I once met a man wearing a t-shirt which said, "Mehr Wasabi bitte". An impressive man

This is from 2017. Nothing has changed, except there's a real one now

Zuckerberg's lost it. Making a battery the size of Manhattan?

The Transformer algorithm, which is at the heart of almost all genAI models, is quadratic. O(n^2). That means it doesn't scale. Making a bigger battery isn't going to fix that. Your brain runs on 20 watts.

Trying to say, "everything else will stay exactly the same, except we are scaling up" 40 years ago, would mean that your phone would weigh as much as a car. A ridiculous idea, don't try to make portable phones, it's impossible.

It's ironic though, I always thought I'd retire and become a teacher

Him sitting behind a backwards desk is appropriate for a man who fellates microphones

The Bellman's Rule:

What I tell you three times is true

-- Lewis Carroll

#RM3 #NonBinaryLogic

This is what forced binary choice looks like

me: "cylinders are flat"
It's another freshman's dream

ChatGPT: To a freshman, who might be learning about Euclidean geometry, the phrase is obviously false. A cylinder is clearly curved, right? But to a graduate student, familiar with concepts like Gaussian curvature, it's an intriguing truth: a cylinder is "flat" because its Gaussian curvature is zero, just like a plane.

Robert Rosen talked a lot about the distinction between simple and complex systems. Essentially, simple systems can be built from parts, like a clock. Complex objects, on the other hand, are things with many models. Like time, or music, or a plant. No single description is enough.

He went further, saying that when studying a complex system, one can "throw away the matter and study the organization" to learn those things that are essential.

He was talking about Category Theory and its use in studying the organization of organisms.

A complex system, you might say, is not decomposable into parts. In this manner the complex :: simple distinction is akin to the difference in monoidal :: Cartesian tensor products in a monoidal closed category. Cartesian products have projections, you can take them apart. A monoidal product does not. You can only make inferences about their structure (make models)

#robertrosen #categorytheory #RM3 #complexsystems

By the way, root canals, if done properly, are completely painless.
Until it's over and the guy says "That'll be $1000"

When you go to a psychologist for help and he has a wall screen TV showing stock prices in his office
This is like my dentist revving his Lamborghini after giving me a root canal

https://kurage.nimh.nih.gov/tomh/guru_of_news.txt

I don't know who wrote this. But I laughed so hard I choked

The Guru of News
or
The Guru of Gnus

Zorphwar

Twenty more successful missions,
and I move up to Sector Commander!

https://kurage.nimh.nih.gov/tomh/zorphwar.txt

https://youtu.be/_cAl3FlQZ9Q

yes, change society for the better, and use pentagons

https://youtu.be/W3I3kAg2J7w

Grant Sanderson being inspirational

https://youtu.be/CTQcLi6SpX8 2022 Abel Prize talk

RIP Jim Simons

https://www.youtube.com/watch?v=MZiEcx0F_CM Relying on geoengineering is "in a way, madness, but the situation is already quite mad."

https://youtu.be/pQsdygaYcE4

P vs. NP from Quanta Magazine

It is sad that when they talk about Boolean Algebra, they just ASSUME it's binary. That's wrong. Everywhere you look in the media, the scientific, and even philosophical literature, you see this mistake. Assuming Boolean Algebra is binary, or that "Logic" equals "Binary Logic" implicitly because nobody even considers that there IS a (very important) alternative.

Instead of Z2, use the field F4. Boolean algebra (which is about polynomials over a field, not binary numbers) works just fine, and you can easily construct coherent 4 valued and 3 valued logics from that.

When I speak of "absolute truth" nowadays, I must do so using a paraconsistent logical system. That guarantees correctness, but the result can still be inconsistent. Which is to say, sometimes, given the available data, that's all you can conclude, or in even other words,

One of the hallmarks of a good scientist is the ability to say "I don't know".