scotradamus

Energy production really isn't the issue. Energy storage is.

We produce way more energy than we use. We currently do not have an easy way to store it. So why generate more than we need? Well, we need to have power available to prevent brown outs, especially right after work when everyone gets home and flips their light switch (so to speak). Second; it is actually more efficient to keep the generators fully spun up.

This most certainly is not fear mongering. Helium is a non renewable resource. We used to discuss this all the time in the laboratory I used to work at.

I wouldn't so casually dismiss BCS theory. BCS contains one of only a few exactly solvable wave functions. It's used in everything from superconductivity to neutron stars. Actually, anytime there is a fermi sea instability BCS is a good starting framework. It's an triumph of modern physics.

1800 dollars for a 5+% salary bump, a new work you're interested in, and perhaps a shorter commute. Seems worth it.

See my comment above. Image processing alone has saved millions of dollars in engineering hours alone.

I dont really agree with your comment. DL is used heavily in the semi industry for image segmentation, defect classification, simulation speed cut-down, and plasma optimization. It's been a game changer, saving engineers up to 30% of their time.

Then google would buy all those tickets at the reduced value and resell them at the normal price.

Worked as a physicist at LANL on energy related issues. Biggest problem is not energy production, it is energy storage.

OK, I'll add a bit to my previous comment. Is there a limit to how small a mammal can be? If so, what sets the scale?

TL;DR – effective pumping of blood requires precise allowable branching network of capillaries in the circulatory system. At a certain size this network goes from AC to DC and sets the scale for metabolic rate and lifespan in all animals with a circulatory system.

The scale is set by impendance matching of the circulatory system. The heart pumps blood. It is AC. To prevent the AC wave of blood from reflecting back when the capillaries branch out and get smaller, the branching network of the capillaries requires the cross-sectional area of the mother branch to be equal to the sum of the cross-sectional areas of the daughter branches. But when capillaries get too small the heart loses this AC advantage and goes from AC blood wave to purely DC.

So the smallest mammal a shrew has 2 capillary branches, 1 AC, 1 DC. Humans have 8 branches 6 AC then 2 like the shrew 1AC, 1 DC. A blue whale has 15 branches 7 AC then 8 like a human (6 AC then 2 like the shrew 1AC, 1 DC). This branching network sets the smallest length scale of a mammal (actually for any animal that uses blood and mitochondria as its metabolic energy source).

Going back, a shrew has 2 branches 1AC, 1DC. A shrew is 4cm in length and its heart must beat 20 times a second, as nearly all the hearts energy for all animals is expended to push blood through the last small DC branch portion of the capillaries. The shrew must be only 4cm and pump blood with the same pressure and speed as a human (which is the same speed and pressure as a blue whale) through the shrew’s tiny little heart. That is, smaller the animal the larger the metabolic rate. The larger the metabolic rate, the more wear and tear and repair on cells, the shorter the life span.

It's true, and the microscopic theory was done by the same physicist who found the macroscopic relationship, Geoffrey West (one of my personal heroes an current director of the Santa Fe Institute). The theory also explains the upper and lower bounds for the possible size of an animal. If you want to read about it he wrote an amazing book called Scale. I _highly_ recommend it.

"Science progresses one funeral at a time" -Max Planck

A ton of great innovation comes from Universities and the National Labs. I would argue most innovation comes out of basic research funded by the government. Interestingly, in a lot of ways this research is subsidized R&D for American corporations and startups. Much of what gets funded in basic research is set by lobbyists. It makes sense, not many companies can sink millions or in some cases billions of dollars into R&D that may never have commercial applications. There was a really good book review about this in a Physics Today issue a few years back (I cant currently find it).

You're forgetting the two most important; energy storage and transport.

This is interesting, and hadn't thought about the cashing out of index funds with age. Given that idea though, the money IS being spent, put back into the economy.

As a data scientist that heavily uses DL in my day-to-day, yes you need to know math. We are having difficulty hiring, and few candidates can answer basic questions: what is an FFT, what is the central limit theorem, what is an eigen value, what does the trace of a matrix tell you, etc. Many candidates use ML but have no idea what they are applying to the data.

Dazzling managers by multiplying matrices together and calling it AI.