A Technology Blog by Alan James
Traditionally web applications were built using backend technology. A request is made to a server and based on the URL being requested the server responds by generating a resource, usually HTML.
To do this servers need to “route” requests to different resource generation logic. This logic is what backend developers build.
This worked well for many years, but as browsers became more advanced code started being moved from the backend to the frontend.
I’m normally not a fan of gamification. It feels like the person trying to teach me thinks I am incapable of learning and needs to trick me into it. However, I just discovered a great programming game that teaches you creative problem solving.
AND IT’S FUN!
Update: node-webkit is now called NW.js and uses io.js and Blink instead of Node.js and WebKit. All other information is still correct.
JavaScript can be run in a browser normally, on a server with Node.js, and on mobile devices with Cordova, but where else can we use JavaScript to program?
Well, plenty other places, but the last big place to write programs is the desktop and today I’m covering writing desktop JavaScript applications with node-webkit.
Last time I covered what Node.js is and briefly covered creating server side JavaScript applications with it. But as part of my recent desire to write everything in JavaScript I have also started writing mobile applications in JavaScript using Cordova.
I’ve written mobile applications in various technolgies and even have some on the Google Play Store, but I like writing mobile applications in Cordova more than other technology I’ve used, which are Java and Flex with Adobe AIR.
JavaScript used to be used only within the browser. But nowadays JavaScript can be used pretty much anywhere programming is supported. It can be used in a browser, server, mobile device, and even desktop.
I’m going to write about building JavaScript applications for mobile devices and desktops in future posts. Today I will discuss how to make server applications in JavaScript.
In my last post I talked about the great annyang library by Tal Ater, which is a keyword detector and abstraction layer for Chrome’s continuous speech recognition system.
Annyang is a great client side JavaScript library for detecting a spoken keyword and performing a specific callback function. But what if you want the same functionality server side?
Use annyang-node of course.
Recently I’ve been working a lot with continuous speech recognition and keyword detection in Google Chrome. Several months ago I found a really great library for streamlining is the process, a library called annyang.
For a long time I’ve had a CSS and JavaScript library that I’ve maintained for my personal and professional projects. The CSS included UI components, like buttons and switches, and the JavaScript were ports of all the PHP functions I had written so I have the same functions client side that I had server side.
Recently I released these libraries under the banner of a larger project, MogulMVC.
Since releasing them on GitHub I’ve been able to install them in any project by just running a git clone or git pull in my CSS or JS directory, and this makes it very easy to keep them updated, but it was still lacking something.
Since deciding to use WebSockets instead of AJAX calls with Node.js I have discovered some interesting things. Some of these things are good, and some are bad.
Node.js is great at times, but in some ways it’s a real pain in the ass.
Sure, it can create WebSocket servers, and it seems to be on the cutting edge of everything, but try to update an array of objects in a database then output a message and you’re in trouble.