Hacker News Reader

Comment by dahart | Hacker News Reader

Anecdotally I’ve heard some CS professors rant about how we teach things all wrong to CS students, especially math and CS, by going bottom up and focusing on abstract building blocks and tools before presenting enough context or real-world motivating examples. One told me about some research into math education that flipped the normal curriculum for math students and went in a very different order than what we all learned in school, and the claim was that students were learning faster & better. If true that might be a tad interesting and surprising.

View on HN

I graduated in CS in Italy. I had to take two mandatory physics courses in the late 80s. Those courses were equations-first with basically no real world examples. The teachers were from the Physics department and the courses for us were watered down versions of what they were teaching to their students there. Those courses were boring as hell, all math and nothing else. Luckily I knew what those equations were about thanks to the physics courses at high school: more watered down but firmly grounded in the real world. Probably the aim of those courses, in the first two years of the degree path, was to make people drop out. The statistics were one graduated every three freshmen. The goal of university was to make new scientists and everybody else, please go away we don't have time for you. That changed later on.

Yes these are legitimately real problems in university education, and not always talked about openly: that optimizing for student learning is not necessarily a primary goal, that ‘weeder’ courses exist to dissuade some people from certain majors, and that some departments and some professors aren’t interested in training people for industry, but only for academic research. Some of that was going on in my undergrad program as well, and it has been shifting toward being more industry friendly over time.

Are weeder courses really a problem? Some students are just not capable at that moment to take certain degrees. Maybe it's a lack of preparation from highschool, in which case the student needs to take some remedial fundamentals, or maybe the student just simply isn't interested in the field they got in under, either because of parental pressure or a misunderstanding of what the field actually entails.

Weeder courses signal to the students "Hey, if you're struggling in this class, it's not too late to switch majors". Every student "weeded out" frees up a disproportionate amount of resources for the rest of the students who are ready and willing to learn.

There’s a valid case for them, I wasn’t claiming they don’t serve a purpose from the department or university perspective.

But sometimes they are a real problem from the student’s perspective, especially when the class is made artificially difficult to pass, and this does happen. They are not always used to prune students who can’t handle a degree, they are sometimes used to limit enrollment in a program that doesn’t have enough capacity, which isn’t an issue with student preparation or capabilities.

I witnessed a situation when "surplus" physicists employed by the university were looking for some better justification of their employment and lobbied hard for extensive mandatory physics courses for maths and informatics majors. In the extent of, say, 20 % of the entire curriculum.

Their proposals were beaten back, but it took a lot of effort.

Czechia, 1998-1999. No tuition involved, study was still completely financed by the taxpayer.

On some level it does seem reasonable for Universities to pursue their own goals, and let industry find or sponsor some means of training their own workforce

I will say that I never understood math until I took a computer maths course and I never understood computers until I took a languages and os class. Seeing how math can be used to prove algorithms, how functions are just functions, and how math is just a language to describe actions and laws made it make sense to me. Same with seeing how computer languages are just ways to abstractly think of the actions and data a computer can have.

I absolutely love computer graphics for learning math and programming, it ties so many things together. Linear algebra, for example, becomes so tangible with 3d transforms.

As someone who didn't understand math until I messed with programming, I absolutely agree academia goes about teaching math completely wrong.

The absolute lack of context and meaning to anything that was done in math education meant I never appreciated them, they were literally just wastes of time to me. Anything beyond simple arithmetic made no sense to me, because I was never taught how they made sense.

It was only once I started playing around with programming when everything suddenly clicked. All at once, concepts like variables and equations had meaning, I could sincerely understand what the numbers and letters stood for.

Math education in general really would benefit students of all ages more if they had more of the Hows and Whys before engaging in all the Whats.

It's a problem across all subjects.

Out of necessity, academia historically selected for students who could learn from being read to just once without much context, and there was little pressure to fundamentally change the education format. It's very easy to better serve most modern students by actually designing the program for those students for a change.

Debates over this get muddled by zero-sum concerns. E.g. catering to the average student hurts the top performers; putting applications first replaces rigorous education with training.

That seems to be a problem with math in every context, from child schooling to wikipedia articles and specialized books.

Reply on news.ycombinator.com