And I don’t mean just diet or sleep. I think what we enjoy, what we’re drawn to, the people we spend time with, and even our inner life play a part. All of it shapes us in one way or another, and sometimes it shows
I think there’s something to physiognomy. how we eat, sleep, live, and spend our time eventually shows up in our faces, expressions, and the way we carry ourselves.
You start noticing these little patterns over time, often without even realizing it
Classical education was built around a simple idea:
A young person should learn how to think before being told what to think.
That meant studying:
— Logic to recognize bad arguments
— Rhetoric to express ideas clearly
— History to understand human nature
— Philosophy to examine difficult questions
— Literature to understand other people
— Mathematics to develop reasoning
These subjects were not merely preparation for employment.
They were preparation for life.
One thing the engineering community will (re)discover is that every program can be ① hardened (edge cases squashed, inputs tightened, errors handled) and ② optimized (profiled, benchmarked, rewritten) basically ad infinitum.
There are real costs in every direction: time, attention, opportunity. You can literally explore input spaces that will never be hit or optimize performance of something that will never be used ("you have no users".) Even in an infinite token world, you need to know when to stop and what tradeoffs to accept. Agents will happily drill no matter what.
In my experience as a creator, I learned that whathever you do you must genuinely be passionate and enjoy the process, no matter the final goal. Chasing just that final goal is a road to perdition most of the times.
New video: basic reversing of a Windows kernel driver.
WESP, the new endpoint security driver in Insider builds, has no device objects and no dispatch routines. Its imports tell you how it really talks to user mode: a filter communication port.
Full write-up: https://t.co/ZWtza7feUt
will be surprised if it ever grows more than 10%, directly paying for AI. They will pay with their private data, not money, and not to AI companies but to Alphabet, Meta, etc.
Today, more than ever, humans need to:
• Set goals
• Work toward them
• Overcome obstacles
• Achieve them
The important thing isn't domination over other people. It's the experience of having meaningful control over one's own life.
Took a while, but here it is, a mega write-up on language models for text classification and, of course, Jev.
It's basically a visual guide to RNNs, CNNs, transformers, and calibration, with hands-on experiments on accuracy and efficiency.