The goal of Yet Another Haskell Tutorial is to provide a complete introduction to the Haskell programming language. It assumes no knowledge of the Haskell language or familiarity with functional programming in general. However, general familiarity with programming concepts (such as algorithms) will be helpful. This is not intended to be an introduction to programming in general; rather, to programming in Haskell. Sufficient familiarity with your operating system and a text editor is also necessary (this report only discusses installation and configuration on Windows and *Nix system; other operating systems may be supported -- consult the documentation of your chosen compiler for more information on installing on other platforms).
What is Haskell?
Haskell is called a lazy, pure functional programming language. It is called lazy because expressions which are not needed to determine the answer to a problem are not evaluated. The opposite of lazy is strict, which is the evaluation strategy of most common programming languages (C, C++, Java, even ML). A strict language is one in which every expression is evaluated, whether the result of its computation is important or not. (This is probably not entirely true as optimizing compilers for strict languages often do what's called "dead code elimination" -- this removes unused expressions from the program.) Haskell is called pure because it does not allow side effects (A side effect is something that affects the "state" of the world. For instance, a function that prints something to the screen is said to be side-effecting, as is a function which affects the value of a global variable.) -- of course, a programming language without side effects would be horribly useless; Haskell uses a system of monads to isolate all impure computations from the rest of the program and perform them in the safe way (see the chapter Monads for a discussion of monads proper or the chapter Io for how to do input/output in a pure language).
Haskell is called a functional language because the evaluation of a program is equivalent to evaluating a function in the pure mathematical sense. This also differs from standard languages (like C and Java) which evaluate a sequence of statements, one after the other (this is termed an imperative language).
The History of Haskell
The history of Haskell is best described using the words of the authors. The following text is quoted from the published version of the Haskell 98 Report:
In September of 1987 a meeting was held at the conference on Functional Programming Languages and Computer Architecture (FPCA '87) in Portland, Oregon, to discuss an unfortunate situation in the functional programming community: there had come into being more than a dozen non-strict, purely functional programming languages, all similar in expressive power and semantic underpinnings. There was a strong consensus at this meeting that more widespread use of this class of functional languages was being hampered by the lack of a common language. It was decided that a committee should be formed to design such a language, providing faster communication of new ideas, a stable foundation for real applications development, and a vehicle through which others would be encouraged to use functional languages. This document describes the result of that committee's efforts: a purely functional programming language called Haskell, named after the logician Haskell B. Curry whose work provides the logical basis for much of ours.
The committee's primary goal was to design a language that satisfied these constraints:
- It should be suitable for teaching, research, and applications, including building large systems.
- It should be completely described via the publication of a formal syntax and semantics.
- It should be freely available. Anyone should be permitted to implement the language and distribute it to whomever they please.
- It should be based on ideas that enjoy a wide consensus.
- It should reduce unnecessary diversity in functional programming languages.
The committee intended that Haskell would serve as a basis for future research in language design, and hoped that extensions or variants of the language would appear, incorporating experimental features.
Haskell has indeed evolved continuously since its original publication. By the middle of 1997, there had been four iterations of the language design (the latest at that point being Haskell 1.4). At the 1997 Haskell Workshop in Amsterdam, it was decided that a stable variant of Haskell was needed; this stable language is the subject of this Report, and is called "Haskell 98".
Haskell 98 was conceived as a relatively minor tidy-up of Haskell 1.4, making some simplifications, and removing some pitfalls for the unwary. It is intended to be a "stable" language in the sense that the implementors are committed to supporting Haskell 98 exactly as specified, for the foreseeable future.
The original Haskell Report covered only the language, together with a standard library called the Prelude. By the time Haskell 98 was stabilized, it had become clear that many programs need access to a larger set of library functions (notably concerning input/output and simple interaction with the operating system). If these program were to be portable, a set of libraries would have to be standardized too. A separate effort was therefore begun by a distinct (but overlapping) committee to fix the Haskell 98 Libraries.
Why Use Haskell?
Clearly you're interested in Haskell since you're reading this tutorial. There are many motivations for using Haskell. My personal reason for using Haskell is that I have found that I write more bug-free code in less time using Haskell than any other language. I also find it very readable and extensible.
Perhaps most importantly, however, I have consistently found the Haskell community to be incredibly helpful. The language is constantly evolving (that's not to say it's unstable; rather that there are numerous extensions that have been added to some compilers which I find very useful) and user suggestions are often heeded when new extensions are to be implemented.
Why Not Use Haskell?
My two biggest complaints, and the complaints of most Haskellers I know, are: (1) the generated code tends to be slower than equivalent programs written in a language like C; and (2) it tends to be difficult to debug.
The second problem tends not to be a very big issue: most of the code I've written is not buggy, as most of the common sources of bugs in other languages simply don't exist in Haskell. The first issue certainly has come up a few times in my experience; however, CPU time is almost always cheaper than programmer time and if I have to wait a little longer for my results after having saved a few days programming and debugging, that is okay.
Of course, this isn't the case of all applications. Some people may find that the speed hit taken for using Haskell is unbearable. However, Haskell has a standardized foreign-function interface which allow you to link in code written in other languages, for when you need to get the most speed out of your code. If you don't find this sufficient, I would suggest taking a look at the language OCaml, which often out-performs even C++, yet also has many of the benefits of Haskell.
There have been many books and tutorials written about Haskell; for a (nearly) complete list, visit the Haskell Bookshelf at the Haskell homepage. A brief survey of the tutorials available yields:
- A Gentle Introduction to Haskell is an introduction to Haskell, given that the reader is familiar with functional programming en large.
- Haskell Companion is a short reference of common concepts and definitions.
- Online Haskell Course is a short course (in German) for beginning with Haskell.
- Two Dozen Short Lessons in Haskell is the draft of an excellent textbook that emphasizes user involvement.
- Haskell Tutorial is based on a course given at the 3rd International Summer School on Advanced Functional Programming.
- Haskell for Miranda Programmers assumes knowledge of the language Miranda.
Though all of these tutorials are excellent, they are on their own incomplete: The "Gentle Introduction" is far too advanced for beginning Haskellers and the others tend to end too early, or not cover everything. Haskell is full of pitfalls for new programmers and experienced non-functional programmers alike, as can be witnessed by reading through the archives of the Haskell mailing list.
It became clear that there is a strong need for a tutorial which is introductory in the sense that it does not assume knowledge of functional programming, but which is advanced in the sense that it does assume some background in programming. Moreover, none of the known tutorials introduce input/output and interactivity soon enough (Paul Hudak's book is an exception in that it does introduce IO by page 35, though the focus and aim of that book and this tutorial are very different). This tutorial is not for beginning programmers; some experience and knowledge of programming and computers is assumed (though the appendix does contain some background information).
The Haskell language underwent a standardization process and the result is called Haskell 98. The majority of this book will cover the Haskell 98 standard. Any deviations from the standard will be noted (for instance, many compilers offer certain extensions to the standard which are useful; some of these may be discussed).
The goals of this tutorial are:
- to be practical above all else
- to provide a comprehensive, free introduction to the Haskell language
- to point out common pitfalls and their solutions
- to provide a good sense of how Haskell can be used in the real world
Additional Online Sources of Information
- A Short Introduction to Haskell: http://haskell.org/aboutHaskell.html
- Haskell Wiki: http://haskell.org/
- Haskell-Tutorial: ftp://ftp.geoinfo.tuwien.ac.at/navratil/HaskellTutorial.pdf
- Tour of the Haskell Prelude: http://undergraduate.csse.uwa.edu.au/units/230.301/lectureNotes/tourofprelude.html
- Courses in Haskell: http://haskell.org/classes/
It would be inappropriate not to give credit also to the original designers of Haskell. Those are: Arvind, Lennart Augustsson, Dave Barton, Brian Boutel, Warren Burton, Jon Fairbairn, Joseph Fasel, Andy Gordon, Maria Guzman, Kevin Hammond, Ralf Hinze, Paul Hudak, John Hughes, Thomas Johnsson, Mark Jones, Dick Kieburtz, John Launchbury, Erik Meijer, Rishiyur Nikhil, John Peterson, Simon Peyton Jones, Mike Reeve, Alastair Reid, Colin Runciman, Philip Wadler, David Wise, Jonathan Young.
Finally, I would like to specifically thank Simon Peyton Jones, Simon Marlow, John Hughes, Alastair Reid, Koen Classen, Manuel Chakravarty, Sigbjorn Finne and Sven Panne, all of whom have made my life learning Haskell all the more enjoyable by always being supportive. There were doubtless others who helped and are not listed, but these are those who come to mind.
Also thanks to the many people who have reported "bugs" in the first edition.
- Hal Daumé III