I'm a bit confused. What exactly defines a "higher-level" property, if not that it can be reduced to lower-level properties?

No it isn't? I did not mean you would be able to make predictions which came true 100% of the time. I meant that your subjective anticipation of possible outcomes would be equal to the probability of those outcomes, maximizing both precision and accuracy.

I disagree with your entire premise. I think we should pin down this concept of "levels of perspective" with some good jargon at some point, but regardless...

You can look at a computer from the level of perspective of "there are windows on the screen and I can move the mouse around. I can manipulate files on the hard drive with the mouse and the keyboard, and those changes will be reflected inside information boxes in the windows." This is the perspective most people see a computer from, but it is not a complete description of a computer (i.e. if someone unfamiliar with the concept of computers heard this description, they could not build a computer from base materials.)

You might also see the perspective, "There are many tiny dots of light on a flat surface, lit up in various patterns. Those patterns are caused by electricity moving in certain ways through silica wires arranged in certain ways." This is, I think, one level lower, but an unfamiliar person could not build a computer from scratch from this description.

Another level down, the description might be: "There is a CPU, which is composed of hundreds of thousands of transistors, arranged into logic gates such that when electricity is sent through them you can perform meaningful calculations. These calculations are written in files using a specific instruction set ("assembly language"). The files are stored on a disk in binary, with the disk containing many cesium atoms arranged in a certain order, which have either an extra electron or do not, representing 1 and 0 respectively. When the CPU needs to temporarily store a value useful in its calculations, it does so in the RAM, which is like the disk except much faster and smaller. Some of the calculations are used to turn certain square-shaped lights on a large flat surface blink in certain ways, which provides arbitrary information to the user". We are getting to the point where an unfamiliar human might be able to recreate a computer from scratch, and therefore can be said to actually "understand" the system.

But still yet there are lower levels. Describing the actual logic gate organization in the CPU, the system used by RAM to store variables, how the magnetic needle accesses a specific bit on the hard drive by spinning it... All of these things must be known and understood in order to rebuild a computer from scratch.

Humans designed the computer at the level of "logic gates", "bits on a hard drive", "registries", etc, and so it is not necessary to go deeper than this to understand the entire system (just as you don't have to go deeper than "gears and cogs" to understand how a clock works, or how you don't have to go deeper than "classical physics (billiards balls bouncing into each other)" to understand how a brain works.

But I hope that it's clear that the mechanisms at the lower levels of a system completely contain within them the behavior of the higher levels of the system. There are no new behaviors which you can only learn about by studying the system from a higher level of perspective; those complicated upper-level behaviors are entirely formed by the simple lower-level mechanisms, all the way down to the wave function describing the entire universe.

That is what reductionism means. If you know the state of the entire wavefunction describing the universe, you know everything there is to know about the universe. You could use it to predict that, in some everette branches, the assassination of Franz Ferdinand on the third planet from the star Sol in the milky way galaxy would cause a large war on that planet. You could use it to predict the exact moment at which any particular "slice" of the wavefunction (representing a particular possible universe) will enter its maximum entropy state. You could use it to predict any possible behavior of anything and you will never be surprised. That is what it means to say that all of reality reduces down to the base-level physics. That is what it means to posit reductionism; that from an information theoretical standpoint, you can make entirely accurate predictions about a system with only knowledge about its most basic level of perspective.

If you can demonstrate to me that there is some organizational structure of matter which causes that matter to behave differently from what would be predicted by just looking at the matter in question without considering its organization (which would require, by the way, all of reality to keep track not only of mass and of velocity but also of its organizational structure relative to nearby reality), then I will accept such a demonstration as being a complete and utter refutation of reductionism. But there is no such behavior.

I suspect that a lot of people also come to racism[1] without doing any research at all

That depends on what you mean by "any research at all". I suspect most people who come to racism do so via the logic I mentioned in this comment.

Comparing outcomes of existing systems would be good, assuming that you have multiple systems used by the same population. Some countries have this data, other countries don't. For example, if majority of schools in a country follows a government blueprint, and only a few alternative schools are allowed to coexist, it is not obvious whether the differences between their results are caused by different education, or simply by a selection bias (alternative schools are chosen by parents who are more interested in their child's education). So if you already have data, definitely use it; but many countries don't.

it is likely to turn out that in order to have a good public education system, you need to spend money

A decent quality requires some financial treshold, but mere money does not guarantee quality. You can't have a great school if most teachers need to take a second job to pay their mortgage, or if the school cannot afford to buy any educational tools or any trivial extra expense which could solve problems. On the other hand, it is possible to burn huge amounts of money without getting any improvement. (For example you spend the money on thousands of education-related government employees, and expensive fasionable educational tools of dubious quality, and the schools get only a small part of the budget.) In my experience, when someone proposes giving more money to education, they usually have a very specific idea about how that money should be spent, and it usually requires mandatory buying of something they produce. (This part may be country-specific.)

I don't follow you. What preferences I include is my business, not yours. You don't get to pass judgement on what is rational, rationality is just "accounting." We simply consult the math and check if the number is maximized. At most you can pass judgement on what is moral, but that is a complicated story.

School systems may be mainly rubbish, and school is slavery for children, but just routinely bashing it ignores the fact that we're the most knowledgeable generation there has ever been, and no-one knows how to do it better.

Yes. Sometimes in discussions about school system I am not sure whether the message is "schools are imperfect" or "schools are obviously worse than X" (and what is this X specifically). Because I fully agree that school system should be improved, and in my opinion we should try many experiments and measure the outcomes. (Also there would be some discussion about goals, like: do we want the best education ever, or just a decent education for a reasonable price? how much utility do we give to kids having knowledge versus kids feeling happy -- I agree that both are important, but what exactly is the desired exchange ratio?)

Only when people start suggesting their improvements, then most suggested improvements would actually make things worse, because they ignore some existing constraints, such as: human nature, limited budgets, lack of "perfect" teachers, limited time, etc. As the article says: "Most possible changes are for the worse, even though every improvement is necessarily a change." That's also true for changes of education. So we have a meta-problem: how to teach people to think rationally about learning?

Surely, that's not really the topic. For instance, I am not fooled by my perspective into thinking that I am literally at the center of the universe. Nor are most educated humans. My observations are compatible with me being at many locations in the universe - relative to any edges that future astronomers might conceivably discover. I don't see much of an illusion there.

It's true that early humans often believed that the earth was at the center of the universe. However, that seems a bit different.

Notice that a first person perspective doesn't necessarily have much to do with adaptations or evolution. If you build a robot, it too is at the centre of its world - simply because that's where its sensors and actuators are. This makes maximizing inclusive fitness seem like a bit of a side issue.

Calling what is essentially a product of locality an "illusion" still seems very odd to me. We really are at the centre of our perspectives on the world. That isn't an illusion, it's a true fact.

The issue isn't just about things beyond cosmological horizons. All distances are involved. I can help my neighbour more easily than I can help someone from half-way around the world. The distance involved entails expenses relating to sensory and motor signal propagation. For example, I can give my neighbour 10 bucks and be pretty sure that they will receive it.

Of course, there are also other, more important reasons why real agents don't respect the preferences of others. Egocentricity is caused more by evolution than by simple physics.

Lastly, I still don't think you can hope to use the term "rational" in this way. It sounds as though you're talking about some kind of supermorality to me. "Rationality" means something too different.