Completely agree. The fact is that most people in STEM aren't so much encouraged to go into STEM as deceived into it by a combination of science documentaries, science types bragging that STEM-like subjects are of Superior Intrinsic Value and Lead You To The Meaning Of Life, and parents selling the field to their children as an easy way to make money.

And a very good point about the hacker/elite snobs pissing on other classmates + gender imbalance and lousy social environment, it's not just limited to CS undergrads but is widespread in Ph. D programs as well, you'd expect that these Head Nerd types would have gotten with the program of life by then.

Combine the shitty social elements with a shitty ROI plus an increased awareness of other kinds of worth than being a coding rockstar, and you've got a bunch of disillusioned, not-very-young people who feel they have been deceived into making their life choices.

The last point is also what's making me reconsider being involved in computer science. It seems like even research universities are increasingly doing work that are more low-quality first-startup style grabs at attention, using machine learning (itself a balkanized field devoid of theory) as a semantic stopsign-marker of objective quality, and less of "classical" computer science results that were very mathematically informed.

> It seems like even research universities are increasingly doing work that are more low-quality first-startup style grabs at attention, using machine learning (itself a balkanized field devoid of theory) as a semantic stopsign-marker of objective quality, and less of "classical" computer science results that were very mathematically informed.

I disagree with this. Much of the product of computer science departments these days is mathematical gibberish. Pages of proofs written by extremely poor writers with nary a point in sight.

The heyday of computer science, in contrast, involved academics working on interesting problems and publishing readable work that was accessible to practitioners. For example, the papers that came out Rice's Scaler Compiler Group in the early 1990s are a model of clarity and good taste in academic writing.

I think we actually agree. I too feel that computer science these days is gibberish relying on the form but not substance of mathematical proofs to convey superior status, whereas "why" systems worked back then didn't require mathemtical _gibberish_ to explain, but were actually very well rooted in mathematical reasoning.

As I edit my thesis, I am observing the sad state of my own technical writing ability. Can you point me to a paper you find clear and in good taste?

I think Preston Brigg's PhD thesis is great: http://www.cs.utexas.edu/~mckinley/380C/lecs/briggs-thesis-1...

This is also great: http://hall.org.ua/halls/wizzard/pdf/Vaughan.Pratt.TDOP.pdf

You should try being a male in neuroscience. Ugh. Women are petty gossips.

I also started as CS, and stuck with it all the way to PhD.

I think that CS, CPE and EE are different than the rest of STEM that students get to start playing in their discipline much, much earlier. I know from experience that CS majors are writing code their first semester. Continually creating new things in your discipline sets up a feedback-loop that keeps you interested. I don't think this was true for engineering programs like civil, mechanical or aerospace.

For science, the payoff is much more abstract, since you don't actually do science as a science undergrad. You learn about science.

There are, of course, undergrads who get involved in research projects or extra-curricular activities centered around their major that allow them to actually play around in their discipline early on. But I think those are edge cases.

Nah, EE's dont. They get divorced theory and math most of the first few years. When we do actually get to build circuits early on, they are to mintutiae-focused on some slew-rate characteristics of an op-amp at a given frequency rather than systemic perspective thinking that is needed in the real world with real applications. To much focus on analysis, nowhere near enough on synthesis. A modern EE education is geared toward training autistic, perspective-deprived drones.

Yes incredibly accurate and its a shame. I did have labs working on breadboards and even IC design but it was literally just following a well defined procedure so you dont screw it up

That is a very good point. I am sure that if all I had done was read about programming and see people scribble on a blackboard about it, I would quit and go be a jazz musician. I know I liked the computer as a teen because unlike chemistry, biology, or physics, I could do really cool stuff without putting together a lab. That's an AMAZING advantage of the computer.

The chemistry student I know has been involved in her university's chemistry research labs since freshman year. Mechanical engineer students here frequently do projects, there is a ME lab for them to mess around with. Biology students have some opportunity (unsure how much) to participate in the university's activities.

But, maybe this University is more hands-on than other places.

I think most schools have those opportunities, but it's something the students have to seek out. I could be completely wrong here, I'm basing this observation off of my understanding of what curriculums I did not go through were like.

As a civil engineer, we were building stuff early on. (and typically destroying it too). I remember towers on the shake table, bridges, and connection testing (including arc welding).

There was still about a 50%washout rate.

Thanks for saying so.

For the record, my CS undergrad had similar attrition, but it was during the dot-com boom. I was a freshmen in '99 and there were 500 other CS freshmen. The typical number of students at a given level is, I think, roughly 1/5 of that now.

I just want to thank you for getting it and not spouting off the regular crap that is regurgitated by certain people who have made it to the level you have (TA or higher).

Recent CS grad (May of this year actually)

Its more than people being lazy. Though of course there is some of that but you hit the major and more important points.

I wish the whole of academia were as enlightened as you are

Thank you for your kind words.

Well, some people are lazy. There's no getting around that. Some people would rather take bong hits or play video games than turn in homework. I don't understand those people. Especially when (as some people have told me) they've done the homework.

I think usually in 'lazy' cases the laziness is because they don't care about the bad grade or the lack of learning. The canonical stereotype is the frat boy with his parents paying his way. Stereotype's priorities are not aligned with seeking a degree. They are aligned with parties or making sure parents are happy. So they have no 'skin in the game', and they are not 'in it to win it'.

I didn't run into very many really and truly lazy people. Most people who didn't turn stuff in (that I talked to - please note the qualifier) simply didn't have priorities aligned with getting the degree with decent grades. Priorities in these cases might be "getting drunk", "arguing with girlfriend", "got sick", "other class had test".

My personal opinion is that Universities in the US are both training and education. We set our sights on education - we the academy - but our clients are often asking for training. I heard a slew of griping by the engineers (EE, ME) in a more theory-based linear algebra course. "Give me numbers so I can stick them in my calculator and crunch them" was the usual desire. When people's priorities are not aligned with learning what you are teaching, you get some poor behavior. My priority was "take hard classes, learn what they teach, suck up the weaker grade". So I sucked it up. I also have a bent for wanting to understand principles, which fits in well in academia.

I think the academy needs to contemplate the idea of "white collar training" and take that as potentially realistic option. Something like a three-year bachelor's degree. I am not sure. I think some CS students could use another year past the four just for code-mashing.

One college I heard about has two options for incoming students. Four year BSCS, or five year MSCS. The extra year adds theory.

> * Math education is generally broken in the US. That means that students are not ready for the rigourous thinking and numeracy demanded in STEM. They struggle understanding the basics of mathematics and computational thinking because they never have had it before, so the 'elite' students who do know what they are doing stand out.

Also even the programs considered good, generally AP or IB, are chock full of bad incentives. Most rankings are based upon pass rates, making the dominant strategy teaching students to perform well on the test by analyzing problem frequencies and released grading guides. This ends up being teaching an algorithm rather than reasoning skills.

This really is the most balanced take on this I've seen yet on HN. Thanks.

A lot of times stories like this are received derisively, and it's like some sort of mark of eternal shame to admit that, yes, the gender balance and social scene in my field is important to me.

Sometimes I feel like hackers (and traditional engineers like EE and MEs) aspire to some kind of neo-valhalla. Instead of eternal fighting and mead-drinking it's coding and soda-drinking. This leads to a field that's incredibly flat and simplistic - aka boring.

Thank you for your compliment.

I have coded since my early teens, so a lot of the code-adoration has faded through the simple wear of time. And, like people seem to as they look towards the end of their twenties, I've made a few mistakes and had a few growing up experiences and those have really broadened my mind about career and goals.

The cyber-valhalla is fun and after a while - for most - the fun palls and you stumble out of the soda hall one day and realize that there's a whole world with unexpected complexity and richness that you've neglected in your pursuit of the code. There's a whole book - Microserfs - about that experience. It's worth reading.

Could you talk more about the difference with MBAs? The side I've always heard said that MBAs don't have that much of an easy time unless they're top 3 program graduates and even then they still have to work their way up corporate ladders for years.

If you're into politics, I don't think there's much stopping you from climbing the ranks of a corporate giant, MBA or not?

Honestly, I don't have a deep understanding of the post-graduate MBA world. Rumor has it that you have to spend a good deal of time actually in the office or on call. I do know, however, that MBAs form the typical growth path for larger corporations.

So (thought experiment, not backed with anecdote or data) if I can get a (say), History degree and a MBA and have a similar track as a CS degree and a MBA... I know which one provides a better ROI in terms of time invested.

From the article: "I was trying to memorize equations..."

That's doing it wrong. I agree with you that math education is probably broken. I imagine that math is very boring if memorizing equations is what you are doing.

Some things you wind up needing to memorize. There's no real way around it. Things like the laws of algebra. Other things simply get ingrained in you by long practice. Still others you can derive by understanding the nature of the system and how to put the math together to solve the problem. If someone doesn't really understand how things work - something that is usually a combination of student + teacher - the simplest way to pass is to memorize. If you aren't interested in the why's of things - many undergraduate engineers aren't and don't have time to be - memorization is a natural response to math.

It turns out you can write tests to check for memorizing tricks though. Given a certain mathematical pattern, a certain trick solves it. That is how I remember the GRE as being. I found it a truly atrocious metric.

I think that doing well in Math and Physics is quite different.