CM: Where does SOA (service-oriented architectures) stack up among those technologies expected to shape future client opportunities for Accenture? Rippert: We do a technology vision every year at Accenture from September to December. Anything and everything that's of interest to our client base within our scope of practice is in the review. We narrowed it down to 42 technologies that we felt would be the most important to our clients and to us from now through the next five years. We put them into three categories: Most Important, Important, and Somewhat Important. SOA was number one within the Most Important category.

CM: That's quite a bet. Looking forward, can you somehow qualify SOA's expected impact on business?

Rippert: One of the things we're working on is a graph that reveals which technologies increased efficiency the most. Imagine a graph with time on the horizontal axis and the integration workdays required to implement a standard unit of work on the vertical axis. So call it 1,000 function points, if you want. Beginning in 1980 and going to 2006, what does that curve look like? One of the most significant examples of technology impact would be structured query language, which enabled the use of relational databases in the mid-'80s. I think that qualitatively SOA will be the biggest reduction in 26 years, since 1980. We don't currently go back before 1980; we don't have enough data for that.

CM: Does the technology development community share your point of view?

Rippert: I have a little bit of a debate going on with one of the big software vendors. You know, the question that I asked them was, "Once the standards are finalized, once people are using the full vision of SOA, what's the reduction in workdays for your package from when it was implemented in the net-centric world, or let's say 2003, before the first vestiges of SOA? Versus the workday reduction in the time of full-blown SOA?" Their answer is one-tenth the number of workdays. My estimate right now is one-fifth the number of workdays, so we're talking about a dramatic drop if all the standards mature, which has to happen in the level of effort required to implement and modify software, the amount of labor required.


CM: We know that you've been asked this before, but as a labor-based company, shouldn't you resist SOA?

Rippert: First, it wouldn't matter if we resisted it or not; it's going to happen. It won't be us or any one company that stops it or makes it. But secondly, as a longtime assembly language guy, I guess theoretically I should have resisted Cobalt because it was a lot easier to implement COBOL (Common Business-Oriented Language) than it was to implement assembly language, and we could sell more workdays in an assembly language project for the same level of function. The fact is that the improvement in efficiency just unlocked demand, and we grew much more quickly when we were in the kind of Cobalt, 4GL world than we did in the assembly language world. And I think that we're going to grow much more quickly in the SOA world, and we're going to have much more demand in the SOA world, even though the effort per unit will go down more than it's ever gone down before.

CM: What is Accenture currently doing to advance its own SOA-related work?

Rippert: We're doing two things to get ahead of the curve. First, we're writing a healthcare application just for ourselves to see how far we can push SOA architectures. Second, we're working out the modern approach to SOA development, focusing a lot on model-driven development and on domain-specific languages. Within 12 months, you'll see our release of that, which is a combination of work from my software architects, the fellows in Accenture labs, and our solution delivery guys. I think that it's going to be very, very interesting in how different it is, what a different mentality you have to have to get that major drop in workdays for implemented function by using SOA. You can't keep doing projects the way you used to. It's going to be pretty revolutionary

CM: What's the latest from Accenture labs?

Rippert: We have four areas of focus in the lab: systems engineering, human computer interface, intelligent device integration, and analytics. On the human computer interface side, we've developed an interactive billboard. It can be used for a lot of different things, but one of the applications is as a billboard. It's about eight feet wide and about four feet tall. It uses rear projectors. We wrote the display software, because we couldn't find it to our satisfaction, to create a very smooth display on the screens. And if you walk up to it, you touch it like a touch-screen computer. It has advertising on it and it has options you can touch. You can run a video, etc.

The billboard is now in the American Airlines terminal in O'Hare airport. We're going to turn it on this month with Accenture's own advertising. So as you walk by all the static, unchanging IBM billboards, you'll be able to get to the dynamic, interactive Accenture interactive network. And if you want to find out more about Accenture, you can. If you want to find out the weather where you're going, you can touch the screen and we'll play the latest weather channel forecast from the city where you're going. If you want to look at the news, you can look at the news. So one thing you'll see very soon is our view of a pretty substantial change in the way indoor advertising will be handled. 

NOT FOR REPRINT

© Arc, All Rights Reserved. Request academic re-use from www.copyright.com. All other uses, submit a request to TMSalesOperations@arc-network.com. For more information visit Asset & Logo Licensing.