ClojureScript: Up and Running

Chapter 1. Introduction: Why ClojureScript?

This book aims to get you up and running with ClojureScript, a dialect of the Clojure programming language that compiles to JavaScript. To begin, this chapter will provide some motivation for why ClojureScript exists.

The Rise of Browser Applications

Web applications have come a long way from simple CGI scripts, but they have always been constrained by the stateless request-response model of HTTP. As the “pages” in a web application become more elaborate, the cost in time and bandwidth of reloading an entire page just to update a single piece of information becomes prohibitively high.

One of the first major uses of JavaScript on the web was to ameliorate the cost of small updates, but “web applications” remained primarily server applications for a long time, and for good reason. Deploying an application to a web server is much easier than distributing it to diverse client machines: the server is a controlled environment, with many more options for programming languages and frameworks. But by treating web browsers like dumb terminals, applications were severely limited by how quickly they could push updates to a client.

Recent years have witnessed the rise of what one might call browser applications. These applications typically still have server-side components, but a significant part of their logic runs client-side, in a web browser. The web browser acts like a virtual machine, executing JavaScript code to communicate with a server, render a graphical user interface, and make all the local decisions that do not require a server. The result is a more responsive, more fluid style of interaction for client applications. While the original World Wide Web of hyperlinked documents will likely remain for many years to come, it seems probable that web server applications will be largely supplanted by web browser applications.

The Rise of JavaScript

Browser applications were made possible by dramatic improvements in the JavaScript execution environments packaged with web browsers. High-performance JavaScript engines such as WebKit’s SquirrelFish, Mozilla’s TraceMonkey, and Google’s V8 proved that JavaScript could be fast and launched the browser performance wars. JavaScript began to succeed as a general-purpose application platform where other in-browser execution environments had failed. It was a historical accident that no one could have predicted, least of all the early developers of JavaScript.

Although JavaScript has many flaws, it has a few strengths that allowed it to succeed:

1. It is a small language. Core JavaScript has a limited number of keywords, concepts, and built-in features. This makes it easy to embed in different environments.

2. It is flexible. Features missing from core JavaScript, such as namespaces or classes, can be added using the language itself.

3. JavaScript functions are first-class. Although JavaScript is not a “functional” programming language in the usual sense, the ability to create and compose functions as values grants it immense power.

4. It’s there. Every web browser has had JavaScript built-in since the mid-1990s. Beyond that, the ease of embedding JavaScript in other applications has led to its inclusion in products as diverse as databases and television set-top boxes.

The Need for a Better Language

Despite JavaScript’s overwhelming success, it still has many flaws (see Douglas Crockford’s excellent book, JavaScript: The Good Parts (O’Reilly)). It was a product of unpredictable evolution, not a carefully thought-out design process. And the vast reach and diversity of JavaScript runtimes is both a blessing and a curse: it will be difficult to create a new and improved version of the language that can replace all of the “legacy” versions deployed around the world.

So JavaScript is here to stay, probably in its current form, for some time. Some have gone so far as to say that “JavaScript is the assembly language of the web” (see Scott Hanselman’s article, “JavaScript is Assembly Language for the Web”). So now we are beginning to see the rise of tools and languages that treat JavaScript as a compilation target, much like bytecode on a virtual machine or object code in a traditional compiler. For example, the Google Web Toolkit compiles a subset of the Java language to JavaScript. We even have entirely new languages, such as CoffeeScript and Dart, designed to target JavaScript compilation directly.

Any cross-language compiler has to make decisions about where to draw boundaries between the source language and the target language. CoffeeScript, for example, is deliberately designed to have semantics very close to those of JavaScript, adding only a cleaner syntax and protection from some of JavaScript’s more egregious flaws. Google Web Toolkit, on the other hand, is designed to hide JavaScript from developers who want to work exclusively with the Java language.

Introducing ClojureScript

ClojureScript is a version of the Clojure programming language, which compiles to JavaScript. Its primary target is web browser applications, but it is also applicable to any environment where JavaScript is the only programmable technology available.

Clojure is a powerful, expressive, Lisp-like language developed for the Java Virtual Machine (there is also a community-maintained port of Clojure to the .NET Common Language Runtime (CLR)). ClojureScript is more than Clojure syntax layered on top of JavaScript: it supports the full semantics of the Clojure language, including immutable data structures, lazy sequences, first-class functions, and macros. Explaining how to use these features in ClojureScript will be the subject of this book.

Clojure was designed to have a symbiotic relationship with the JVM: it does not try to hide all the details of its host platform. In the same vein, ClojureScript does not try to hide all the details of JavaScript or the browser execution environment. ClojureScript uses the same native types as JavaScript, such as strings and numbers, and can call JavaScript functions directly without any special “wrapper” or “foreign-function” code. ClojureScript is also designed to work closely with best-of-breed JavaScript optimization tools such as the Google Closure Compiler. These relationships will be explored in Chapter 3.

In summary, ClojureScript provides developers with a language that is more powerful than JavaScript, which can reach all the same places JavaScript can, with fewer of JavaScript’s shortcomings.