6 things you should know about Node.js

Get the highlights of Node.js -- the runtime system that gives JavaScript coders an on-ramp to the server side

JavaScript is eating the world, with new tools and enhancements arriving at a breakneck pace. With Node.js, an open source runtime system invented in 2009 by Ryan Dahl, that reach has extended to the server side.

javascript

Node.js has become wildly popular, with coders everywhere using it to create APIs and build a new matrix of interoperability across the Internet. Joyent has been the chief sponsor of Node.js from the beginning. In this week's New Tech Forum, Ben Wen, vice president of product marketing at Joyent, outlines six things you should know about the phenomenon shaking up backend development. -- Paul Venezia

Node.js is a runtime system for creating (mostly) server-side applications. It's best known as a popular means for JavaScript coders to build real-time Web APIs.

But Node.js is not a JavaScript framework; indeed, several authors have written excellent frameworks specifically for Node.js, including Express.js, Restify.js, and Hapi.js. So what exactly is this phenomenon finding its way into Web applications, operating systems wrappers, microcontrollers, and robots?

At it's core, Node.js is a stripped-down, highly customizable server engine -- a proto-server, if you will -- because out of the box it doesn't do anything until you set it up. This proto-server processes in a loop, ready to accept and respond to requests. Any of those requests themselves may initiate other requests to some other part of the system, such as to read a file off of disk or to send a signal to spin a motor on a robot arm. That loop, known as the event loop, is the "runtime" part.

Node.js ships with workhorse connectors and libraries such as those relating to HTTP, SSL, compression, filesystem access, and raw TCP and UDP. JavaScript, already tuned for a Web browser's event loop environment for GUI and network events, is a great language for wiring up these connectors. You can snap connectors onto the event loop almost as easily as you can snap Lego parts together. Doing so lets you create a simple, dynamic Web server in just a few lines of JavaScript.

In short, Node.js is a runtime system that makes it easy to build a network or other event-driven application servers. Here are the six things you need to know about it.

1. JSON has won

JSON (JavaScript Object Notation) is a practical, compound, wildly popular data exchange format. JSON has enabled JavaScript developers to quickly construct APIs and foster interoperability at scale -- a key objective for Node.js coders. JSON's stark simplicity can be expressed in just five railroad parse diagrams, notably without the self-conscious preening of XML and its scheming friends (SOAP, XSD, WS-*, RELAX-NG, and their endless committee meetings).

JSON and JavaScript have reinforced each other's importance. Back in the early days of the Web, dynamic data in the browser had to be manipulated, filtered, and operated on by the only reasonably comprehensible non-plugin language available: JavaScript. Regardless of its original network-presentable format, data needed to be marshalled into a JavaScript object. The dependence on JSON for general purpose data description gave rise to document-oriented NoSQL databases such as MongoDB and CouchDB. It's all JSON all the time today.

2. JavaScript is everywhere

JavaScript is a quirky, object-oriented, C-like language. It's the only choice for developing applications in the browser, with a new framework introduced every week to woo developers. And with Node.js, JavaScript has spilled over to the server. Competing implementation teams have driven JavaScript interpreters forward, so that Google's V8 engine is respectably fast -- fast enough to reside at the core of Node.js.

JavaScript also has the internal capability to handle the event loop mechanism in a straightforward way. Other languages have this capability, which are used by their own evented systems. Python has Twisted and Ruby has EventMachine. But because of history, both of those event-loop systems come freighted with relatively easy ways to make a particular kind of performance mistake, while JavaScript remains relatively free of this peril.

JavaScript also runs across many OS environments, having historically had to support them in the browser. This, along with the libuv library to help abstract away some of the operating system differences, means that Node.js has a broad footprint.

But the biggest force for JavaScript's migration to the server side is human. Programmers have to do less mental context-switching between a Web browser and the server. There are even efforts to unify the environments between client and the server so that code can run equally well in either location, further simplifying the model and leading to increased productivity.

3. Sharing is encouraged

The ethos of the Node.js community is "share gleefully." It's frighteningly easy to share packages of library code -- technically, culturally, procedurally, and legally. The Node Package Manager is included with Node.js and has grown to a repository of nearly 50,000 packages, making it likely that another developer has already packaged up a solution to your problem, or even some less common ones.

Node.js' namespace philosophy is essentially the absence of one, letting any author publish under an unused module name in the shared public repository. Sharing code under the MIT open source license is highly recommended in the community, which also makes cross-pollination of code relatively worry-free (and lawyer-free) from an intellectual property perspective. Finally, the community is highly engaged in binding interesting C libraries like computer vision (OpenCV) and the Tesseract open source optical character library. The latter, for example, makes possible weekend projects like Imdex that process images from the Web so they can be automatically searched for written content.

4. Node Package Manager works broadly

Speaking of managing library dependencies, the Node Package Manager deserves to be called out. Node Package Manager is the root of almost all deployment systems for Node.js and underlies the many PaaS (platform-as-a-service) providers for Node.js, actually making it somewhat easy to move smaller applications between providers. Its simple, dependable package management has let the Node ecosystem grow extremely well in recent history, to the point that the underlying public service now needs to scale to the next level.

5. 'Batteries not included' minimalism

Node.js applications and Node.js Core itself are broken down into small modules that are composed and shared. Each package and tool can be scoped tightly and crafted to be manageable. These can then be baked together -- often without too much unnecessary kneading. The low-barrier, carefree nature of creating a module also encourages experimentation in the community, and there's quite a bit of overlap and experimentation in the package population. When executed well, each package typically handles one task (e.g. node-optimist.js: 'light-weight [command-line] option parsing').

6. Instrumentation

Finally, Node.js is well instrumented for production usage. What that means is that there are the tools to help bring an application to full production readiness and performance. As with any maturing technology, there are areas where more documentation, tools, and best practices could be helpful. But as Node.js heads towards its next major release, it's on fairly solid footing.

Node in context

If you know JavaScript, Node.js is a gentle on-ramp to asynchronous computing for the Web. And it just so happens that Node.js is suited to solve Web problems exactly of this type: integration and glue challenges, with cascading calls to API after API.

Where does Node.js not work so well? It's not totally appropriate in places where a single-threaded calculation is going to be the holdup, like some kinds of successive approximation or classification. In those instances, it's more efficient for Node.js to drop the request to an independent library that's dedicated to the task, where it can be distributed across hundreds or thousands of processors.

Node.js is maturing quickly and is being deployed in more and more mission-critical and revenue-critical systems, such as eCommerce Black Friday infrastructures. It's easy to get started with Node.js, and yet Node.js is deep enough to handle modern Web complexities. If you're building your next generation Web site -- especially APIs for mobile and Web integration -- or if you are creating something new that depends on underlying services itself, Node.js is a runtime system that could very well work for you.

New Tech Forum provides a means to explore and discuss emerging enterprise technology in unprecedented depth and breadth. The selection is subjective, based on our pick of the technologies we believe to be important and of greatest interest to InfoWorld readers. InfoWorld does not accept marketing collateral for publication and reserves the right to edit all contributed content. Send all enquiries to newtechforum@infoworld.com.

This article, "6 things you should know about Node.js," was originally published at InfoWorld.com. For the latest business technology news, follow InfoWorld.com on Twitter.

This story, "6 things you should know about Node.js" was originally published by InfoWorld .

Join the discussion
Be the first to comment on this article. Our Commenting Policies
See more