Literally the manifesto of a supervillain but unironic. Fortunately, I don't think the libertarians will conquer space given how things are going. They might accidentally start a space war though. Then we'll get space UN the hard way, slanted towards China and US.

2/Institutional experience from past wars do matter, but only in the short and medium term. In the long term, as long as the combatant has the industrial might and political endurance to stay in the fight, they will eventually gain experience the only way they can: the hard way.

4/Female labor participation and more attention per child is good social progress on a micro level and for the individuals involved. But for various reasons this has caused the associated economics to be unviable not just by multipliers, but by exponents. Just my personal ideas

3/A good govt school needs rigorous entrance exams=lots of paid tutoring. Not even taking into account other supplemental costs. Also, many women even in my parents generation defer having a child until its almost too late (late 30s/40s) as women are expected to have good careers.

2/However, each parent now wants the best future for their child (instead of just being free labor). To secure that is orders of magnitude more expensive than any subsidy can achieve. i.e. in education, a good international school is THB700K-1M per year. Avg TH income is 12K per month. Do the math.

3/ Hard economic times create zero sum thinking and radicalism. Manifests in tariffs and unrest. Made worse by the interconnection of global trade. Globalization of trade is optimization at the cost of robustness. I suspect self-sufficient economic blocs may be more resilient even if inefficient.

2/Germany pushed for Lebensraum and war in the East. Japan coveted resources in Manchuria and the Dutch East Indies. Militarist elites concluded it was "expand or die". Or at least relegate yourself to a second-rate power dependent on Anglo domination, which they weren't willing to do

The quality of an LLM response, I've noticed, is only as smart as the human writing the prompt. Perhaps a LLM lab is a way to bootstrap and automate the making of insightful and intelligent prompts, perhaps by combining LLMs likely to hold different knowledge and training data

We become people who contort ourselves to fit optimization problems built with assumptions that weakly fit reality. Perhaps we even convince ourselves those optimization models fit reality because we want the illusion of objectivity.

some combination of random search for solutions and genetic algorithms to select and combine best solutions over many rounds. vaguely remember seeing some video about janky robots learning to walk after a thousand rounds. tho this was years before deep learning et al. became The Big Thing