2004 has definitely been a big year for us in several areas where I think we set some milestones. In terms of product innovation and mass enablement, we passed J2EE 1.4 certification ahead of IBM and BEA. We also made significant contributions to key specifications like EJB [Enterprise Java Beans] and have really taken an important role in the standards community being on the executive committee of Java; one of 15 companies that decide the future of Java.
The second area is from a customer adoption standpoint. 2004 has seen a shift from rolling out JBoss in production in a couple of boxes to standardizing across the company -- The National Laser Group and LaQuinta Inn & Suites are two examples there. Also, being part of the bigger movement of open source.
We increased our community involvement in 2004 announcing Hibernate, the jBPM acquisition and JEMS, an umbrella for these projects and communities. How important is J2EE certification to enterprises implementing JBoss?
J2EE certification is a seal of compliance with the 23,000 tests put forth by Sun. It's been quite a big effort on our end. We committed most of the company's resources to pass it and we've done it. We are ahead of the competition being the first app server to be certified. To the marketplace, it shows a seal of quality -- it's safe to put in production and is certified by Sun.
J2EE 1.4 certification was a Web service standard. To pass certification, we had to rewrite Axis, the open source implementation of the SOAP stack. A lot of the infrastructure for those Web services is present in the JBoss app server. What are your customers saying about Web services?
We follow Web services with interest and some detachment. We are technologists and open source people. We include features in products when customers want them badly. We don't sell licenses. We view Web services as a vendor push to differentiate. We see vendors talking about Web services, but we're not seeing customers talk about it. They ask for technology lower in the stack, because it's a great integration technology; expose business services as a WSDL [Web Services Description Language] and have a coarse grain of computing with Web services. It's not like our customers are saying we need WS-Orchestration right now. It's a wait-and-see attitude. If the market goes that way, we will too.
Let Microsoft and IBM duke it out. We'll go with it.
It simplifies marketing and sales, and, yes, it's a branding effort. From a technology standpoint, we take a plug-and-play approach.Think of JEMS as a motherboard, a bus, a microcontainer or microkernel. On top of a thin motherboard, you can plug in modules like you plug in cards to a motherboard: Cache, Hibernate, etc.
Everything plugs in at an API level into that bare generic motherboard, which is a bundle of JBoss microcontainers. It's all plug-and-play. We don't require you to use all or nothing like the competition does. A lot of customers use Hibernate on WebSphere, and we're happy to support that.
The second part is a branding effort. JBoss in 2005 wants to rewire the brand associations. Our brand is associated with an app server. It's a great one, a great brand, but what we want to do is transform from a product brand to a suite or platform brand so when think about middleware, you think about JBoss.
The second thing is we want to associate the JBoss brand as a methodology, which is a professional open source. Unlike a packager, we don't take stuff that's on the Web, put it on a CD and say, 'Here's our package.' We operate as a vendor; we do our own R&D. Does your open source branding ever hurt your business?
We are known as an open source company, and our brand is associated with our app server. In fact, it's exactly what we're trying to do -- change from an app server brand to an open source brand. How much does that hinder us? Well, our business comes from open source. In many cases, enterprises use JBoss, and don't know JBoss Inc. structures the technology they're using. Companies use a technology, and when they standardize, they want to know a technology is viable and has a company message behind it. If you buy from us, you are investing in the technology by funding the developers. This is unlike proprietary companies where you buy licenses. We don't sell licenses. There's also growing interest in business process management in IT. What drove you to acquire jBPM?
Web services in relation to BPM for us is a language for exchange, whether you're talking native Java sockets or CORBA, it's transferable from BPM. BPM is a deeper construct for developers than just a language used to exchange data formats.
With that in mind, BPM is used out there and there's a lot of demand. Workflow is a core construct available for developers to use. The challenge in that market is that there are many [BPM] vendors, about 200 with a niche. It's a fragmented market, and also an opportunity. As an open source player, we accelerate consolidation. Consolidation is mega trend going on in the industry. We've definitely seen it in the app server market where we're down to three: IBM, BEA and JBoss.
BPM is for real, but there are too many vendors. We hope to see consolidation.
FEEDBACK: Do you agree with Marc Fleury that Web services is primarily an integration technology right now?
Send your feedback to the SearchWebServices.com news team.