A boy named SOA

Dr Jason Weisse, vice president for IBM's Enterprise Integration Solutions organisation, has a worldwide responsibility for managing the design, development and deployment of IBM's SOA technology. He has been working closely with Dubai Municipality on its SOA initiatives, which are due to be piloted later this year. He talked to Colin Edwards during one of his many recent trips to the region.

  • E-Mail
By  Colin Edwards Published  June 1, 2006

|~|Wiesser,-Jason----IBM200.gif|~|Dr Jason Weisse, vice president for IBM's Enterprise Integration Solutions.|~|Arabian Computer News: How much interest is there in SOA in the Middle East?

Dr Jason Weisse: We've been working here for some time on a number of initiatives in the Middle East region. When we started piloting SOA about two years ago, we had a summit in the United Kingdom and quite a few clients from this region came to it. What we discovered fairly quickly was that there was a tremendous appetite for innovation down here because users are facing many, many challenges from delivery and business standpoints. They had already decided that they did not need to go through the pains and errors that their brothers and sisters elsewhere in the world had made and so they were really trying to see if SOA gave them a way to leapfrog many of those issues particularly those surrounding what to do abut legacy systems.

ACN: What's happened since then? Was that initial interest translated into action?

JW: We have begun a number of projects down here in the region. These are pretty much in every core sector - government, banking and industrial - where the customer is taking what I refer to baby innovation steps. This is a sensible approach towards an innovative SOA strategy transformation.

ACN: But I've seen IBM advocating that these initial steps should not be too small? What do you mean by baby steps?

JW: Let's be careful what we mean by steps. From a business standpoint the business step needs to be meaningful enough to generate sufficient value so that the business owners see the merit of supporting this kind of transformation. At the end of the day SOA is a cultural transformation. It's not just a technical project. It's not just a bunch of technical people saying: 'let's launch a new technology - that will be fun - and oh, by the way, you can spend a lot of money on it and next year - when the next flavour of the month comes along - we'll replace it with something else.

We are seeing that the size of that first step depends on what needs to be accomplished. A case in point was MCI Telecommunications. A baby step for them was launching a new product. It was a pretty damn big baby step. I would say that it was probably the biggest baby I've every seen in my life

ACN: What sort of product was it?.

JW: It was a VoIP-based product. Michael Capelas, the CEO at the time had decided that for the company to get out of bankruptcy and then to to transition, it needed a big play in the VoIP commercial space. No one had been able to offer an IP-based commercial product. The only problem he had was that MCI had six different provisioning systems, six different billing systems, 32 different databases and 16,000 applications - none of them were connected. So it was going to be a bit difficult launching a new product with systems that didn't work together. We enabled him to launch that in less than 18 months

So that was a big step. The next step for them was that they sold the company. They clearly proved the point and one of the reasons they sold the company was that they had begun the transformation to SOA.

ACN: What kind of steps are being taken in the region? Is they being taken within traditional legacy environments for core applications or are they in greenfield territory?

JW: The beauty of SOA is that it really doesn't matter where you're coming from. As a result we're seeing them come in from both legacy and deep legacy structures as in banking, finance and government to fairly infrastructure-free zones. In fact we're working with one prospective project right now where they are literally starting from scratch. It is a very large entity and they are literally starting with a greenfield. They have much of the same kinds of challenges as the legacy structures have in decision-making, but they have an easier job of setting governance and setting standards and basically make decisions about how to deploy infrastructure. They don't have a lot of legacy interface issues such as other people have.

ACN: What stage are the pilots at here?

JW: Bear in mind that all the pilot projects we do are production ready. This means that when they are done they can go online live and have been tested and are ready to go live. The first project will go live later this year. This is the first large project in the region.

ACN: Can you share any details on the nature of this project?

JW: It's Dubai Municipality. It is a highly innovative project. Last year they won a Sheikh's award for technical excellence as a result of their innovation strategy around this. It is a joint partnership between us and the Municipality. It's Linux based, which is very exciting for us and it is a very important initiative for them. It's very exciting because the Dubai Municipality has the opportunity to become a centre of excellence. And I think they are excited by the prospect of taking up such a leadership responsibility and to be able to drive this kind of centre of excellence.

ACN: So should SOA initiatives be undertaken as a vendor/client partnership? Some people say they should do it alone - internally?

JW: My knee jerk reaction to this is that they are right and they are wrong. This technology and this model is very complex, but all of a sudden everything is SOA. Every vendor has a SOA. They're naming children after SOA, I've got a dog called SOA. Companies are changing their names to SOA.

The reality is that few have a sense of what they are talking about. I think businesses and governments need to be excruciatingly mindful of strangers bearing gifts or, in this case, vendors bearing gifts. Partnering around this kind of transformation requires a level of due diligence to ensure that you are working with partners who have not only been down this road for a number of years successfully, but who are willing to really partner by absorbing portions of the risk to ensure success.

ACN: Before joining IBM in late 2002, you were involved in shaping Microsoft's .NET product architecture. Why the move to IBM?

JW: Microsoft does not do this type of technology. They never did. They are a technology company driven by commodity and as such they provide value services and assets, but when you start dealing with enterprise-capable requirements and complex integrated systems and mainframes and servers and data structures, it is just not their work.

I was chief enterprise architect there and realised - a number of us realised - that we were either going to retire and take up duck farming or we were going to find ways to make it happen. I was approached about four years ago and given the opportunity to determine if this was the direction IBM needed to take. I was given a great deal of support by our CEO and IBM bet on the deal. Three years later we have proven that this is the direction the industry is taking. This is not another flash in the pan.

ACN: Is SOA going to be the final architecture?

JW: I think that it will be the final framework. I think that what it does is it creates the ultimate degree of ubiquity. It's extremely infantile right now. What I mean by that is that SOA, in the same way as electricity is in its adolescence right now, is in its infancy. It will achieve a degree of maturity over probably the next 25 to 30 years as it becomes more transparent.

Having said that it will change the entire industry. We're seeing for example more and more independent software vendors and development houses starting to develop services and components and we see the notion of composite applications starting to dominate the way ISVs are building technology.

ACN: Are you convinced that SOA will deliver on the integration promise?

JW: Am I convinced that business understands and sees the value it will deliver on the promise? Yes. Am I convinced that business will see the value quickly? No. Am I convinced that ultimately whoever replaces the people who didn't get it the first time will see it? Yes.

I've made a statement on a number of occasions publicly about the fact that if the people sitting in a room I'm talking to don't get this and decide not to move down this road today, whoever replaces them will. It's a juggernaut. It's moving forward whether you want to be part of it or not. The reason it's moving forward is because the entire community is behind it. This isn't another server client or Corba. This is like Alice's Restaurant.

ACN: What is going to have the biggest impact on computing and the business in the years ahead - SOA or are there other technologies in the sidelines?

JW: Grid and virtualisation are how things are going to shape up over the next couple of years. Grid is probably going to be adopted more rapidly over the next 10 years because it is cheap and efficient. And do you know what? It works. Virtualisation is another example. Should it matter where something is sitting? The reality is no. If you want to deal with the ubiquity and the value SOA brings you have eliminate anything that has to do with physical proximity. You move to a fully virtualised environment. That way services become available as needed and this is what our chairman, Sam Palmisano, talks about when he refers to ‘on-demand’ computing; he's talking about on-demand business. He's saying: “Oh you're a business person; you're responsible for marketing new products; you're not a technical person; and you need to market a new product. Great - go to your little business modeller; drag a couple of big animal pictures together until it does exactly what you want you push a button - and there, it's done.”

ACN: Is it really going to be that simple? Is that what we can expect when SOA starts to be rolled out?

JW: I think in the next 10 to 15 years it will be that simple. I absolutely think it will be. Just let me remind you what computing was like in the mainframe days. When you and I started in IT, I was programming on Fortran and I thought that was pretty good. Then Pascal was introduced - oh my God natural language text programming - I've seen heaven. That was not long ago. You got to remember that Mosaic was the browser of choice 15 years ago. So do I believe it will happen? You don't have to be a rocket scientist to predict this one.

Add a Comment

Your display name This field is mandatory

Your e-mail address This field is mandatory (Your e-mail address won't be published)

Security code