Quicksort is the new Hello World (opens in new tab)

    qsort :: Ord a => [a] -> [a]
    qsort []     = []
    qsort (p:xs) = qsort lesser ++ [p] ++ qsort greater
        where
                lesser = filter (< p) xs
                greater = filter (>= p) xs

I know nothing about Haskell but I don't think this code implements the original quicksort algorithm which sorts the input in-place. Moreover the two-pass of filter over the list and the concatenations cause unnecessary overhead. Thus, even if the sample code is simple and elegant, a real-world implementation would surely be a bit longer than the given example.

Yet another article that completely misses the point. The point of "Hello World" is to show a noob how to fire up the editor, compile, and see something happen.

And btw, here's the answer in F#:

  let rec qsort = function
       | [] -> []
       | x::xs -> let smaller,larger = List.partition (fun y -> y<=x) xs
                  qsort smaller @ [x] @ qsort larger

tl; dr, but a quick note: the presented algorithm takes the first element as the pivot element p:

   qsort (p:xs) = ...

This is not recommended as it results in worst-case behavior on sorted input lists. (Another commentor correctly pointed out that it requires extra memory for the intermediate lists, too.)

I think your point is valid and of course interesting for programmers, but KR's "Hello, world!" is still relevant for students and novices. Maybe reading/writing on a file would be more useful nowadays, since it does not demand knowledge of algorithms.

For a veteran programmer learning a new language like Haskell, I agree completely.

For a novice programmer learning their first language, they probably don't understand the quicksort algo and probably have no concept of arrays or any data structure. Visually displaying "Hello World" is very simple and teaches you two basic parts of a language: - Syntax - Printing stuff

On another note, someone learning HTML for the first time will have no reason to do quicksort.

Although this version is very simple, concatenating and creating lists at runtime ruins performance (and it takes extra memory). Please make it a good example by doing a version of Quicksort with in-place partition.

It is like fizzbuzz/binary search in that most people think they can write it and they get it wrong.

Quicksort requires careful selection of pivots, and is unstable. (Many scripting languages (perl, python) are opting for stable sorts)

Check out Bentley&McIllroy's Engineering Quick Sort for a guide about the problems implementing a production ready quicksort.

If you're going to teach them something easy, simple and relatively hard to implement badly, teach them merge sort.

Then teach them adaptive merge sort.

I usually implement a tic-tac-toe game as my first project when learning a new language. Tic-tac-toe, properly implemented, will give you a pretty good tour of a language. But with that said, Hello world, still serves it purpose.

"For every complex problem, there is an answer that is clear, simple-- and wrong."

This article is misguided.

Hello World serves two (and only two!) purposes:

    1. It's traditional
    2. It lets you check everything is working

It isn't a way to evaluate or compare languages. How could it be? It only uses a very small fraction of the language. Note that the Hello World for C is rather long and inelegant. But C itself is definitely an elegant language. So its predictive power is poor.

QuickSort is neither traditional nor something a beginning programmer could use to check everything is working. Therefore it is not a substitute for Hello World.

- - -

If you want to get a feel for a language with one code snippet, allow me to introduce the Trabb-Pardo Knuth algorithm:

    In their 1977 work "The Early Development of Programming Languages",
    Trabb Pardo and Knuth introduced a trivial program which involved
    arrays, indexing, mathematical functions, subroutines, I/O,
    conditionals and iteration. They then wrote implementations of the
    algorithm in several early programming languages to show how such
    concepts were expressed.

http://en.wikipedia.org/wiki/Trabb_Pardo%E2%80%93Knuth_algor...

    ask for 11 numbers to be read into a sequence S
    reverse sequence S
    for each item in sequence S
        do an operation
        if result overflows
            alert user
        else
            print result

Hello world is valid for checking that you have everything installed properly and it's the fastest way to make sure you get the most basic syntax of the program.

It's not like the first thing you're going to do is write out all of quicksort to see if it's easy. You're going to fiddle with numbers an arrays a bit first, particularly if your language has some kind of interactive console. In fact, I wouldn't be surprised if the first thing you do in the language is actually 1+2.

There's quite a bit of syntax you need to learn before quicksort becomes a relevant test.

If you're looking for a good language to build a quick website backend in you probably don't even care about that, you'll care more about having nice ways of passing information around and checking that it has a decent inbuilt sort method.

I don't think "Hello World" will ever cease to be useful for the simple task of making sure everything's installed and operating correctly. Quicksort is certainly beautifully concise in Haskell, but it's not in all languages, and I want a low probability of errors when I'm just getting things up and running for the first time.

I still write and run Hello World in languages I am well versed in when I start working in a new environment, purely as a sanity check. In many cases it is the simplest starting point to ensure you have a working environment.

As far as familiarizing myself with a new language, I usually write a Hello World, then add some variables, and some functions or whatever is relevant, and just focus on the syntax of the language, or in the case of a platform or framework, take a look at building trivial components.

Not a language, but a platform example: I recently explored ExpressJS (If you're not familiar, its a framework for node.js web apps.) and my workflow was: Make a hello world server, Add a few routes and a static handler, Add some views, Add some middleware, Build it into a basic blog. (Because it's always a blog first, right?)

The whole process was quite quick and gave me a very strong jumping point for ExpressJS. Implementing quicksort or something similar as my first crack would probably have been pretty useless. So for me its all about context; I tend to start at Hello World and quickly grow something simple, but relevant to the language or platform.

I think I'm not alone when saying that the first thing I write in a new language I learn is some sort of project that requires a good understanding of the language's function-call model, of its object model, of a few specifics (strings, regices, data structures, arithmetic [depending on what I find interesting]), and that requires me to use some part of the standard library not readily available or to decompose my code into multiple files (or both). This helps me understand better what I'm doing, what I like about the language and what I don't. Examples of such projects include IRC bots, Lisp Interpreters, twitter clients, etc. (The one exception is Haskell, where I haven't gotten far enough to understand Monads well so I just wrote a completely functional RPN calculator).

I'm proficient in Hello World in 24 languages. And now you say my knowledge is obsolete?

Hello World is there to validate that you can setup, compile and run a program in your new language of choice. Complicating that process by trying to write something non-trivial defeats the purpose of the exercise. I call BS on this article.

Someone already pointed out that you try out "Hello World" to see if everything is working correctly. Anyway, the author fails to make the distinction between is and ought to be, it's sloppy.

This test gives an extreme advantage to lazy functional languages that they don't enjoy in any realistic context.

Trying to come up with a single test that is good turns out to be surprisingly hard. In modern languages I'm looking for the ability to do clean functional code, stateful object code, async programming, a solid module system, and a type system that doesn't get in my way. I really can't imagine a single 5 line program that would encapsulate all of these areas.

The only way to learn a new programming language is by writing programs in it. The first program to write is the same for all languages:

Print the words

    hello, world

This is a bug hurdle; to leap over it you have to be able to create the program text somewhere, compile it successfully, load it, run it, and find out where your output went. With these mechanical details mastered, everything else is comparatively easy.

The C Programming Language [Kernighan, Ritchie]

After I write my "Hello World!" program I usually graduate to writing a tictactoe game. Tictactoe has enough stuff to learn the basics of most languages.

QuickSort is still a very small problem that is not going to show much about the language.

Gosu's example code does a good job at covering a lot of aspects in ~50 lines of code:

http://gosu-lang.org/intro.shtml

Do we all agree to teach from quicksort, to complete newbies of programming, while most experienced people had hard time to understand? Or simply just read this thread and imagine if you are just 12.

Sadly, many programmers don't understand how quicksort works. This makes it somewhat unsuitable to be a hello-world.

I think something simpler would be a better Hello World equivalent. The first program I generally write in any functional language (usually as a sanity check) is one to calculate the factorial of a number.