WSO2: Dipanjan, you've really taken advantage of the intersection between open source and the flexibility offered by an OSGi componentized approach. How is that different than working with a proprietary alternative?
DS: What I demonstrated today was how an open source SOA stack can talk to a non-open source, commercially available SOA stack. It was really a matter of choice for us to deploy certain functionalities on the open source product using some of the features it provides by default or to follow the OSGi bundle to have a more appealing, comprehensive, user-oriented kind of UI for our Service Federation Framework. Today I was using the OSGi model, which in my opinion is far more appealing to the user.
WSO2: What is appealing about the ability to create and reuse different content and capabilities?
DS: The power of OSGi is the fact that it can be independently built, independently deployed, without creating disturbances in the other functionalities provided by the software. This is something that plugs in, kind of hand-in-glove, with the existing product. So the moment you develop an OSGi bundle for a specific purpose, and you deploy that, the functionality that it provides looks like an integral part of the product itself. If you don't like it, you just take out the OSGi bundle, and that particular functionality will vanish. It's the modular nature of programming which OSGi allows us, which is something that is attractive and we have taken advantage of.
WSO2: Do you see this use of OSGi expanding more with the open source community?
DS: It is a part of the open source family is because of the nature of open source, which allows you to handle the source code, which typically is not something allowed by the closed source applications. But I wouldn't be surprised if some of the OSGi capabilities down the line would be used in a closed source product as well. They may not allow you to customize it to your specific need. So OSGi, the way it is used in open source, I would say is far more developer friendly.