2023 letter | Zhengdong
The first and most important sense is touch. Touch tells the edges of a good abstraction. Computer engineers find this intuitive. From Dijkstra we have “Separation of Concerns.” Distinct components of a system should handle distinct concerns. From others, Don’t Repeat Yourself, You Ain’t Gonna Need It, Keep It Simple, Stupid, and so on. I don’t think it’s controversial to say that all engineering really is, is creating the perfect abstraction. The end of the year is a time for imagination and polemic. So take this to a further extreme. The act of research—no, all intelligent behavior—is about the perfect abstraction. Dijkstra said as much. Separation of Concerns is from his essay “On the role of scientific thought.” In it he explained “what to [his] taste is characteristic of all intelligent thinking.” So let me give a flavor of the problems of distinction that research engineers face. Then, I will propose an abstraction that I cannot describe in any way except perfect. Everybody knows
2023 letter | Zhengdong 2023 letter | Zhengdong Wang [home] | [all posts] 2023 letter 2024-01-01 21:44 GMT Research engineers, like other human beings, have five senses. The first and most important sense is touch. Touch tells the edges of a good abstraction. Computer engineers find this intuitive. From Dijkstra we have “Separation of Concerns”—distinct components of a system should handle distinct concerns. From others, Don’t Repeat Yourself, You Ain’t Gonna Need It, Keep It Simple, Stupid, and so on. I don’t think it’s controversial to say that all engineering really is, is creating the perf
Explore this link on the map →saved by
related reading
- LessWronglesswrong.com
- The machines are fine. I'm worried about us.ergosphere.blog
- Toward a Critical Technical Practicepages.gseis.ucla.edu
- Sporadicamichaelnotebook.com
- Designing AI for Disruptive Scienceasimov.press
- The Plan - 2023 Version — LessWronglesswrong.com
- A Pragmatic Vision for Interpretability — AI Alignment Forumalignmentforum.org
- How should we evaluate progress in AI? | Better without AIbetterwithout.ai
- Complex Systems | The Point Magazinethepointmag.com
- Togeliustogelius.blogspot.com
- Vivek on X: "some notes on getting into frontier ai labs" / Xx.com
- The Paradox of First Principles (And Why Nobody Wants to Hear That You're in Pursuit of Greatness) - Sohail Mohammadsohailmo.ai