A couple of weeks ago Thomas Erl, one of the main authors on all things SOA, donated his Service Modeler invention to Red Hat. As Thomas says:

"Service modeling is an increasingly important area for SOA. By providing a simple, platform-agnostic tool that can be used by a range of IT professionals to conceptualize, define and manage inventories of services and service compositions, this technology has the potential to drive more widespread adoption and improve the end results of SOA initiative"

The modeler should play a very important role in the development and lifecycle of services; something which, until now, has been missing or developers have been forced to use inadequate or ad hoc techniques and tools. As Thomas mentions and others agree(ourselves included), Web Services do not infer SOA (apart from the fact that one is an implementation technique whilst the other is an architectural style). So although some of what the modeler provides is available in pure-play Web Services environments, that doesn't help those users who don't want to, or can't, use Web Services. Plus, even in a Web Services environment, we're confident the modeler provides good value add.

This is a good thing for Red Hat, but also for the open source community. The invention and any patents around it will be covered by the standard Red Hat Patent Promise. But it's gratifying that Thomas chose Red Hat as the home for the modeler. As he said:

"Red Hat is an ideal place for SOA-related open source innovation today. I have great confidence in Red Hat's ability to oversee the open source project being based around this service modeler and transform it into a mass-market solution."

At the moment there is no implementation of the modeler invention, but we'll be trying to rectify that soon. The aim is to fold this technology into the Red Hat SOA Platform and our ESB as soon as we can with the help of the community. Innovation continues to grow!