Reply to comment

Frank Millar
posted 8 years 25 weeks ago
I agree with Dennis that Tom's assertion, "SOA is perceived..." is interesting because the perspective is outside of my experience. The way that I look at this stuff, Services Oriented Architecture is a dependency of the Business Architecture, which is in turn, a dependency of the Enterprise Architecture. The larger the organization, the less likely that org will be ready, willing, or able to do a full EA. Nevertheless, an EA that is cursory/superficial/incomplete for the whole organization can set an important foundation for a deeper dive into a targeted area, starting with the business architecture. Among other elements, the business services of this BA should be clearly identified. This sets the stage for an SOA that is traceable all the way back up to business operations strategic goals that support the enterprise strategy. Services Oriented Modeling and Architecture (SOMA), an IBM favorite, offers a step-by-step approach for creating a SOA, and this method can be dovetailed nicely with a business architecture that includes its business services clearly identified. I've held that this dovetailed approach has merit for some time, only to have my perspective on this reinforced during my first reading of TOGAF9. That standard more or less debunks certain SOA methods (I think the TOGAF9 authors might actually be referring to SOMA!) as too technically oriented and insufficiently business oriented. I'm comfortable that this dovetailed approach addresses the issue expressed in TOGAF by deliberately threading the helpful SOMA method back into the services elements of the Business Architecture. I believe that TOGAF is working on a more business-centric focus for SOA. It will be interesting seeing what The Open Group ends up putting in the standard. [Updated on 4/24/2010 8:53 PM] [Updated on 4/24/2010 9:11 PM] [Updated on 4/24/2010 9:14 PM]


Shopping cart

View your shopping cart.