Hacker News Reader

Comment by czhu12 | Hacker News Reader

I also was at UBC for that class and left with a totally different impression. I thought it was awful and set up introductory students very poorly for subsequent CS classes, especially data structures and algorithms that were taught in imperative languages.

When I TAed, a lot of incoming 2nd year students didn’t know how to do for loops yet, which really set them back for basic algorithms.

I had a much deeper appreciation for functional languages after theory of computation classes in my senior / 4th year.

View on HN

I also TA'd 110 and I firmly disagree. CPSC110 teaches you to view a particular problem as a series of different possible states, and solving for those states (while placing an emphasis on seeing base cases and working up from there). Students learn about data structures like binary trees by the fifth week, and the ninth week they're already able to solve sudoku puzzles using generative recursion. We even touched on the n-queens problem towards the end of the course.

Racket serves its purpose well as a simple and fast to learn educational language; it's easy to see and understand recursion. It's also easy to see what the execution order of statements in your program.

I will say though that some problem sets were a bit brutal in terms of time taken to complete them.

Agreed. My main observations from students in my cohort were that those who viewed programming as a set of instructions struggled, whereas those who viewed the course material as a series of manipulations on mathematical objects grasped the material much quicker. In general, the course focused a lot on learning from first principles, which I appreciate much more nowadays.

Most data structures taught with imperative languages are just that, data structures for imperative languages, which run on a single CPU core and do not utilize multiple cores well, or at least not without locking mechanisms. One can usually find one of them and copy paste the code, when needed. Much more interesting are purely functional data structures, that can be easily parallelized. That is what universities should teach. I feel like we are quite backward in education with data structures. I can watch a whole data structures lecture/class at no cost besides my own time on YouTube, for famous professors, but when I want to apply it in an FP setting, I will still be dumbfounded and looking for other solutions, which I do not know how to arrive at.

"When I TAed, a lot of incoming 2nd year students didn’t know how to do for loops yet, which really set them back for basic algorithms."

They hadn't done simple recursive functions over lists? What had they been doing?

Been a while since I had to read introductory material for Racket (or any similar language), but as I recall functions with a base case exit and recursion is usually among the first things they teach.

Lisp variants only really have linked lists. So yes you can do recursion over them, but for someone who has never done a C-like language before, the arrays and for loops will take getting used to

Lisp "variants" have lists, hash tables, multi-dimensional arrays, strings, bit sequences, streams, etc.

Is it the case that all the cons cells at the base of the language are compiled into more efficient structures?

cons cells are not the base for vectors, strings and a bunch of other datastructures. Cons cells are mostly only used for nested lists or similar.

I guess I'm curious how the implementation of Vector works in Lisps. I thought it would be some sort of combination of Atoms and Cons cells.

I looked into the Racket Repo to find something, but had a hard time finding the code.

In Racket, a vector is a fixed-size contiguous array of memory. The values in each slot of a vector are tagged pointers so values like fixnums don't require any indirection to reach (i.e. they're unboxed). Racket also has specialized vector types for storing fixnums (fxvector), floats (flvector), strings (string) and bytes (bytes).

Other Schemes and Common Lisp have similar constructs. Just because a language is a lisp doesn't mean its primitives have to be implemented using cons cells.

As an example of other Schemes, here are the details of the representation used by Larceny (a Scheme):

https://www.khoury.northeastern.edu/home/lth/larceny/notes/n...

In a traditional-style Lisp which has atoms and cons cells, "atom" is not a specific type. It is a category which means "not a cons cell". Everything that is not a cons cell is an atom: symbol, string, number, character, vector, I/O stream, function, class object, ... Notice vector in there.

The purpose of a vector in Lisp is to provide contiguous storage with integer index addressing and O(1) access time.

Cons cells are typically used to provide lists with O(n) access time.

No they don't. The basic data structures in Racket are listed here, https://docs.racket-lang.org/reference/data.html . There are also specific array implementations in modules, like the generic array interface in the array module or arrays for math in a math module.

Recursing over a list is a way to learn how to implement for, while and friends. If you know this technique understanding for is just understanding a subset of what you are already familiar with. It can be used to iterate over an arrays as well as lists, e.g. with ranges: https://docs.racket-lang.org/reference/pairs.html#%28def._%2...

Racket can also be used to teach object oriented programming and programming with structs, if the aim is to teach patterns used in C-like languages generally it's not a bad fit. Well, except advanced stuff like pointer witchery. Though you could probably implement a teaching language that does it with arrays or the byte code directly if you wanted to. It might be a good way to improve on error messages for pedagogical purposes.

Only if stuck in 1950's Lisp.

Racket has a ton of different types of for loops for iterating over data structures etc. and collecting the results (or ignoring them). They're implemented under the hood using recursion but that's hidden from the user and it should be trivial to go from then to other languages. Maybe this class they're talking about didn't cover them?

https://docs.racket-lang.org/guide/for.html

it was indexing and mutating arrays / hashmaps that students had trouble with. Certainly a very different programming model than what’s presented in racket, with recursive iterations.

Of course with years of experience it’s easy to interpolate one from the other but these are students who’ve had 4 months of exposure to programming that have to jump from functional languages into operating systems and data structures and algorithms, all of which are taught with imperative languages

I started with BASIC on C64-/VIC-style machines, then went on to dabble in C, C++, Java, Visual Basic, Delphi Pascal, PHP, before I came into contact with CL and Emacs. Recursion-iteration-equivalence was a mindblower for young me, suddenly I realised more ways to use it for problem solving in the algolians than I had seen before. If I had gone the other way around I'm sure I would have saved a lot of time.

This is surely a single point of anecdata, but it makes me suspect that it was more about how teaching was done than a 'functional vs. imperative' thing. I also suspect pointers and memory management to have been bigger hurdles than how to format code for iteration, unless the Racket course introduced techniques like quoting.

> When I TAed, a lot of incoming 2nd year students didn’t know how to do for loops yet, which really set them back for basic algorithms.

I'm surprised anew on the rare occasions I come across statements like this. The first time was when I applied for my first programming elective class. The teacher actively tried to dissuade me from it, saying that by the end of term they'd be lucky if they even got to for-next or do-while loops. I was shocked, having first understood these from BASIC code listings in the back of magazines, with only one or two weeks per year having a few hours of access to an Apple II on which to actually run anything.

I tend to fall into the trap of feeling that loop constructs should be so trivial, that understanding should come from a tiny fraction of one class session plus maybe a few minutes tracing the code by hand. So, I thank you for being my irregularly scheduled reminder that people's first exposures to such things can be very different.

SICP doesn't teach assignment till chapter 3 if I recall correctly, and that was used for years at MIT. I don't think it set CS back there, so maybe it was more a teaching issue?

Reply on news.ycombinator.com