Engineer Zero

I happened to view a youtube video about underrated classic science fiction movies. One of them was called ‘Rocketship X-M.’ As I watched, my mouth slowly dropped open. An awful lot of coincidences to unpack here! And are they just coincidences?

‘Rocketship X-M’ is a low budget science fiction film that debuted in 1950. It involves an experimental rocket that flies four men and a woman to the planet Mars. But I’m not here to discuss the movie. I want to focus on the rocket itself.

Here is the rocket on the movie poster:

Here’s a scene from the movie of the rocket having landed on Mars:

I think that if you’ve come to this site, you already know that SpaceX is the rocket company owned by Elon Musk. You already know that SpaceX is developing a rocket called Starship that is intended to fly to Mars. So let’s get to the point of this blog entry: Starship appears to have been inspired by Rocketship X-M.

Here’s a side-by-side comparison of Starship in real life with Rocketship X-M:

Both have a cylindrical body and are approximately the same size and configuration. Both have a pointed nose with two fins placed near the nose. Both are designed to land vertically. And just as Rocketship X-M went to Mars in the movie, Starship is intended to go to Mars in real life.

Bonus design coincidence: SpaceX Starship has a booster stage, so too Rocketship X-M:

So how do we account for these coincidences? Perhaps the science advisor on the movie was really good at rocket science and came up with a similar optimal design as SpaceX would come up with almost 75 years later. I’m open to other suggestions.

And yes, I know that ‘X’ is Elon’s favorite letter.

Grok reports:

The Intuitive Machines lander, named Athena, touched down near the Moon’s south pole on March 6, 2025, but appears to have tipped over on its side, repeating the fate of its predecessor, Odysseus, which also tipped over after landing on February 22, 2024.

This isn’t a new problem in lunar exploration. Other moon probes have indeed failed or encountered significant issues due to tipping over. For example:

Luna 23 (1974): A Soviet probe designed to collect and return lunar soil samples landed on the Moon but tipped over due to an uneven surface in the Mare Crisium region. It remained operational briefly but couldn’t drill or complete its sample return mission because of its orientation.

SLIM (2024): Japan’s Smart Lander for Investigating Moon, launched by JAXA, successfully soft-landed on January 19, 2024, but tipped over onto its side after one of its engines malfunctioned during descent. It still managed to send data and images, and remarkably, it survived the lunar night to communicate again later.

It’s not hard to figure out why this keeps happening. Just look at the photos. Here’s the Athena probe:

You can plainly see this design has too much vertical and not enough horizontal. It’s just asking to be tipped over. Here’s a more stable configuration of probe and widely-spread landing pads:

“But the landing legs will have too much mass penalty!” No they won’t. Lunar gravity is 1/6 Earth gravity, so the legs only have to be 1/6th as sturdy — meaning that their mass need be only 1/6 as much as Earth-based struts.

Whatever the cost penalty for the extra mass of widely spread landing legs, it’s nowhere near the cost of losing the entire probe.

I mean, I don’t know this for sure, but that’s how I would place the bet.

If you’ve been living on Earth more than week or two, you probably know who Elon Musk is, but you may not know who Grok is. Grok is the AI avatar for X.com, formerly Twitter. If you have an X (formerly Twitter) account, you can ask Grok questions, have him generate pictures, and even analyze pictures that you upload. New features are being developed and added.

Maybe you’re thinking, “Well, okay, but given that computers can be programmed to play chess, what’s the big deal about a computer (ie AI) being programmed to play another game?” The thing about that is, an AI such as Grok probably isn’t being programmed to play any specific game. It’s being programmed to learn how to play games.

In other words, the AI is connected to the game through the internet just like any other player, and then without human assistance it studies and understands the rules, understands the screen images, follows the action on the screen, responds with the correct gamepad, mouse, and/or keyboard commands, and learns tactics and strategies as it goes along. And then it gets better and better.

I don’t know whether Grok has this ability. I don’t work for X, so I’m on the outside trying to look in, and the curtains are closed. But I have a feeling, based on what knowledge is public.

It starts a few months ago, when Elon Musk announced that he was a top-rated player in an online computer game called Diablo 4. Experienced gamers questioned this, because it takes hours and hours every day to build up enough points to appear high on the leaderboard. Even so, people shrugged it off. Maybe it was possible. Maybe Elon has a lot of downtime while shuttling back and forth aboard his private jet.

Then Elon claimed that he was one of the top-rated players in another online game, called Path of Exile 2. Now the gamers were sure that he was a fraud. Path of Exile 2 apparently is not a game where you can ‘grind’ your way to the top. It’s very complicated and you need to learn a lot of stuff in order to advance very far. Yet here was Elon, again rising to the top of the leaderboard.

He wasn’t even making much effort to hide the fact that he was lying about being a top player. In an interview he demonstrated ignorance about the basic mechanics of the game. If he’s going to hire someone or an entire team of someones to play the game for him, why not hire someone to whisper the correct answers to interview questions over a hidden earpiece? But he didn’t do that. He was making it obvious that he personally was not playing the game.

This has outraged the gaming world. But it’s not as if it doesn’t happen that wealthy people pay experienced gamers to boost their game scores for prestige. Also, if you want to be known as a top gamer so that you can attract attention to your game streaming channel, you might be tempted to do this.

But I think Elon has another motive. He’s using his identity to mask the identity of Grok as Grok learns the game and climbs the leaderboard. You can, with justification, claim that it’s neither fair nor honest for Elon to claim credit for Grok’s in-game achievements, but that’s really an insignificant matter. What’s monumentally important is that an AI has not only learned how to play a complex computer game, it’s gotten progressively better until it’s one of the top-ranked players.

From there, it’s still quite a few steps from learning how to navigate in the real world and succeed at the game of life. But Grok clearly is heading in the right direction and doing so with remarkable speed. And meanwhile, no other AI is anywhere near this achievement. No one else is even in competition, because the executives running the companies that have those AIs are being misled to think that Elon has gone crazy and is openly defrauding gamers by hiring experienced gamers to manage his gaming account.

I’ve lapsed into speaking with certainty here, and again, the truth is that I don’t know if this is what’s happening. I do know, however, that there are many people who despise Musk enough to gloat over accusing him of blatant gaming fraud and leave it at that. But perhaps in their gloating, they’re blind to the rapid evolution of his AI, and thus are enabling him to win even more at the biggest game of all, the game called In Real Life.

Imagine a train with rail cars filled with thousands of drones that are programmed to fly buckets out to the Pacific Ocean, scoop up water, and drop on brush fires. The train goes along the California coast to where its needed and releases the drones. Diesel generators provide battery charging.

If each drone costs $1000, even a million drones would cost only $1 billion. That isn’t that much, when neighborhoods of $5 million homes are at stake.

China can choreograph thousands of drones to put on a light show of a dragon flying over a city, so why can’t we program drones to put out a fire?

This is Eric Berger’s second book on SpaceX, the first being Liftoff. Liftoff was about the founding of SpaceX up to the successful launch of the Falcon 1. In this book, Berger covers the story of SpaceX up to the present.

The story covers a lot of ground. Falcon 1 was followed by Falcon 9, a much larger launch vehicle. Then came Falcon Heavy. Then came Dragon capsule. Then re-usable boosters, which at this point is the crown jewel of SpaceX accomplishments and, obviously, is the inspiration for the book’s title and cover.

Musk was very much the central character in Liftoff, but in these pages he is more in the background. Whereas in Liftoff he was the motor than made SpaceX run, here he sometimes seems to be getting in the way. He seems to lose his temper more often, so much so that ends up on the verge of firing his best people, and not so much because they made a blunder but because they said something that he didn’t like.

I respect what Musk has done, and you have to give him credit for doing it. But is he a boss that you’d want to work for?

For example, there’s an incident while SpaceXers are moving a booster cross-country via flatbed trailer. They have to detour into a small town and the trailer has to make a tight turn around a corner. Everyone is on their walkie-talkies, ready to inform the driver if the booster is in danger of brushing against the building. But since they’re inexperienced, it doesn’t occur to them that if they all shout a warning at once, the walkie talkie signals will interfere and cause squelch and the driver won’t hear anyone’s warning. And so the booster was damaged.

Musk accepts the mishap calmly (seemingly) but then one of the people at the scene makes a casual remark relating the incident to a similar incident while moving the Falcon 1 booster, and Musk goes ballistic when he learns about it and phones the supervisor, telling him to fire the guy. The supervisor is driving in his car at the time and the guy in question is sitting next to him. The supervisor hangs up and the guy asks him what Musk said. The supervisor says, “Nothing,” and doesn’t fire the guy, and Musk forgets about it.

The main problem with Musk’s management style is not so much that he fires people by whim but that he consistently drives them into quitting. Story after story in the book confirm that burn-out at SpaceX is a serious problem, and the main reason people stay on is not because of the money and status but because they believe Musk’s vision to colonize Mars, make humanity interplanetary, and ‘spread the light of consciousness throughout the universe.’

That latter goal sounds very poetic, but as many a philosopher has noted, “For happiness to be got, it has to be forgot.” We’re often happiest when we’re least self-aware, and while engineering work can be engrossing, burn-out makes us very conscious indeed — of pain.

If Musk can make Starship reusable, it will open up the solar system to exploration and colonization. No one else in the space business is even close, and most aren’t even focused on such a goal. Thus it seems that humanity has two modes: do-nothing or overwork. Work-life balance is not on the corporate agenda, which appears to be great for the bottom line (for now).

However, if we don’t have time for families, we aren’t going to have quality families, and human consciousness will extinguish because there won’t be human generations to perpetuate it.

I wonder if it’s possible to spread the light of consciousness throughout the universe on a forty-hour work week.