Emphatically, no.

First, portals are evolving. I reflected in a previous post that “The nature of portals has changed, and with GateIn (eXo + JBoss), we’re seeing portal’s transformation from an aggregator of external applications to a platform for building and delivering services that add value to the user organisation.”

In this world, the question is not whether a developer has to choose between portlets or gadgets but rather, which services are best delivered as gadgets and which as portlets.

At eXo, we do not make a great distinction between a gadget or a portlet. They are merely categorized in a way that makes it easy for users to go and grab more gadgets or portlets to customize their application. The user doesn’t care.

From the developer side, we use portlets for more robust components that take longer to build while gadgets are lighter and take less time to build. The difference comes from the technical aspect behind the scene. Even if we see some similarities in the last specifications, originally portlets were more Java server side components that generate fragments of HTML code while gadgets dynamically generate that HTML on the client side thanks to JavaScript and REST server calls. The main difference is then in the application lifecycle as portlets have to be deployed as WAR packages on the application server while the gadget is just an XML file that can be dynamically added and edited online. Gadgets allows our business users to build applications on the cloud very quickly.

Next generation portals will support both models but this is just one of the building blocks of the platform of tomorrow. Horizontal services like content, mail, calendaring, or profiles are other pieces that can be mashed up together to easily build a custom web application… online. We’re baking functionality into GateIn that will enable it to be the ultimate “mash up” platform for combining and recombining components. Times have changed, let’s be more agile.

Share this blog post with :

More »