I actually agree with both you and the person you are replying to.

You can learn in both academia and industry, but you learn completely different things. In industry, you typically learn more about business and current "trends" in technology, whereas in school you improve your fundamentals and have the opportunity to learn more hard skills, like machine learning, programming languages, etc.

I'd argue there is a lot more opportunity to learn in school than industry though, if only because you can go at your own rate in school so if you wanted to learn about current technology in school you could easily do so on your own. Still, I think a lot of it really boils down to having responsibilities that you are getting paid for vs not being paid but having the freedom to do what you want.

I certainly have learned a lot since leaving school, but the majority of it has not been due to being in industry but learned on my own time.

It may sound horribly depressing, but it rings true to my own experiences. That said, I think it also depends on what you consider to be "learning."

Most engineering jobs require a lot of horizontal learning: develop enough knowledge of a topic to do something useful for the company then move on. While academic learning revolves around developing depth in a particular subject.

If you'd like to learn enough about a topic to contribute something truly new to the field, 99.9% of engineering jobs will not satisfy you.

The way I see it is like this:

So-called "programming" jobs can be science, engineering, a trade, or a completely different job masquerading as "programming."

Science is what tends to happen in research divisions and academia. These jobs revolve around the invention of novel algorithms, research into theory, and so on, sometimes without even having a practical application in mind. The work product is often papers, but is also sometimes new languages or proofs of concept.

Software engineering jobs tend to be what you see in "senior developer" positions in industry. They're about applying known concepts (i.e., existing languages, implementations of approaches from the literature, and so on) to build a defined system in sometimes-novel ways. The work output is usually a product or service that you have some degree of responsibility for.

Software as a trade is what most junior programmers and general employees at larger companies do. This is the application of very specific learned skills to implement pre-solved problems (i.e., a ticket to "make X work like Y using Z method"). The final product is closed tickets / implemented features.

And still another large group of "programming" jobs are actually more about knowing the problem than the solution. These are really jobs about using programming to do something else. Process and product optimization as well as a lot of engineering support jobs fall into this category. The work product is an improved process or system that happens to use a computer.

You can learn in any of these jobs. In the science job, you learn the fundamentals and then think of advancements to the field. In the engineering job, you study the fundamentals to apply them to solving a problem. In the trade job, you build the biggest possible toolbox of code to apply to each solution you're given to implement. And in the non-programming job you learn a whole new field and then apply code to it.

All of these jobs can be lucrative, fulfilling, educational, and fun, but you need to know which one you're getting into ahead of time. "Programming" and "industry setting" are awfully vague and useless terms.

Most work is what you make of it but don't fool yourself into thinking that you are going to learn and grow at an optimum rate in an industrial setting. Your emotional reaction to the reality of the situation does not change the fact that learning is maximized in an academic setting.

With what end goal? I could also never leave a library, ever, and maximize my learning that way. And I'd die alone and inexperienced. Of course there's some give and take...

I trade my "pure learning" opportunities to be able to still grow and develop while earning an income that empowers me to live my life as I desire. I get to solve real world problems and challenges, rather than open ended ones.

I'm not trying to take anything from academics, but the most useful result will always come from a combination of academia and business. Neither would exist without the other, and there's and endless amount to learn from either side.

Any end goal. The fact that there are no distractions to purely pursue a specific goal is what makes an academic setting special. There are no managers and business specialists hovering over every single decision being made.

That doesn't jive with everything I've read - some of which here on HN - about universities. Grants, tenure, cost-cutting by administrators, ego, are all said to be sources of politics and distractions. And teaching, of course. Are they wrong?

They are not wrong. If I were to go back then the situation for me would be quite different. If your intent is to go for a tenure track position then you have to put up with all sorts of stuff but if your goal is to learn and write a thesis with the intent of going back to industry then you'll have a much easier time. The usual teaching and class load is easy to handle and you don't have to worry about grants and all the other stuff especially if you have a bit of savings.

The difference is between maximizing tenure-track job prospects vs learning. Surprisingly those two goals can sometimes be at odds and that is usually what you hear about on HN. People that were trying to maximize for both learning and future job prospects inadvertently running up against academic bureaucracy. If you have a clear goal then you'll have a much easier time.

This is laughably inaccurate. There are a ton of politics in an academic setting. Pressure to publish, to get grants and raise funding, grades and the financial pressures of being a grad student. You make it sound like people in school have unlimited time to sit back and think about big problems. It just replaces quarterly profits and management goals with conference paper deadlines and academic departmental goals.