Luffel Industries
I love the early-internet vibes of Neocities, and so I'm channeling that manic and all-encompassing energy into a few questions that I hope to write about in the future. I'm not an expert in most of these topics, so there are probably more than a few errors. You'll also notice that my target reader is a young person. For example, a child of mine.
Questions
How should we organize data? What is a schema? What will computers be like in 50 years? Is AI going to take over? How are animated films made? What's wrong with contemporary programming languages? How many file formats are there? What is the most compact way to store data? Is open source a moral imperative? What's important about instruction sets? What can physical simulations tell us? Is high-frequency trading a waste of human talent and energy? How much energy do humans use for different tasks? What is knowledge? What is wisdom? Why should we be knowledgable? What is truth? How do we arrive at the truth? Why is truth hard? How can we avoid being wrong? What is bias? How do our brains work? What is memory? How do we learn? What kinds of knowledge are there in the brain? Are animal brains like human brains? What use are habits? How can we make good habits and break bad habits? How can we tell good from bad? What are our goals? What does history tell us about people's goals? What do babies need? How can we provide good environments for children? What does the body need? How can we stay well-fed? What is required for living with other people? How do societies work? How does industrial society work? What role do material flows and energy flows play in society? How is the power of decision-making distributed in a society? What are the ecological limits of our planet? How does biochemistry work? How do chemistry and physics work? How does electricity work? What are the limits of our understanding of physics? Will we ever meet aliens? What is unknowable or unprovable? How do computers work? How does mathematics work? Is math real? Will artificial intelligence take over the world?
What is a schema?
In computing lingo, a schema is a description of how different pieces of data are related. For example, a schema might say: "a person has a first name and a last name, and two parents who are themselves people". Schemas are most common in databases, where we want to represent some objects and relationships in the real world.
Schemas are uniform: they represent all similar things (like people) using the same data fields (like "name" or "parent"). The real world is never as uniform as a schema. The real world is messy and irregular. So a schema will be unable to represent some situations in the real world. Even so, schemas are very useful. They make it possible to write software (since a schema provides reliable assumptions about what can or cannot be stored). They also allow more compact and simpler storage of data in the database.
For more on how schemas interact with the complexity of the real world, see: Gay marriage: the database engineering perspective.
What can physical simulations tell us?
One of the most important tasks that we give to computers is the simulation of physical objects. For example, we simulate how a tall building would bend in the wind, or how wind will change the weather by moving heat and humidity across the sky. We also ask computers to simulate themselves. For example, before making a computer chip, we simulate how the electricity will flow through the wires and transistors.
In each of these examples, we ask the computer to follow some physical rules that we have learned from scientific experiments and from theory-making. Our simulations are only useful if they are realistic, and they will only be realistic if they contain true physical rules and quantities. We also need our simulations to finish quickly. That way, we have enough time to take the knowledge we gain and use it in the real world.
TODO: Big list of simulations, use as few physical rules as possible for speed, validating simulations, setting up simulations, stability
How do societies work?
Peter Berger begins "The Sacred Canopy" with the sentence: "Every human society is an enterprise of worldbuilding". He continues (in an uncorrect 1960s argot):
Society is a dialectic phenomenon in that it is a human product, and nothing but a human product, that yet continuously acts back upon its producer. Society is a product of man. It has no other being except that which is bestowed upon it by human activity and consciousness. There can be no social reality apart from man. Yet it may also be stated that man is a product of society.
This is a simple but important idea. While society is just "the stuff we do" it feels to us like an external world. It feels like an objective fact, because we grew up seeing the patterns of behavior that constitute our society, and we (rightfully) expect those patterns of behavior to be continued by new people we meet.
TODO: the sociological imagination, psychology, motivation, roles, power, labor, material flows, agreement, coordination, politics
What do babies need?
At a material level, babies need to be the right temperature, they need food and water (milk, ideally), they need to sleep regularly (maybe every hour or so). They need (or at least, prosper with) regular cleaning (mainly removing poops and peeps), clean air (less wood smoke, please).
At a somewhat less material level, they need to be in a place that's not too bright or loud or generally stimulating. Though sometimes babies like to be rocked gently and cuddled. Basically, they need to feel like they are in a safe place, and that other people are around to take care of them.
There are a lot of books and podcasts and other advise-giving-media about taking care of babies. Why so much? Well, most everyone agrees that babies are cute and delicate and helpless, and that to fail (as a parent / caregiver) to meet the baby's needs would be a huge waste, both of the effort of gestation, and the baby's essential cuteness.
That explains the desire to do a good job. But that doesn't explain why people (apparently) believe that they won't intuitively or naturally do a good job. That's harder for me to explain. Maybe partly because during gestation there's a sense of waiting at a train station for the baby to arrive. You have a few months, why not read something? You've got baby-on-the-brain, how about a baby book? There must be other reasons why people doubt their intuitions too, but I need to think more about that. Also: is this an international obsession, or just a particularly American self-help-culture phenomenon? There's this notion that many Europeans are more chill as parents and that Japanese parents are much less terrified of child predators. I haven't read much about either of those, so I'll leave off here.
How does biochemistry work?
All living creatures (and plants too!) have complicated chemicals inside of them. Some of the chemicals are made inside of the cells of the creature, and other chemicals are only available from food that the creature eats. For example, most plants create sugars and most animals can't create sugars, and need to eat sugars to survive. There are three categories of chemical molecules are that most important for animals: carbohydrates (like sugar), proteins (like the keratin in your finger nails), and fats (like olive oil).
Your body makes proteins by linking together smaller molecules (called amino acids) in specific orders. The instructions for linking together the amino acids are given by a molecule (called DNA) that describes how to make all the different proteins that your body needs to survive.
Will artificial intelligence take over the world?
That's hard to predict. The people who think that AI may control our lives in the future make (at least) four assumptions: 1) computer programs will continue to improve at making good decisions 2) there's no fundamental difference in the quality or type of decisions that a computer program can make (relative to a person) 3) eventually a computer program will have very different goals from people 4) eventually a computer program will be given enough power in the material world such that it can acquire more power more quickly that people can counteract it.
All of these assumptions seem basically reasonable to me. But both #3 and #4 depend a lot on the particular behaviors of the people and societies that work with AI. We also have examples of handling the risk of #3 and #4 when dealing with other humans, not just AIs. People can also acquire power quickly and can use that power to harm others. The difference is that an AI may be harder to counteract in case of a power struggle. For example, an AI doesn't need to have a body in a fixed location, but could distribute itself across the world, avoiding physical control. Similarly, an AI could quietly acquire more decision-making capacity, for example, by exploting insecure servers.