snicker7

We are not smarter than our ancestors. Human brain size is down 20% from peak, and is continuing to shrink.

The issue with the textbook market is that the people that pick them (departments, professors) don't actually have to pay for it. If colleges were forced to subsidize even a small % of textbook costs, we'd see an order of magnitude reduction in prices.

https://www.scattered-thoughts.net/ (Jamie Brandon)

How does this work for data involving both a EU citizen and non-EU citizen?

A lot of JS bloat could have been eliminated if it had lisp-like macros.

People intuitively understand the capabilities of cruise control. Can the same be said if FSD?

By "very small minority", do you mean "one of the two only viable political parties in the United States"?

Another option is to use a software-defined abstraction over cloud services. Canonical’s JuJu, for example, allows large chunks of infrastructure to be deployed identically across numerous cloud providers (including private cloud).

Book recommendation lists are about inflating one's ego. I would look uncultured if I recommended the books I actually enjoy and derive value from (or, more poignantly, actually read).

I /highly/ recommend SICP as a first introduction to programming.

The same land is used to make human food. We get the grains, they get the hay.

Cattle mostly eat what humans can’t digest — grasses, not grains. That energy would go to “waste” otherwise. And the water consumption stats often include rain water (“green water”).

This is really cool. Being able to convert Pluto notebooks into a “book” is super useful.

One feature I think will be useful would be the ability to include other notebooks, and still maintain reactivity. It would make “how to build an X” type tutorials more feasible.

Guix configuration is in Lisp, a well-established language (60 years and still strong). Moreover, Guix package code is written in a clear, declarative style. Nix package definitions, on the other hand, typically embed shell scripts in them.

If the COLA adjustments are granular enough (e.g. zip code), I could see this as a civil rights violation. They'd be punishing employees who live in black and latin communities.

Work for the government. I used to work on software that uses AI to catch financial crimes. Super satisfying, but pays a lot less than private sector (which is why I left).

As for array index checking, Julia gets it right. It uses a pair of macros (@boundscheck and @inbounds), which empowers the user elide bounds checking. Moreover, it is extensible to beyond just arrays. I’ve used these macros in my own hand-rolled implementation of a ring buffer.

Hubble is trying real hard now that JWT stole the spotlight.

The pandemic will never be over.

Programming is the easy part. It’s the collaboration and communication that get you. Working part time means a higher ratio of your time is spent communicating and syncing up rather than doing “actual work”.

English isn’t abelian.

I don’t see the point of this post. Julia is a compiled language, Python is interpreted. Comparing them on CPU-intensive algos is unfair.

The problem with these social experiments is that these payments are not guaranteed for life. Being out of work is bad for your career, but if you receive passive income for rest of your life, I think that would change the calculus.

If Netflix increases their price, than so can their competitors. And many people are now subscribed to multiple services instead of just one. In practice, this is not a 2X increase in 10 years — it is 5X.

Julia matrices are laid out as column-major (like Fortan, unlike C). So, it is better to get into the habit of thinking of a matrix more as a vector of columns than a vector of rows.

There a lot of cases in scientific computing where using "glue code" of any sort can kill performance. You miss compile-time optimizations that can perform whole-program analysis.

As a result, despite Julia being very FFI-friendly, the Julia community is trying to move towards a "pure Julia" ecosystem. Sometimes this attitude makes sense (Julia-native BLAS), other times not (Julia has no HTTP/2 packages).

Not even. ROCm does not support any current-generation enterprise cards.

Compare this to Nvidia or even Intel where I can run GPGPU API's (CUDA, oneAPI) on low-end, consumer-grade hardware.

The problem, ultimately, is that AMD does not even care.

Mathematicans read very slowly relative to, say, computer programmers. A page can easily take an hour or more. Bad notation (e.g. variable names) can interrupt one's flow.

AMD consumer cards are unusable for GPGPU. Utter garbage. It’s not just mining. I would not have bought them if it weren’t for their open source driver. I hope Intel Arc fares better.

How does running --tune affect substitutes?

Things moved fast in China (rather Shrnzhen) due to a period of ultra-liberal economics that Xi is now dismantling.