Jonathan Blow - Preventing the Collapse of Civilization (English only)

I took a couple of compiler courses in university, and I think for the most part it made me a better programmer, mainly because it enforced understanding of how specific programming concepts and constructs are generally implemented. That, and a more intense look at the "heap of abstractions" you're generally working with when you write a higher-level language. I've always advocated that, if you want to really understand the tools you are using, you need to understand at least one level below the "surface abstraction" that you are working if. Even better if you can understand two or three levels down. There's a great talk from game developer Jonathan Blow [1] that describes how knowledge is lost "generationally" due to our lack of understanding of the black boxes we build things on top of. Not sure I 100% agree with his thoughts, but it's an interesting take. [1] ▶ this video

in “An Intro to Compilers (2017)” · View on HN

It costs so much mostly because programmers don't know how to write software cost-efficiently. As everybody follows the same path, VCs start to believe the cost is increasing when in fact it's the programmer's knowledge of cost-efficiency software design that is decreasing. Take a look at this: ▶ this video The costs to develop software are ridiculous and unrealistic. It turns out that if enough people act as ridiculous and unrealistic that becomes the new normal.

in “Ask HN: Why is development so expensive?” · View on HN

Of course that's true - we probably don't need to start re-teaching the abacus... but I'm careful about getting too complacent. When everything has been going well for a long time it's tempting to think this will continue forever. But one, say, global pandemic can bring everything down - and some of our modern conveniences can be gone... possibly for centuries. An interesting (programming) talk by Jonathan Blow on the subject of losing knowledge! ▶ this video

in “Amateur radio skills prove useful during bushfire emergencies” · View on HN

The Soul of Erlang and Elixir, by Sasa Juric video: [link] HN discussion: [link] Preventing the Collapse of Civilization, by Jonathan Blow video: ▶ this video HN discussion: [link]

in “Ask HN: Best Talks of 2019?” · View on HN

If you'd like to venture further into this rabbit hole I can recommend "Preventing the Collapse of Civilization": ▶ this video And "The Thirty Million Line Problem": [link]

in “Docker is vulnerable to a symlink-race attack” · View on HN

Well-known game developer Jonathan Blow has a lot to say about software and technology more generally being in decline due to skills being lost when older generations retire [1]. In my own experience, writing code with pointers in it used to be a skill expected of entry level programmers, and getting the hang of it was something of a rite of passage. Around the time Java started being used as a teaching language, it became best practice to rely on the language's standard libraries instead of rolling one's own linked list or whatever, and I instructed university students accordingly. As things have developed, the design effort put into Rust seems to imply that pointers are now regarded as a grand challenge problem. These days I don't tie my own neck ties any more but wear the clip-on kind because the knots have been developed and tested by experts. [1] ▶ this video

in “Ask HN: What are the “lost arts” of our field?” · View on HN

No. This is the absolute wrong direction. In effect, you're giving a third-party absolute control over your code and your ability to work on it. You're also disconnecting the developer's mental model of how all of this stuff works (arguably why there's so much bad code floating around). Eventually, that will devolve to people not knowing how any of this works and the industry will collapse. It's an abstract line of argumentation related to this, but Jonathan Blow nailed it on this idea a few years back: ▶ this video

in “The End of Localhost” · View on HN

While not incompatible with ECS, the DOM and this renderer go all-in on the javascript event-loop. You would have to write your own run loop, which executes the systems on every frame (ideally creating a DAG and executing in parallel while possible), and leave the event loop behind, with all the niceties like `onClick`, to go full ECS. Otherwise you'll create some Frankenstein monster of part ECS, part event-loop, part declarative React. Additionally, you can throw OOP in that mix as well, because Three.js has it's own whole OOP-style framework, that you're strapping declarative React on top of with this renderer. Reminds me of Jonathan Blow's talk on the end of civilization via endless layers of abstraction[1]. I really think, when it's ready, a Bevy[2]-style system either native or compiled to WASM with WebGPU will be ideal. And while I'm airing opinions (forgive me), I think writing shaders now is like SQL 30 years ago. Developers left optimizing difficult--according to them--SQL to database administrators by abstracting it away into ORMs. If history is any indicator, I think we'll be having the same arguments on Hacker News 30 years from now about 3D frameworks vs writing shaders directly as we're having now about ORMs vs writing SQL directly. [1] ▶ this video [2] [link]

in “React Renderer for Three.js” · View on HN

The author of this article, Samo Burja, also gave a great lecture in 2018 called “Civilization: Institutions, Knowledge and the Future” ( [link] ) which was mentioned in Jon Blow’s popular 2019 talk on software complexity, “Preventing the Collapse of Civilization” ( ▶ this video ). Burja’s talk is worth watching if you’re interested in his analysis of technology and history over an even longer period, and it provides some context for his remarks at the end of the article about intellectual institutions and the benefits of centralisation.

in “The Centralized Internet Is Inevitable” · View on HN

You've probably seen Jonathan Blow's talk on the subject: ▶ this video I'm super interested in the subject: [link] . In the framework of this thread I focus on 2 of the 3 layers: problem and software. Hardware is out of scope. For now. Recently I've started thinking about BIOS (inspired by [link] ). So perhaps it's only a matter of time.

in “Bootstrapping a Forth in 40 lines of Lua code (2008)” · View on HN

Quality Software : Yes, this is becoming more and more important and is overlooked at many levels. With more of our infrastructure, lives and the future of humanity in the hands of a bunch of software & hardware. We need to take things a tad bit more seriously. I believe this is not yet a big problem but it is becoming one. ▶ this video [link]

in “Ask HN: What Are the Big Problems?” · View on HN

I agree with the sentiment that they only augment our capabilities, but then we end up with another problem, which is that the people who write the software are not fully aware of how it all works. Relevant to my argument is Jonathan Blow's `Preventing The collapse of Civilization` talk in which he discusses the disappearance of knowledge as generations pass: ▶ this video

in “AI writing code will make software engineers more valuable” · View on HN

> I'd expect the bar to preserve any specific content to be a lot lower than what we had in the past. At the moment, I agree. However, as older civilizations rose and fell, their knowledge lived on in physical support, directly accessible by anyone who was taught how to read the language, therefore the assumption was that "as long as the language lives on, books are a good archiving support". When our modern digital civilization falls [1], I'm not sure most of it would be transmittable through long-term electronic archiving supports that may be easily readable (as easily as a written book at least) by future civilizations, because not only the language has to live on, but the technology needed to access the content also has to function. I also agree that some contents will always be more relevant to "backup". As for everything that happened on Facebook (same can be said for most social platforms), it is likely to entirely disappear with the company. One could even see it as a parallel civilization, that has no means of self-preservation other than that of their ruling entity's business interest. [1] Jonathan Blow - Preventing the collapse of civilization - ▶ this video

in “How the invention of the book changed how people read” · View on HN

This reminds me of recent talk by Jonathan Blow [1], where he talks about how we've made very little progress in the field of software and anything that appears to be progress is just software leveraging better hardware. It's quite scary how low our standards have gotten. [1]: ▶ this video

in “"DigitalOcean Killed Our Company"” · View on HN