Dear Architect, When the first Architecture Journal
was published in January 2004 (back then, called Microsoft Architects Journal
), it contained an article, “Understanding Service-Oriented Architecture
,” in which two CBDI analysts explored a novel concept of application integration. At that time, most organizations ignored SOA and its implied concepts. Yet, in those companies that had some degree of SOA awareness, there were few or no ongoing projects that went in that direction. A few years later, it was rather common to hear of a given organization thinking to realign its whole IT strategy with an SOA model. While they were clear on the goal, the business justification, and the benefits to be had, the initial attempts were still not clear on the right path to take or (even more CIO-enervating) the accuracy of initial cost estimations. More than five years after that introductory article, thousands of SOA projects have been kicked off, and several of them are still being completed. These days, we have a better understanding about its ROI, low-hanging fruits, hidden costs, and other realities; and the basic, incipient SOA discussion is today specialized in a variety of threads—from service-related ones, such as versioning and discoverability, to broader, portfolio-level threads, such as business-process management. As if everything about SOA hasn’t already been said, other IT trends insinuate a sudden irruption that could twist the original course of SOA by using alternative techniques in noncritical scenarios. These include the modest, lightweight approach that REST offers, as opposed to the rich but heavier OASIS WS-* standards—the latter having matured and evolved with SOA as a purpose. This issue of The Architecture Journal
covers several of these debates. The main difference with the article that we published in the early days is that, this time, thoughts have emerged as a consequence of a practice; in 2004, thoughts had emerged as a consequence of a vision. Both perspectives are necessary: the vision, to understand the goal and what we intend to achieve, and the practice, to help us understand the real dimension of constraints and how to mitigate (if not avoid) them successfully. These eight articles discuss aspects such as business alignment, service modeling, governance, federation, infrastructure, reusability, convergence, and coexistence with cloud computing and event-driven complementary approaches (among others), and they’re accompanied by guest columns that analyze other tangential aspects. I’d like to stop here to thank Eliaz Tobias, IT architect and SOA expert from Microsoft Israel, for his collaboration as guest editor-in-chief. As we promised two issues ago, we were in the process of leveraging more digital formats. In that sense, this set of articles and guest columns is complemented by a series of short videos
on yet other aspects of this SOA topic. Since our first issue, SOA has walked a long trail, and we still expect a long journey ahead. However, this checkpoint along our way is a great opportunity to share and reaffirm certain concepts that will be useful for the remainder of our journey. I hope that you enjoy these articles, dear reader. As usual, you may send us your comments at mailto:archjrnl\@microsoft.com.
. Diego Dagum Editor-in-Chief