In Werner Herzog’s Rogue Film School, he taught his pupils how to pick locks or forge signatures. Herzog says the reason behind that curriculum is that some films can’t be made if you don’t “have the right criminal energy and the skills”.
(His movie Fitzcarraldo, whose production Wikipedia calls “troubled” only to then mention that a 320-ton steamship had to be pushed over a hill in the Amazon rainforest and that, at the end of filming, the chief of a native tribe told Herzog they could, you know, take care of the problem for him and kill Klaus Kinski, that movie couldn’t have been made, in Herzog’s words, without “massive forgery”.)
Now, that’s what Herzog says as to why you need criminal energy and “the skills”. My interpretation of the curriculum is different. And yes, I’m going to use my criminal energy here to re- and possibly misinterpret the founding idea behind a school I neither visited nor was involved in.
I think that he’s teaching how to pick locks and forge signatures and be rogues because you need to live a life, ideally a full life, in order to make films.
Because if you don’t have that, what are you going to make films about? About school? About learning how to make films? No, you have to have something worth sharing, something worth talking about: experiences, ideas, opinions, doubts, delights, heartbreaks, and questions.
And I think it’s the same for programming now.
Yes, it’s always been the case that programming alone, the typing of code and the construction of computer programs, wasn’t enough.
But there was so much typing, and syntax, and libraries, and function signatures, and all that stuff to wrestle with that the ratio was off: for one bit of idea you had to spend ten bits of programming.
Now it’s much more balanced and you have to figure out what to do with all the programming you can now accomplish.
So, if I had to build a Rogue Software School, I’d get my students to learn about business, about startups, about marketing, about markets, about the places where software is made and sold; I’d want them to learn about how software is used and why, learn about the trends involved in all of that, learn how to market and sell software.
And I’d also want them to spend time in places where nobody talks or even thinks about software. Work on a construction site, unload trucks, help in a restaurant, work in a bookstore, run a youth soccer club.
Find something that gives you a reason to build software, because programming isn’t enough anymore.
Note: This will be the last Joy & Curiosity this year. I’ll be back in January. If you want me to come back earlier, tell your friends about this newsletter.
Completely forgot to mention this last time: we shipped Boxes! What’s a box? It’s simply a computer, a VM. You can tell Amp now to deploy something to a box and it’ll do it. You can have as many apps as you want running on a box. Apps, daemons, scripts, whatever you want. It’s just a computer. There are no containers or other isolation. Agents absolutely love that. I have a tiny box on which I run my newsletter-link collector app, my metronome app to practice drums, a festival store app for my local club, and a bunch of other things. Best of all: on each box there’s an agent running. No longer do I have to carefully edit text or other content or data in production. No, I can just point the agent at it via the built-in widget and say “format this nicely” and it’ll do it. Jellyware!
Bad ideas that aren’t bad anymore. Yes, yes, yes, it’s what I’ve been saying, yes, rethink everything, yes, a fundamental constant in a very big equation has changed and you need to rerun the calculations.
ADVICE WANTED: “If you could write one essay to every scared new grad and young person in software… What would you tell them?” I’m going to be a judge for this writing contest, which not only means that I’m Very Important and Smart but also that I’m Cool and Sexy and maybe I’ll wear a wig like the judges in England (or whatever they’re called) when reading submissions. And it also means that I’d love for you to submit an essay. I’m serious. I really am Smar— no, I mean: submit an essay!
Levent Alpöge claims to have proven the Riemann Hypothesis. I can’t tell you whether it’s a valid claim or not. It’s 250 thousand pages long, which is quite the hurdle to declaring you agree or not, but, maybe even more importantly, I also don’t know anything about math. What I can tell you is that I recommend you read the first two pages of that PDF. “[…] this time Claude far outpaced me near the finish line. No, Claude flew, and danced, and sparkled near that finish line. I had us on the line of combining two quite different ideas present in a toy setting, one of which was already new and shocking, making very simple one of the shadows of Riemann over which we mathematicians felt most triumph. I guided us to the right situation analogous to that toy setting, and then Claude began to leave me in awe.”
Joscha Bach: “It still feels to me as if dropping through an event horizon. That is, there is nothing specific about any given moment, but the rate of change and weirdness is constantly speeding up, and we won't ever return to the before times.”
Scattered thoughts on how doing math feels these days. This was a great read. And, funnily enough, I came across that post a few days before the whole Riemann Hypothesis AI sensation extravaganza.
This is absolutely wild: “BREAKING: Rockstar Games says Grand Theft Auto VI sold $3.79 billion in 24 hours, making it the biggest entertainment launch of all time. Pre-sales of the game reached 32.5 million units.” It does seem like everyone is talking about GTA VI and I can’t help but think that I agree with the people shouting “bring back monoculture!” because it is nice having shared experiences like this.
The School by Donald Barthelme. This did something to me. The amount of inspiration you can get from a 4-page PDF, huh?
Hillel Wayne on his Least Favorite Example in Software Engineering. Great post. In response to Hillel, Richard Feldman also shared his least favorite example. Examples are important! So, so, so important. In case you haven’t read it: Examples are the best documentation.
I started reading the first of Knausgård’s My Struggle books. It all started with a post I saw a few months ago. It was a picture of a young girl reading the book and the post made the rounds due to (I assume) the dichotomy of a young girl reading a book whose original title in Norwegian is Min Kamp, which is, well, the Norwegian version of Mein Kampf, which is, well, … So I started reading the Wikipedia page, then I read it again and again over the next few weeks and became increasingly fascinated. Finally, right before the flight to London last Sunday, I bought the first volume for two euros and started reading. I was hooked immediately. I’m now halfway through the book and I can’t remember the last time I was this into a book (which is also why there are fewer links in this issue). On nearly every page I wonder: how does he do it? How does he remember? How does he weave the events like this? How does he create this sense of suspense when describing a normal afternoon of a teenager? Why is this so insightful? It was a huge bestseller, Wikipedia told me, but I don’t know what the general perception of the books is, but to me it feels like great literature. It’s very good.
Yes: “Theater is simply not a current artform, it’s a historic one. To write a “play” is to purposefully engage in anachronism. Current artforms include sigma edits, moodboards, nanofiction, DM dialogues, mass shootings, AI fruit commedia dell’arte, video essays, thirst traps, etc.” The last sentence I’m not sure about, at least not entirely.
Ben Thompson: Obituaries.
I really like these two terms here: everflux software and translucent sources. Everflux software means that “instead of stepping from release to release, months apart, software now evolves continuously.” And translucent sources: “Today Claude will happily pick apart any Electron app, and any binary shipped with debug symbols has become a reverse-engineering playground for LLMs. In other words, much so-called closed-source software is now open source in all but name, and keeping code closed makes less and less sense.”
Henrik Karlsson on learning from DFW: You learn more about writing from close reading than from writing advice. This made me buy a subscription to his Substack. Great writing, great piece, as always by Karlsson. And yet… Something always rubs me the wrong way when someone wants to ascribe specific intentions to the author of a piece of writing, music, art. That’s not exactly what Karlsson is doing here, but I couldn’t help but think of all the different pieces of Hendrix analysis I read, listened to, or watched over the years. “And then he did this because of… and after that he did so he could…” It all sounds neat and makes you go a-ha! But then you watch an interview with Hendrix and listen to enough of his music and have to come to the conclusion that, well, it’s more likely that he didn’t think about any of this at all. It just felt right to him.


