If you have been following the blog circles lately, there is a big buzz about SOA being dead. It all started with a recent blog post by Anne Thomas Manes in which she says “although the word ‘SOA’ is dead, the requirement for service-oriented architecture is stronger than ever.”
SOA at its very core is simply an architectural approach and not a technology stack nor a vendor-recommended product or platform. As Anne says, “they missed the important stuff: architecture and services.”
As I have always maintained, an SOA implementation can be as simple as a few business services that wrap business or application logic, and in its most complex form it can be an entire ecosystem of technologies selected based on thoroughly analyzing needs and that most importantly support service-orientation principles.
So whether we call it SOA or something else, I firmly believe that the need of the hour is service-orientation. Anne seems to think so as well as she goes on to state, “service-orientation is a prerequisite for rapid integration of data and business processes; it enables situational development models, such as mashups; and it’s the foundational architecture for SaaS and cloud computing.”
Another blog post by Lori MacVittie titled The death of SOA has been greatly exaggerated, speaks more eloquently about the real state of SOA. I really like the way she sums it up as “SOA is not dead yet; it’s merely reached the beginning of its productive life and if the benefits of SOA are real (and they are) then organizations are likely to start truly realizing the return on their investments.”
Stepping back, maybe this is a good time to get back to basics. Let’s get the data part of our infrastructure right, so that architectural approaches that borrow from service-orientation principles may look forward to guaranteed success by ensuring that timely, accurate and consistent data flows into applications and through business processes.
Taking this further, David Linthicum in his post So, if SOA is Dead, Should We Focus more on the Data? says “If we’re not going to do SOA, then back to the basics I say. I think that it may be a better approach to focus on concepts and approaches that are more simplistic in nature, such as data integration, and perhaps use that as a jumping off point to SOA, or SOA like substance in the future.”
I believe that we are onto something here at last. About a year ago one of our prospects had approached me for help stating that they were in SOA-hell. In discussing the problem in detail with the CIO and a few enterprise architects, it became clear that in an effort to enable increased customer satisfaction through web-based self-service, their IT organization had invested heavily in a laundry list of SOA tools and technologies. However, they were still faced with data inconsistencies, inaccuracies and staleness, specifically around their customer data that lay fragmented across diverse information sources. This scenario represents the growing frustration within IT organizations as they grapple with making excuses for having invested heavily in vendor-hype-driven SOA tools and technologies, without analyzing and solving real business problems.
As I have discussed in my earlier posts, if a data integration platform can quickly and seamlessly deliver data as a reusable and standards-based service or data service, an IT organization can look forward to reaping the benefits of both service-orientation and the availability of timely, accurate and consistent data. Additionally, with the focus now squarely on operational efficiencies, i.e., “doing more with less,” the platform also needs to enable the reuse of integration logic and skills, to ensure that resources are leveraged to the maximum.
What are your thoughts?