[7thaccount]

I found the article to be hilarious.

It's existence obviously means some professor didn't allow them to use the standard library and they tongue in cheek show how irritating that is.

I'm sure it's possible they made it up, but we had similar restrictions where I could use an Arduino for my engineering Senior Design Project in college, but no Arduino or module libraries - just our own C. We passed, but barely. It also was weird as it didn't match how normal people would solve the problem.

[Jtsummers]

Yeah, this line made it click:

> Now this is starting to look like Professor of Programming Languages code!

A lot of tier 2/3 CS schools with programming language courses (a survey of multiple languages across various paradigms) teach Prolog this way (and I've seen some fantastically terrible examples of Lisp code coming out of those same courses). It's unfortunate because, at these schools, this is often the only formal course where students branch out of C/Java/C++/C#/Python/JS (and they may only get 2-3 of those over the rest of their courses). It leaves the students with a gross misunderstanding of the languages and dread when they see them again.

[yuye]

>I'm sure it's possible they made it up, but we had similar restrictions where I could use an Arduino for my engineering Senior Design Project in college, but no Arduino or module libraries - just our own C.

When I started compsci, it was the first year of a all-new curriculum for CompSci.

Literally the first week: We had to program an Arduino to play a song on a speaker by toggling GPIO, within a week. No assembly or any of that high-level mumbo-jumbo. We had to read the datasheet, look at the instructions and write the hex representation for each instruction in a txt file.

We had a "linker" tool: It took a basic bootloader (containing some helper functions for GPIO), turned our hex-txt file into actual binary, then just copies that to a set address. The bootloader did nothing more than init the GPIO pins and jump into our code.

We were given the locations of where these helper functions lived, nothing else.

It was to give a quick and intimate understanding of how CPUs actually work and why we use compilers and stuff. It worked really well, but it was so much of a pain that they changed it to allow students to use asm a year or two after.

[tengwar2]

<Four Yorkshiremen accent>

You were lucky! We had to design and build the computer out of Z80-family components first before we could play around in binary.

There was actually a reason - this was back in '83/84 in a physics lab, and the idea was to be able to build your own logging / control device. PCs existed, but were terribly expensive.

[seba_dos1]

> It also was weird as it didn't match how normal people would solve the problem.

If I can solve a problem by building from preexisting LEGO blocks then I'll probably do that, but it wouldn't be a valuable learning exercise. Students aren't being given problems in need of effective solution, these problems are tools to facilitate understanding.

What you described could be pointless if it made you work on reimplementing trivial stuff that you'd already mastered long time ago instead of focusing on actual problem. Writing your 100th hashmap implementation or yet another I2C driver isn't particularly valuable. Since you mentioned "barely passing", I don't think that was the case though.

[7thaccount]

It's one of those projects where you never have enough time throughout the year. Coding what we did from scratch wasn't very easy either. I could now probably do the coding in a weekend with the right library code. You learn something either way I guess.

[Blackthorn]

There's nothing wrong with teaching like that. If the class is supposed to teach you something that isn't just "use the stdlib", then it makes sense to ban it.

[7thaccount]

I'm guessing the author is aware and just venting. It doesn't make it less frustrating though. It's almost like someone asking you to dig a hole with chopsticks.