bsdpufferfish

I don't blame you if you aren't entertained. But this analysis is like thinking Moby dick is about hunting whales.

you know what else sucks? living in a low trust society where people game these return systems (which we also pay for).

No disagreement.

People don't typically start a Phd at 34. You're at a different level of maturity now.

"have no hope" actually means someone at the department refuses to sign your papers. I have never been treated with so little respect in industry.

Research in a Phd program is 90% long term habit, 10% creative musing. Marriage structure helps with that 90% part.

Why are they always trying to sell these things to audiences who are not interested?

higher dimensional vectors or matrices are still not tensors.

> property rights

No you are conflating two topics. Property rights is not propping up stocks and bonds. It's protecting ownership so you can invest in building things over the long term without them being taken.

Similar reasons encourage the government to protect the value of their currency.

> the point of fiat is to be an easy medium of exchange

The stated goal of the federal reserve is price stability, not exchange.

> Hoarding paper or shiny objects is still speculation with risk

Yep but it's a difference of how much risk.

> having strong social safety nets

Those carry their own kinds of risk.

> I shouldn't be incentivized to hold dollars.

0% is hardly a massive incentive to hold cash. If we can't agree on that then ill be incentivized to hold a currency that does care about its value.

Of course there is always risk. But one of the supposed goals of fiat money and property rights is to reduce that risk, ensuring the ability to save into the future.

There are countless examples of countries where that trust was broken.

> not prerequisites

Almost the entire document is undergrad numerical topics (finite difference methods, maximum likelihood, Jacobins, hessians, newtons method, etc). This is all well covered material that is soundly not research.

> what's stopping them from keeping the money in an index fund

Risk of principal loss.

The comment is referring to playing loose with the currency, not a drop in supply or an increase in demand.

It's not good for society that ordinary people have to speculate their hard earned money just to keep it for retirement.

Even if it's good for GDP to have everyone spending every penny, it's not good for you.

Relatively constant demand, and difficult to increase supply.

Yep, societies just don't function until you establish a central bank with fiat currency and fractional reserve banking. It's a law of human nature.

fiscal hawks, like who? and with what power?

its really not. the math is a distant abstraction of a social subject

Differentiable programming is hardly a "subfield" it can be explained in a paragraph if you know calculus well. If there is any subfield, it's in researching specific compiler optimizations.

sales costs money

The most interesting part of this is the comments about software shifting from a normal career to a prestige target for wealthy families, and that this demographic shift has massive consequences on technology design and services.

> you know better than the query planner, for now until eternity,

Nope, you just have to know it's fixing a real problem today.

Having a query regress in performance below a KPI would be worse than not taking advantage of a further optimization in the future, due to out of date hint.

> Good engineering means having the intellectual curiosity to exhaust these possibility before resorting to hints

Why is that better?

Luckily we don't have to rely on such grandiose claims. Just try it out. If you find a query that you can tune better than the planner for your data set, then it's a better outcome.

What if I have a performance problem I need to fix now?

Additional reasons:

- there is less alternative hardware I want to use. I want Apple Silicon processors, materials, and there just isn't much high quality competition.

- Because of inflation the Apple premium isn't as high as it used to be. You get a Mac mini or MacBook Air at very competitive prices (RAM is still painful).

- Linux Desktop software is more competitive and fills some of the roles that needed macOS before.

> Free market hardly work there.

This is hardly a free market. Libraries receive a budget to spend _ dollars on books every year. This entire publishing industry exists to capture that cashflow, because consumers aren't buying it with their own money.

The university version is more complex, but the bottom line is you do not have consumers spending their money on products they use.

This is like comparing an oscilloscope to the iPhone 15 because they both have a screen.

lambda calculus is one way to describe computable functions. this course chose to focus on Turing machines.

category theory is historically not important to theory of computation, and it has questionable use if you aren't familiar with abstract algebra

Swift can be written in many ways with different performance tradeoffs. The slower ways looks like Objective-C or functional programming. The faster ways look like ergonomic C or C++.

Why are you interested? I recommend the short book "Godel's proof".

Godel's work is wrapped up in historical context which is both interesting and distracting from the core idea. For example, Bertrand's Russells' work and book isn't really essential, it's just the system which Godel worked in to do his proof.