Hacker News Reader

Comment by tasey | Hacker News Reader

Wow, people saying that it is "really simple". I feel dumb.

I think that if you have just passed your undergrad degree, these questions are indeed not _that_ hard, because the stuff is still quite fresh in your mind.

If you have left university for some time, chances are you did not do any math in that time inbetween, if you are a software engineer. Software engineering requires logical thinking in a different way than mathematics, so we do not really keep our mathematics skills up to date.

Looking at document, I can remember touching the necessary mathematics to solve this, but not having done any mathematics of the kind makes me too rusty in mathematics to be able to solve it.

With a bit of revision, someone having seen this somewhere in undergrad will probably be able to solve it.

An undergraduate degree in maths, sure. But if your degree is in computer science, I think it's likely you'd never come across any of this during the course. Some universities may teach it, but there are probably plenty that don't.

Any self respecting university that teaches a formal computer science degree would teach this level of mathematics. Mine went beyond and happily forced students to learn introductory quantum mechanics as well, which annoyed the CS students to no end.

Same story here, as computer science students we got a hefty dose of mathematics, and was in some respects close to what students of physics and mathematics got.

For example, when compsci students get Linear Algebra in their second year, they are getting the same lectures as Physics students in their first year. (Same room, same exam).

We also had shared physics courses in the first year.

I believe that physics, mathematics, and computer science should have some shared courses, not just having compsci students take mathematics. For example, for programming, we were a joined group of mathematics,physics and compsci students in the first year.

They are all courses related with logical reasoning anyway, so a compsci student should not hate mathematics and physics that much. Most of the complaints I noticed were the physics students complaining about _programming_ rather than the compsci students complaining about physics.

I agree; my undergraduate course (not in the US) wasn't even called Computer Science, it was "Software Engineering", but we definitively learned enough math to pass that test easily after the second year.

MIT, for example, doesn't require their CS students to take a differential equations class. So you're wrong in that respect. The rest is certainly doable though.

There's nothing in this exam that requires specific diff eq knowledge.

Though if you've taken a course on variational calculus then question 2 is completely routine and you're at a great advantage relative to someone who hasn't taken such a course.

Doesn't question 2, problem 2?

No? That's calc 2 level problem, not a diff eq one. Partial derivatives are hardly diff eq territory.

Sorry, I'm not american. What does calc 2 normally refer to in the US?

I'd love to know where you went to school and earned an ABET accredited CS degree (assuming US) and didn't take calculus, linear algebra, or have any coursework in probability and statistics.

I studied in the UK. I think our systems are rather different because a lot of that stuff is taught in secondary school, so you're expected to know it before starting uni.

Which is all well and good if you stay on at secondary school (or attend a sixth form college) until you're 18, then go straight on to uni. But that doesn't apply to everyone (some people leave school at 16, or take a break before going to uni). So I can imagine that a lot of CS graduates here in the UK would have studied these aspects of maths quite some time back (and therefore can't remember them very well) or not have studied them at all.

That makes sense, well, that it happened, at least. Typically those topics are covered in US schools before university as well, just they are covered in more depth and rigor again in university and with the applications of those topics to CS as well.

In the US we have this body called ABET which helps define a standard engineering curriculm.

One of my close friends studied in the UK. University of bedfordshire, which was also a computer science degree. Compared to me (Belgian university), he indeed barely got any mathematics, but I always thought that was just because of the university.

Sure, you might not have studied differential equations though.

At my university, all of this content was part of the CS education except for maybe the eigenvalue stuff depending on which courses you picked.

I don't see anyone saying they are "really simple". But they are indeed at the expected level of knowledge for somebody who has had completed few advanced math courses. I did my bachelor in electronics engineering and all these problems look familiar and could probably solve them after refreshing my dusty math skills.

Reply on news.ycombinator.com