[youreincorrect]

It's absolutely insane to pretend most software engineers need to do any of these things. We're making billions of dollars in aggregate without knowing this stuff. Most people aren't writing their own data compression algorithms, and we sure as shit aren't writing cryptographic algorithms because we're told to leave that one to the experts (i.e., mathematicians).

I'm guessing 80% don't even know bit-field flags and never have to use them, despite them being relatively simple. It would also take someone moderately capable at understanding math less than an hour to "learn." I learned them when I was 13 because I fucked around with MUD codebases, and I don't think I am special for it.

[feoren]

Do you realize that once you write code to solve a problem, that exact problem never needs to be solved again? Either you're solving new problems (and "new" can be slight -- new situation, new context, new hardware, new people, new company, whatever), or you're doing the compiler's job. If most people aren't solving new problems, then their job is bullshit. I frankly don't even understand how you can be confident that code copy-pasted from Stack Overflow actually does what you need it to do without understanding the fundamentals.

> we sure as shit aren't writing cryptographic algorithms because we're told to leave that one to the experts.

Shouldn't we all be striving to be an expert in something? If you're not working your way toward expertise, what are you doing? Why are the absolute fundamental basic building blocks for understanding how computers work and what programming languages are doing something that only "they" need to bother to learn?

[youreincorrect]

Most of the problems software engineers are solving today are business problems defined by stakeholders in a business.

And yes, I agree we should be an expert in something. Harping on binary seems like a waste of time, however. I would certainly like that people are interested enough in the field that they spend their time learning as much as they can about CS, but I'm under no illusion that I'm going to automatically be more productive or better as a software engineer because I know bit-fields.

PS: Thank you for downvoting me.

[feoren]

> Harping on binary

We're "harping" on it because it's so basic. There are a hundred equally basic things you should also know.

> Most of the problems software engineers are solving today are business problems defined by stakeholders in a business.

Even understanding the business problems usually requires a lot of fundamental knowledge in a wide variety of subjects. Being a good software engineer is hard.

And regardless of the problem, if the solution is going to be in code, you can't get away from binary. I actually don't think most programmers should be learning assembly language, because you can actually completely abstract away from it (and you should, because you don't know which assembly language you're programming for). But you can't abstract away from addition, the alphabet, binary, strings, algorithm complexity, and other basics.

PS: I didn't downvote you. I don't downvote people for disagreeing with me. I like disagreement, and besides, it would reduce the visibility of my pithy responses! I only downvote comments that aren't even worth arguing with. So the very fact that I'm taking the time to respond means I have at least some respect for your argument.