Is SOA Being Employed as a Tool for Streamlining or Cutting Budgets, or is SOA a Victim of the Budget Knife Itself?

Neither. SOA is being employed because it's in fashion to do SOA. Tomorrow, it will be in fashion to use Social Interaction Architecture and all applications will be developed around the concept of contextual usage patterns versus consumer/provider patterns. It would be fantastic if SOA actually lived through its entire maturity cycle, but most likely it will be superseded by the next popular trend.

Both. Remember that SOA was forged during the dot-com bust (2001-2003), seen as a way to leverage and interconnect existing IT assets that were being surfaced as Web services. At that time, nobody had the budget to buy new things, or undertake expensive integration projects; they were being forced to work with what they had. Still, we've seen a backing away from the big-budget boil-the-ocean SOA attempts, going back to its roots -- lightweight, addressing one pressing business problem at a time.

We are seeing SOA being used more for streamlining which then results in cutting their operational costs. I guess the answer to this question is dependent on your perspective. Since we provide an automated approach to SOA governance our customers are clearly going to be interested in making SOA efficient so they can quickly and easily meet the needs of the business.

I would agree that we are seeing both. The trend away from the large "boil the ocean" SOA projects is clear. No one has the luxury of extra time and dollars to deal with these type projects. It's all about solving specific business problems, and wringing inefficiencies out of existing processes.

Few organizations that are looking at SOA as being more than a technology initiative, but more of a mindset shift within their business and IT organizations are onboard that this is a long term joourney spanning multiple iterations rather than a one-time project that you implement and deploy. As such, they are treating SOA as something that can streamline budgets in the longer term.

To the contrary, there are many organizations that are taking a short term perspective at SOA and thinking of it is an expensive project or program with high upfront costs and hence cutting those down in today's economic environment and cost pressures.

The right answer, unfortunately, is not straightforward and requires a practical blend of the strategic direction with the tactical imperatives.

I think that Joe and Jim are spot on – SOA “projects� that are not nearing completion are in serious risk of evaporating, especially if they are not tied directly to a business process that is clearly driving revenue or saving money. This has been the case with large scale SOA projects from the beginning – outside of IT people’s eyes glaze over when you try to explain the drivers for SOA projects.

But that in no way means that SOA as an approach and as an implementation model is dead – it is here to stay.

Certainly yes and yes as far as we've seen. They are expected to use SOA to wring more costs out of existing IT assets, leverage and reuse partner services, and assemble software with as late a commitment of dollars and resources as possible.

At the same time, cost overruns of some "SO-" starter projects over the last 5-8 years that were either poorly planned or overdesigned (and rigidly tied to a single architecture or vendor approach) means that now, any SOA initiative is itself put under the budget knife to accomplish its goals.

That budget cut is difficult to take given the increasing complexity and rate of change in a service-oriented world. But it won't make SOA go away (whatever we call it) because there are already too many good examples of value out there.

I’d think neither - unless it is considered as a fashionable thing to do as JP Morgenthal describes.

I'd like to put the church back in the center of the village, as they say over here, and remind that SOA isn’t really a project, but rather a journey and methodology.

To better understand and position SOA, I often use the real world analogy. Imagine a world in which we have to supply all our needs on our own… that would set us deeply back in time, don’t you agree? If you think about it, modern society is based almost exclusively on services: trade, health, transportation, finance, and these services are founded upon specialisation, standardisation, regulation and scalability, among others. Come to think of it, we live in a SOA! Yet in the IT world, we’re still very much in the primitive autarchic era in which every solution/application has to supply all its needs on its own. Now, imagine how far a business technology (that’s the next thing beyond IT) could evolve with SOA and ubiquitous services – may be as far or even beyond what modern human society achieved by building on a SOA.

But SOA cannot be an objective by itself – just like in the material world, it is useful only when consumers actually need and consume the services. And continuing the analogy, the smaller the enterprise – the less it can spend on services that do not present a short term ROI. Coming back to IT, maybe SMB's are the pathfinders - adding SOA elements as part of specific projects with clear business cases.

So in short, SOA is here to stay and carry us to the next level – it is certainly far from dead as some provocative statements claim.

My opinion: "SOA is not even a factor in the larger discussion of budgets."

Budget decisions are made based on strategic, operational, and tactical objectives - in that order of preference. SOA really should not play a role in that decision as it is the means to an end. That is, SOA is the means to achieving IT alignment with the business with the end goal being becoming more "agile" in meeting the business objectives. So, the budget decision might be to marginalize the priority of a certain set of business objectives, which in turn might nix an SOA initiative. Contrarily, as other business objectives gain priority, SOA should be a consideration in the implementation of those objectives. So, the nutshell is that whether or not SOA gets used as a tool to stream operations or gets the knife itself is really a "side effect" of business level decisions related to long-term goals and objectives.

I'm late to the discussion party for this topic so I have the advantage of reading all the other comments. That said, I didn't see a comment that picked up the theme I was going to express until I read Tarak's. I struggled to answer Joe's question because it was similar to asking "will eating food be cut in this economic downturn"? IT still needs to deliver capability to meet core business needs and SOA becomes a viable "how" for IT to do this, just as eating becomes a viable response to hunger. What tends to get cut are two things 1) projects that are lower priority in terms of business impact and 2) IT improvement projects that may make people's lives easier but are not directly linked to KTLO (keep the lights on). This is similar to saying that dining out and take out cuisine may be cut in a tight budget crunch and supplanted with the lower cost but possibly less convienent cooking at home. Only as a last resort will eating be curtailed, much like only as a last resort will core business process and business application investments be curtailed.

So, SOA may or may not be cut as a result of a project being cut, but just as core busines projects will continue, so will SOA for organizations embracing SOA as a "how" for delivering against those project requirements.