Loosely Coupled Thinking

This is a classic example why ROI is an awfull metric: you can always argue that given enough time you will get the benefits.

Personally I think the problem lies on why, how and when the enterprise adopt SOA. Going all SOA for instance is a recipe of disaster. On the other hand, implementing it in a project to see how it goes for no other reason, will probably give no value to the enterprise.

I would say enterprises must base SOA adoption in real business needs, where there will be clear value added to the enterprise in the short term. If its value isn't clear in the short term, then some consulting services on the subject must be used (or a better one).

That said, the problem is above SOA. The following article enterprise architecture article, gives some great advices on how to get business value: http://msdn2.microsoft.com/en-us/architecture/aa479371.aspx

Overly Hyped and a low rate of return?

I have been consulting for 20 years.  Been in many different companies from small to the largest.  It would be interesting to see how they would compare the benefits from SOA to projects of the past when very little companies really ever tracked the past projects very well.  

A company might show results where projects were completed on time.  However, how soon did they need to provide an update release to fix or add functionality that was missed the first time?  

Grant you, there are some companies that do a great job at delivering projects.  

When it comes to evaluating a ROI, for me those results will surface in the future.  Once in place, faster development, much lower support costs, which then leads to the ability to put more into development for new functionality and increases the bottom line.

I would suspect those who make this claim also believe a new application can be CORRECTLY developed in 30 to 60 days.

bkaplun@columbus.rr.com

I could get used to this rolling out of bed into my office thing BizTalk Server The highly anticipated

Another SOA Zen Moment:

Making something easy to change makes the overall system a little more complex, and making everything easy to change makes the entire system very complex. Complexity is what makes a system hard to change.