3 ms·

I really dislike these types of papers. "Pfft look at these neanderthals studying complex system X by observing surface properties and trying to make deep infer

I really dislike these types of papers. "Pfft look at these neanderthals studying complex system X by observing surface properties and trying to make deep inferences! Nothing like us mighty computer spacemen!" A circuit is a highly systemic, regular design which can be precisely studied in isolation. It is not so easy to carve up a human brain, cell, or metabolic process so that one can study this thing hidden to the human eye. The only tools available require working within artificial constraints to have any hope of picking up trends. Biology is incredibly messy full of overlapping systems, sometimes operating at cross purposes.

civilitty 3y ago
> It is not so easy to carve up a human brain, cell, or metabolic process so that one can study this thing hidden to the human eye. It's basically impossible when studying biological systems. Can you imagine trying to understand human digestion in isolation from the gut microbiome? Even if you had perfect information about what the digestive system does, you'd still be completely missing half of the big picture.
- Solvency 3y ago
  Same goes for nearly all cellular processes. Metabolism is virtually impossible to define as a series of discrete elegant deterministic procedural steps and properties. Same goes for insulin, and all other hormones. You can't even try to think of something like insulin as massive polynomial function where you might eventually understand all of the variables involved and how they affect said function. In fact the function could completely change at any time depending on other esoteric confounding factors in the body. It's absolutely insane. In fact almost all biological functions in the body can be described as trillions of chemicals and chemical pathways interacting in infinitely complicated ways that just so happen to occasionally result in enough homeostasis to support the macroscopic survival of said organism because of a sloppy brute force evolutionary system that took centuries to unfold.
geysersam 3y ago
But is that the point of the perspective? My understanding is that they're asking the question: are the tools we apply to the task able to solve another related but much simpler task? If the answer is no then we should probably take another look at our approach.

civilitty 3y ago

> It is not so easy to carve up a human brain, cell, or metabolic process so that one can study this thing hidden to the human eye. It's basically impossible when studying biological systems. Can you imagine trying to understand human digestion in isolation from the gut microbiome? Even if you had perfect information about what the digestive system does, you'd still be completely missing half of the big picture.

Solvency 3y ago
Same goes for nearly all cellular processes. Metabolism is virtually impossible to define as a series of discrete elegant deterministic procedural steps and properties. Same goes for insulin, and all other hormones. You can't even try to think of something like insulin as massive polynomial function where you might eventually understand all of the variables involved and how they affect said function. In fact the function could completely change at any time depending on other esoteric confounding factors in the body. It's absolutely insane. In fact almost all biological functions in the body can be described as trillions of chemicals and chemical pathways interacting in infinitely complicated ways that just so happen to occasionally result in enough homeostasis to support the macroscopic survival of said organism because of a sloppy brute force evolutionary system that took centuries to unfold.

geysersam 3y ago

But is that the point of the perspective? My understanding is that they're asking the question: are the tools we apply to the task able to solve another related but much simpler task? If the answer is no then we should probably take another look at our approach.