First it was WOA vs. SOA, now it's Mashups vs. SOA. When will people start focusing on business problems and stop having religious wars about which technology is best? All of the above mentioned technologies are great tools for your IT toolbox. They should be used where they make sense. SOA is an architectural approach. Mashups and WOA are technologies that work well with SOA. Why can't we all get along?
One of the items on our SOA roadmap is extending our architecture to include mashups. We envision leveraging popular mashups like Google maps but we also hope that our customers can extend our products and services with their own mashups. Take a look at all of the mashups that are available for Salesforce.com.
I've said it before and I'll keep on saying it until somebody proves me wrong. The problem with these emerging technologies is not the technologies themselves, it is the people. Only people can take great concepts like SOA and Mashups and create a bunch of unnecessary noise that scares the masses away. Instead of arguing about semantics, people should try to better understand these concepts and figure out how they can apply them to solve real business problems. At the end of the day, the business doesn't care what technologies you use. They just want their tools delivered faster, easier to use, and at their finger tips wherever they are.
There has been so much hype about SOA and now WOA that sometimes I think we focus more on fantasy then reality. Yes, both of these technologies offer huge potential benefits but they are both just another tool in the toolbox. Neither one of these will solve world hunger as some media types tend to think.
The WOA craze seemed to start shortly after Anne Manes pointed out that many of the companies she talked to were failing in their SOA efforts. Then came the quick fix, WOA. Many pointed out how the business can grasp the concepts of WOA and not SOA. Others pointed out that it is easier and quicker to implement then SOA, which they claim takes years to implement. I believe WOA is a great compliment to SOA, but I fear many people are thinking about WOA instead of SOA.
Implementing SOA takes vision, leadership, planning, and architecture, something that many IT shops don't have the patience for. Instead they opt for quick and dirty habits that lead to the inflexible, costly to maintain systems that so many of us are stuck with today. I fear that if we bypass SOA instead of complement it with WOA, then we are are going to wind up with "Quick-n-Dirty 2.0", a web enabled mess. There are no shortcuts to developing a sustainable, flexible, and maintainable architecture. We should use these technologies where they make sense. They are not the end-all, do-all answers to our problems. The excitement that I hear about WOA reminds me of the day trading craze of the late 90's. Everybody was looking to get rich quick. We need to make sure that we don't get too crazy with WOA and quickly build nice eye candy systems that we can't maintain, have no control over system performance and SLAs, and can't fix when they are down.
I know this sounds like doom and gloom. I am actually a fan of WOA. But I am concerned that many companies will dive in head first without much thought about architecture, infrastructure, support, change control, dependencies, and many of the other critical areas that a sound enterprise must consider when it introduces new applications into the production environment.
Remember, we are architects not day traders.
