How to walk through walls

On hacker mindset

In March 1991, Robert Rodriguez, then 22 years old, decided to write and shoot three feature-length home movies to gain experience making full-length films, in case he ever received an offer to direct a real one.

Nine months later, having finished El Mariachi, the first part of his planned trilogy, Rodriguez found himself in the office of Robert Newman, a Hollywood agent. Watching the trailer Rodriguez had cut, Newman, who would go on to sell the movie to Columbia in a deal worth $1.8 million, asked:

“How much did it cost [to make] again?”

“$7,000.”

“Really? That’s pretty good . . . most trailers usually cost between $20,000 and $30,000.”

“No,” Rodriguez said, “the whole movie cost $7,000.”

In nine months, he had written, directed, and sold a 90-minute action film that cost a third of what a film trailer would. How was that possible? At the time, the cost of film stock alone would normally run into several hundred thousand dollars for an action film like El Mariachi.

During the press tour, the journalists thought the story about the $7,000 was too outlandish to be true, and Rodriguez had to show them a behind-the-scenes video to convince them. In that video they could see that the reason Rodriguez, the son of two Mexican immigrants with ten children in San Antonio, had been able to make a commercial big-screen action film from his private savings was that he had a hacker mindset.

Hacker mindset

I learned the term hacker mindset from Gwern, a pseudonymous blogger, who wrote about how people like Rodriguez think in his 2012 essay “On Seeing Through and Unseeing.”

To explain the hacker mindset, one example Gwern uses is people who set world records in video games, doing so-called speedruns.

Unlike normal sports, where the athletes are usually at best twice as fast as a healthy adult, video game speedruns can be so insanely much faster than a normal playthrough that a normal person can’t even understand what is happening on the screen. The last game I played was Legend of Zelda: Ocarina of Time, in my parents’ house in the late nineties. I put in some 30 hours, and if I remember correctly, I didn’t even finish it—but when I look at the current speed run record, I see that Bloobiebla & MrGrunz have finished the game in 20 minutes and 9 seconds. I can’t wrap my head around how that is possible.

And I don’t get much wiser when I look at the recording and notice that a substantial part of it is them running backwards with a hen on their head. If we want crazy outcomes, I guess we have to accept crazy behavior.

The difference between Bloobiebla & MrGrunz and me is not, primarily, that they are faster than I. It is, Gwern points out, that they see the game differently. When I played Zelda, I saw “villages” and “hen.” But they have a hacker mindset, so they know that there aren’t actually any villages and hens.

The game, Gwern writes, just “pretends to be made out of things like ‘walls’ and ‘speed limits’ and ‘levels which must be completed in a particular order.’” But what it actually is, at a deeper level, is bits, code, memory locations, processing units, and so on and so forth.

And because they see the game at this level—and understand how it’s put together—Bloobiebla & MrGrunz can make moves in the game that I couldn’t, such as “deliberately overloading the RAM to cause memory allocation errors” (perhaps this was what running backwards with a hen on their head did) which, Gwern writes, can “give you infinite ‘velocity’ or shift you into alternate coordinate systems in the true physics, allowing enormous movements in the supposed map, giving shortcuts to the ‘end’ of the game.”

And lo and behold, soon after Bloobiebla & MrGrunz drop the hen, they fall through a “wall” and land in the final “level.”

Because I’m watching the abstraction that the game is pretending to be—a cute fantasy world with villages and swords and horses—this looks bizarre to me. But it makes perfect sense when you understand what the game is at a deeper level.

Most systems can be viewed at multiple levels. There is a superficial system which pretends to be made of one thing (walls, hens). But actually, it is really made of something else (bits, memory allocations). And if you learn to understand that underlying system, you can find ways to use the lower-level details to steer the system in a way that looks incomprehensible to those who only see the more superficial system.

Robert Rodriguez’s classmates must have experienced a bewilderment of this kind when they saw him go down to Mexico with $7,000 dollars and return with a film showing in cinemas across the US. That was not a move that was part of how they’d been told the game that is the film industry works; but it was a move that was perfectly compatible with the facts of cameras, lights, and Hollywood deal-making if you understood them at a deep enough level.

Rodriguez could speedrun a film career, walking through proverbial walls, because he saw through the game to its underlying mechanics. He had the hacker mindset. He was willing to get his hands dirty and learn the practical realities. He saw that a lot of what the other film students took for reality were just fictions they’d been taught at school.

This is a bit vague. Let me give some concrete examples of ways he saw through the system his classmates took for reality.

At film school, they were taught to work with a crew, where someone specialized as a cameraman, another as a sound technician, and so on. But Rodriguez, who had always done movies on his own as a kid, knew that that was just a convention. If he could figure it out, it would be more effective to do all of the technical work himself, which also meant he wouldn’t have to pay for a crew.

This seemed insane to others. On July 24, 1991, three days before leaving for Mexico, his film teacher asked him who would be the director of photography.

From Rodriguez’s diary, published as Rebel Without a Crew:

I know he’ll shit all over me if I tell him the truth—that I’m planning on shooting it all by myself, without a crew. So I told him, “I’m going to be the director of photography, but I’ll probably have a small crew around to help out.” He shook his head. “No, no, no, no … You’re going to fail! Your actors are going to hate you. They’re going to be sitting there waiting for you while you light the set. Don’t be an idiot. Get a director of photography.”

Instead, Rodriguez bought 250-watt bulbs that he screwed into the existing lamps on the sets, and that was that for lighting.

At film school, they had also been taught to shoot several takes from multiple angles so the editor could shape the scene, but since Rodriguez was the editor, he could visualize exactly what the scenes would be, shooting only precisely what was needed, a single take per scene, which minimized the cost of film stock and editing time.

A thousand small optimizations like this meant he could shoot the film in ten days while staying with the lead actor’s mom in Ciudad Acuña during summer break.

More examples

There are similar shortcuts in most domains if you learn to see through the abstractions and unsee the conventional ways of viewing something. There are deeper levels to most systems if you are willing to take them apart.

For example, when I grew up, I was told that there was such a thing as a “job,” and these were listed on “job boards” where you could read about the “qualifications” necessary—qualifications that you got through something called “education.” This isn’t false. You can play the game this way. But it is a very superficial way of playing the game.

A slightly more precise reading is to say that the economy is made up of 8.25 billion people, all trying to solve various problems—and what “getting a job” really means is finding a person with a problem and convincing them that you can solve it for them. This can be done by looking at job boards, of course, where people who collaborate on solving problems (aka work at companies) list some of the problems they want help solving.

Now, you notice more things you could do. You could, for example, go talk to people directly and convince them that you can solve their problems. Or, you could work in public, sharing projects you are building on the internet or elsewhere, so people can see what you do and reach out. If you want to work at a specific company, you could talk directly to the employees you’d want to work with, understand their current problems, and then solve their problems for free, so they lobby their superiors to employ you. People who operate at this level tend to have more interesting careers than those who play the game by looking at job postings.

Another everyday example of the hacker mindset can be seen when agentic people deal with bureaucracy.

Companies and governments like to pretend to be formal, machine-like systems, where things have to be done in a specific way. But this, just as in the case of the video game pretending to be made of levels, is an abstraction. A fiction. Actually, a bureaucracy is just people and some file systems. Calling and asking to speak to a supervisor, or showing up in person, or finding the specific person who handles your case, often lets you bypass the “system.” If you search Patrick MacKenzie’s tweets, you get a long stream of great examples of him doing this—getting the customer service agent at an airline to buy him a ticket from their competitor when his flight was delayed, for instance, or calling pharmacies to create an inventory of the US vaccine stock as he did together with a group of volunteers when the US government failed to keep track of the vaccine stocks during Covid.

How do people develop a hacker mindset?

One thing I personally find useful is to read about people who have it and notice what they do.¹ This has helped me see possibilities that I had been blind to because I had a too superficial reading of the system.

It also helps if you can surround yourself with people who have a hacker mindset. I suspect this kind of cultural osmosis played an important role for Rodriguez. His dad was a self-employed salesman and was always trying new things—they seem to have been a family that encouraged taking machines apart and doing stuff yourself. Rodriguez also had a great boss at his first job:

My first job in high school was at a photo lab and I remember what my first boss, Mr. Riojas, told me one day after he saw some of my cartoons and photographs. He said that I had creative talent, but what I really needed to do if I wanted to be successful was to become technical. He said that just about anyone can become technical, but not everyone can be creative. And there are a lot of creative people who never get anywhere because they don’t have technical skills. Part of what makes a person creative is his lack of emphasis on things technical.

My boss said that if you are someone who is already creative, and then you become technical, then you are unstoppable.

This is another common pattern among people who have a hacker mindset. They have gotten their hands dirty playing around with the technical parts, insisting on understanding every aspect of the work, “weaving [the] system into [their] mind[s] so tight that it’s hard to find the stitches after a while,” as Alice Maz writes about her experience becoming incomprehensibly good at Minecraft.

When Rodriguez made his first feature film at 23, he had already spent a decade making home videos, editing them by using two VCRs, so he could play the raw material on one and record the bits he wanted on the other. By working hands-on, guided by his own needs, he had learned the details of the work and how things could be manipulated in such a way that his films looked good even if he had no crew or budget.

In an appendix to the diary, he writes:

The most important and useful thing you need to be a filmmaker is “experience in movies,” as opposed to “movie experience.” There’s a difference. They always tell you in film school and in Hollywood that in order to be a filmmaker you need to get “movie experience” so you can work your way up in the business. The reasoning being that by working on other films, even as a production assistant, you get to see firsthand how others make movies. Now, that’s exactly the kind of experience you don’t need. You don’t want to learn how other people make movies especially real Hollywood movies, because nine times out of ten their methods are wasteful and inefficient. You don’t need to learn that!

“Experience in movies,” on the other hand is where you yourself get a borrowed video camera or other recording device and record images then manipulate those images in some kind of editing atmosphere. Whether you use old ¾” video editing systems, VCR to VCR, or even computer editing. Whatever you can get your hands on. The idea is to experience creating your own images and/or stories no matter how crude they are and then manipulating them through editing.

That is, you want to avoid learning the conventional wisdom about how something works—which is always simplified and filled with false walls—and instead focus on getting into very close contact with the actual nuts and bolts by doing everything yourself. That is how you will learn to understand the system well enough to “see through” it.

It might sound like a depressing conclusion to this essay: the way to find shortcuts is to first spend ten years learning all of the technical details.

But it is not depressing.

What we’re talking about here isn’t like going to school—it emphatically is not that—suffering through all of the boring prerequisites before you get to do the exciting parts. What we’re talking about is actually doing the fun stuff, playing around with projects that excite you, trusting that you can learn enough to solve your problems. If you keep tinkering, doing one fun project after another, you will eventually see through the system.

Also, it is only the first time that it might take years. After you’ve developed a hacker mindset in one area of your life, it is much easier to see the rest of reality in the same way.

Having seen through the superficiality and clumsiness of the normal way of doing things, you are less likely to trust conventional wisdom going forward and more likely to trust your eyes. You know that there are deeper layers to reality and have a sense for how to access them.

And then all of reality becomes something you can horse around with.

---

Drafts of this essay were discussed with Johanna Karlsson. The copy edits were done by Esha Rana. Any remaining mistakes are mine.

1. Rodriguez’ diary, Rebel Without a Crew, is one example. “The Story of VaccinateCA” is another. I also like “Playing to Win” by Alice Maz. A Guide to the Perplexed with Werner Herzog. Surely, you're joking Mr. Feynman. Robert Caro’s The Years of Lyndon Johnson. Some of these examples are more unethical and problematic than others, so beware. If you lack ethics, hacker mindset can be used in manipulative and anti-social ways. And that’s a sad way to live.