects on the environment. The lower the availability of housing the larger the need for commutes which creates CO2. My conviction of YIMBYism (Yes In My BackYard) has become so strong that I cannot unthink it anymore. Whenever I see an empty spot anywhere in a city my first thought is “Somebody should build a house right here”. I have accidentally become a city designer without the intention, expertise, or capital to build housing.

Video Games

Video games are an incredibly money-efficient way to generate happiness. They usually don’t cost more than 100€ but you can spend hundreds or thousands of hours playing them. Furthermore, they have the potential to increase your agency, as you can choose which character you want to be, e.g. which gender or ethnicity they have. You can explore impressive virtual worlds which are more accessible than e.g. visiting a different continent. They also remove a lot of the annoying parts of real life. You just respawn if you die, goals are clearly defined and the mechanics of a game are much simpler than those of the real world. So if you want to maximize your own happiness, I think video games are one of the best ways to do so.

However, since I’m an effective altruist and thus want to help others rather than maximizing my own happiness, I try to reduce my video game consumption as much as possible. I found that I’m very responsive to the advantages of video games and tend to get addicted rather quickly. Since this doesn’t allow for limited consumption I needed to stop playing those games entirely.

Nuclear Energy

I have always been a proponent of nuclear energy but my conviction has gotten stronger over time. After understanding how nuclear power plants are built in more detail, I’m pretty convinced that they are very safe, i.e. that the expected risk of modern systems is basically zero. Obviously, there is a chance of extreme weather events such as tsunamis and earthquakes or human error as in Tschernobyl but modern designs such as the molten salt reactors seem to be very close to secure against either. Thus I don’t think the biggest problem with nuclear energy is safety but waste disposal. So far we don’t have optimal solutions for storing used uranium rods nor can we recycle them. There are some promising future research directions such as recycling nuclear waste but none of them are close to solving the problem in the near future. So I’m not saying nuclear energy doesn’t have its problems, I just prefer it to its alternatives.

Sustainable energy does currently neither have the capacity nor stability to supply all of our energy demand. Thus some stable alternative is needed. Either we use nuclear energy or burn coal and gas - and I think nuclear wins all relevant comparisons pretty drastically. I know this sounds extreme to some, but I would even be willing to trade of an incident in a nuclear facility like in Fokushima every 20 years because the number of lives lost are so much lower than the number of deaths due to coal related respiratory illnesses.

Nuclear energy has basically no CO2 emissions while Coal and Gas have lots of them. This pollution leads to a drastic increase in respiratory diseases today and much more suffering due to climate change in the future. I find it hard to quantify how large the health consequences of nuclear waster are, but no estimate that I know is even close to the damage done by coal. Thus, I think states should use nuclear power plants until our network of sustainable energy is large and stable enough to take over entirely. To see how large the differences between energy sources are in terms of pollution consider this figure from Our world in Data

Private health insurance

I used to think that private insurance was bad because it is unfair. The most prominent example of unfair treatment is the large discrepancy in waiting time between publicly vs. privately insured patients. By now I am willing to trade off this unfairness for the benefits the private system provides. My understanding is that private insurance costs disproportionately more money than it yields privileges. This means that richer people effectively subsidize poorer patients by being covered for more exotic conditions or having a larger bed during a hospital stay. While the difference in waiting time is very visible and thus feels tangibly unfair, it doesn’t seem to have a large effect on treatment outcomes. If you have an urgent condition you will get treatment independent of your insurance. My mind was changed mostly by talking to doctors who reported that without their privately insured patients they would either have to increase their prices drastically or close down their practice.

Update: A follow-up discussion made strong arguments against my position and I concluded that my moral judgements merely depend on empirical data that I currently don’t have access to. I will try to speak with someone within the insurance industry to resolve my questions.

Recycling and Trash

Even though this is a pretty niche topic, my opinions have actually changed quite drastically on it. First, I was convinced that we should recycle our trash because it is more sustainable. Then Rob Wiblin changed my mind with this article where he argues that recycling is often very inefficient and landfills can be well run for relatively low costs. Then my mind was changed again by a long discussion with my father who has been working in the energy business and built waste incineration power stations for most of his career. He argued that the problem with landfills wasn’t toxic water (this can be managed) but rather spontaneous fires that are impossible to control. Since landfills contain a lot of easily inflammable materials and have high pressure in the bottom they can ignite easily. Once a fire has developed it is nearly impossible to extinguish because it usually develops from the bottom to the top. Once we see the symptoms, thousands of tons of trash are already burning. The smoke of such a fire is usually very toxic and can have strong health consequences for surrounding towns. Thus the best option seems to be to recycle whenever it is feasible and burn the remaining trash in a waste incineration power station where toxic gases are filtered and the resulting energy is put in the grid.

One last note

If you want to get informed about new posts you can subscribe to my mailing list or follow me on Twitter.

If you have any feedback regarding anything (i.e. layout or opinions) please tell me in a constructive manner via your preferred means of communication.