JsonSchema, StrongLoop and Velocity

Pictured - the Google Trends graph for "json schema". I wrote the Model Driven Node (MDN) project exactly two years ago. I used java-based tooling to generate node.js scripts because I knew those tools and the javascript equivalents weren't mature. For instance, JsonSchema for node seemed unfinished but I

Pictured - the Google Trends graph for "json schema".

I wrote the Model Driven Node (MDN) project exactly two years ago. I used java-based tooling to generate node.js scripts because I knew those tools and the javascript equivalents weren't mature.

For instance, JsonSchema for node seemed unfinished but I figured "enterprise mentality" would enter the node.js market and interest in json schemas is taking off as developers realize that centralized message definitions are key to scalability. Last week I revisited JsonSchema and pulled it into my current project. I'm pretty pleased so far.

XSD schemas generated most of my MDN code to minimize hand-coding and avoid a weakness of javascript. Last Christmas, I wrote a brand loyalty prototype using StrongLoop, MongoDB and Nools Rule Engine in about a week. StrongLoop generates the REST api dynamically from json schemas; a better way of accomplishing the same goal.

(Addendum: I wrote this about a month before IBM bought Strongloop. The purchase makes a lot of sense to me. http://www-03.ibm.com/press/us/en/pressrelease/47577.wss)

Java-based Velocity templates generated MDN because javascript alternatives seemed sparse. Imagine my surprise last month when I found this - Velocity-Node A javascript version of Velocity! I haven't used it yet but I'd like to try it in a real project.

The components and trends are coming together for real enterprise Node.js. Open source, especially Node, is displacing proprietary enterprise java frameworks.


Oracle

IBM Websphere

Siebel



I previously worked with IBM's Websphere Commerce Suite. It's an enormously complex product which was developed by

relatively few developers for
relatively few customers over a
long period of time.



I believe Node.js is succeeding because it crowdsources development across

many developers for
many customers over a
short period of time

which creates (as I wrote in 2006)

a different Project scalability.

It's nice to see a prediction from over two years ago coming true. I'm fairly sure my other predictions (demise of the "analytics boom" during the 2016 recession and The Vertical Web) are on track, too.