No text

124.4 meters (408.1 feet) tall, 9 meters wide, 5,500 metric tons at launch with a TWR of 1.6. Should leap off the pad and hit max Q in 45 seconds. These engines are grossly overpowered for the launch mass, which implies another stretch. And they're a work of art.

I have to go see one of these launches one day.

And they're not done. Raptor 4 is in the works. They really are going to Mars. The long dry spell of "boldly go" is coming to an end.

Class of '26, in before proctors!

This is about getting wholesalers to pay them to take the data about where they source their stuff and who they sell it to at what rates so Amazon can replace their step in the chain as well.

People act like life expectancies have been getting shorter.

They romanticize historic and prehistoric diets alike, as if they were utopian; as if people somehow intuited what to eat, or else that the constraints of supply somehow shaped digestive evolution like an intelligently designed metabolic symphony of symbiosis. That ignores the plain reality of volatile supplies -- even after the advent of agriculture, but especially before it -- and the reality that evolution is not driven by perfect health or life expectancy; only by surviving long enough to reproduce.

Even if modern diets are "unhealthy" (whatever that means), that doesn't imply that people were eating healthier at any point in the past. In fact, skeletal records clearly show that human existence has been rife with scurvy, rickets, iron deficiency, and stunted growth. Nutritional deficiencies were the norm, not the exception. Now (many people) have abundance, and that presents its own challenges, but the notion of an ideal, historic nutritional baseline is pure fiction. It's turtles all the way down.

Those cards would be worth far more than the data at the moment.

Everything in the visible universe is subject to lunar gravity now. That's how gravity works.

>arguing it unfairly advantages startups

Way to say your dealers suck.

Simplify. The best part is no part. The parts omitted never fail. They don't require maintenance, supply chains, continuous improvement.

LLMs have never been rules-based "agents," and they never will be. They cannot internalize arbitrary guidelines and abide by them unerringly, nor can they make qualitative decisions about which rule(s) to follow in the face of conflict. The nature of attention windows means that models are actively ignoring context, including "rules", which is why they can't follow them, and conflict resolution requires intelligence, which they do not possess, and which even intelligent beings frequently fail to do effectively. Social "error correction" tools for rule-breaking include learning from mistakes, which agents cannot do, and individualized ostracization/segregation (firing, jail, etc.), which is also not something we can do with LLMs.

So the only way to achieve rule-following behavior is to deterministically enforce limits on what LLMs can do, akin to a firewall. This is not exactly straightforward either, especially if you don't have fine-grained enough controls in the first place. For example, you could deterministically remove the capability of an agent to delete emails, but you couldn't easily scope that restriction to only "work emails," for example. They would need to be categorized appropriately, external to the agent, and the agent's control surface would need to thoroughly limit the ability to delete any email tagged as "work", or to change or remove the "work" tag, and ensure that the "work" tag deny rule takes priority over any other "allow" rules, AND prevent the agent from changing the rules by any means.

Essentially, this is an entirely new threat model, where neither agentic privilege nor agentic trust cleanly map to user privilege or user trust. At the same time, the more time spent fine-tuning rules and controls, the less useful agentic automation becomes. At some point you're doing at least as much work as the agent, if not more, and the whole point of "individualized" agentic behavior inherently means that any given set of fine-tuned rules are not broadly applicable. On top of that, the end result of agentic behavior might even be worse than the outcome of human performance to boot, which means more work for worse results.

The Moon is target practice. We need to get away from innovative bespoke engineering, into industrial mass production with continuous improvement. To do that we need to fly often. Mars just doesn't have the launch window availability. The biggest part of the challenge is that we were born in the bottom of a deep well. To toss enough stuff out of the well for a long journey is critical. Boosters that reliably fly on time often and cheaply enough to get ships and fuel out of the well. Ships that carry fuel into orbit and return over and over since the vast majority of the material we need to send out of the well isn't payloads or ships, it's fuel. Kilotons of fuel. Once the factories and processes are set up for that going far beyond the Moon is fairly easy. But with a narrow opportunity every two years that's not going to happen in a human lifespan. It's not enough refinement cycles per year.

I see this accelerating the Mars objective, not deferring it.

This is the crux. Optimization of supply chains to eliminate inventory makes them frail. Or, to quote Wirth:

Premature optimization is the root of all evil.

The last of the US federal helium reserve - including land and equipment - was sold in 2024.

Is joke of course. Was angling for the same joke.

3He is normal helium atom with an extra neutron, hoped to be used in some forms of fusion. It's not considered radioactive. Emitted by the sun it's trapped in lunar rock possibly at concentrations of up to 50 parts per billion but more likely 5-10ppb. The utility of extracting it from the Moon is hotly debated. On Earth isolating it from normal helium involves the same sort of centrifuge used to isolate isotopes of uranium, radium, hydrogen but there is far less of it than in lunar soil.

This is not actually the case in the subject at hand. It's all normal helium. When cooled enough all other gases will precipitate out as they freeze - including Hydrogen - leaving only helium as a gas and so easily isolated. That's actually why it's valuable since it's the only gas that will boil off at temperatures so low that the conductors immersed in the fluid will superconduct supporting the currents necessary for the intense electromagnets used in imaging and such. /Explaining nerd jokes since 1983