How many integration approaches does IT need? Didn't IT justify massive investments in EAI technology for Integration? Are most message queuing products, orbs, etc used for integration? Is SOA now just yet another integration approach?
Now, use of service-based technology - AKA Web Services - might introduce a greater degree of technical loose coupling to integration, reducing dependency on expensive EAI products, or enable greater virtualization. However, our experience is that folks coming at SOA from an integration approach are focusing on the S but not much the A. There is often little architecture, and the services they deliver are primarily of IT, but not business benefit. Moreover, they are often not that agile (a key promise of SOA) in that all they often do is render existing interfaces and datastructures as "services". Though this is not strictly IT's fault, as they find it difficult to get the funding or committment from business to do much else... Lawrence --- In [email protected], Gervas Douglas <gervas.doug...@...> wrote: > > <<A survey > <http://itknowledgeexchange.techtarget.com/soa-talk/survey-sees-soa-strength/>of > > the audiences at TechTarget Application Development Group member sites > proves some points that I have been making about SOA for a long time. > First, the most common entry point to SOA is integration. "Frye writes > that today's SOA projects are largely about integration. Survey results > show that the top benefits organizations hope to achieve with SOA are > improved data integration (32%), enable legacy application integration > (32%) and integrated disparate department applications (23%)." > > The next common entry point, and one that I have argued is the next > level of maturity from an application perspective, is business process > management. "Meanwhile, there is interest and uptick in Business Process > Management with 29.7% of respondents marking BPM as one of the critical > areas for their organization's technology efforts. At the same time, > 35.8% of respondents counted Business Process Management software among > the types of infrastructure software currently used, with 38% planning > to use it in the future." > > And finally, SOA definitely is not dead! "She notes that SOA use is > strong. Among the survey respondents, 49% said their organization has > one or more SOA projects under way, and 60% characterize their current > or future SOA projects as enterprise level as opposed to > departmental/divisional level (21%), or single, isolated projects (19%)." > > This is just the common sense approach to using SOA to improve > application integration, thereby reducing maintenance and support costs > within IT and reducing duplicate data entry and errors for the > end-users. And using SOA to improve business processes to reduce labor > costs and improve productivity is the next step after integration - > end-to-end process automation most often requires integrated systems. > > Ten years after I started working in this space we are back to the > reality and hard work of making SOA pay for itself in the same ways this > all started - the integration of distributed systems and improvement of > business processes. This is SOA blocking and tackling and there is still > a great deal of it to be done.>> > > You can read this blog at: > http://it.toolbox.com/blogs/the-soa-blog/soa-survey-points-to-practical-approaches-30900 > > Gervas >
