Service comes first
Friday, March 18, 2005 10:30 AM
 |
Streibich's goal this year is no different. The ambitious German who previously served as deputy chairman at T-Systems, is aiming for a 10 percent revenue upswing for Software AG by the end of 2005. And he plans to do so by hopping on the bandwagon that several others in the market are already on--services-oriented architecture (SOA).
His company recently forged a partnership with Fujitsu to jointly develop and market a SOA initiative. This effort, he said, would help break up chunky applications into smaller components which can then be wrapped in Web services and made available via the Internet.
In an interview with CNETAsia, Streibich elaborates on this vision, shares his thoughts on the slow uptake of Web services and explains why the modernization of legacy systems is so important. He also clarifies why he thinks software development should not be regulated.
You've taken over as Software AG's CEO for over 18 months now. Has everything gone the way you want it to go? And what are your priorities for this year?
It has gone the way I wanted, and the priority for this year is growth. We want to go back into growth mode.
When you say "go back into growth mode", has that slipped a little more than you like?
Software AG's (revenues) shrank in 2002 and 2003. Our goal was to stop the shrinking and we achieved that in 2004. When you have a decline, you need to first stop the shrinking and then go back into growth. That is the priority this year. After the Internet bubble burst, IT spending declined significantly, especially in the services segment. That cost us a revenue decline. The key point is we achieved growth this past year, and we want to prepare the portfolio for a ten percent growth next year.
To do that, we've identified three areas of focus. First, we must develop our portfolio. Second, we must better utilize our geographical presence. And third, we have to make more business through our partners, be it in development, sales or marketing.
Can you elaborate a little more about your Fujitsu partnership?
We will jointly develop an offering which is a model-based SOA. This will provide a very flexible way of creating such architectures. For instance, the classic IT infrastructure is monolithic. You have your application systems, non-integrated systems and other various systems.
With SOA you're breaking up monolithic applications into application components. You then wrap them in Web services and make them accessible through the Web. You also have the possibility of automating your processes, and reusing those applications in a flexible way. This is a strategic move for us because probably 80 percent of chief information officers worldwide are defining their new IT strategy based on a SOA.
Speaking of Web services, I met your predecessor and Software AG's former chairman and CEO Erwin Königs back in 2001. He had predicted then that XML (Extensible Markup Language) would be central to all businesses in the next three to four years. Now that it's 2005, do you think we've reached this stage yet?
I think he had been absolutely right with his statement. The point here is, timing. It took a little bit longer for XML to really take off.
Timing is also a critical success factor for business. That is the only problem. When you consider that Microsoft will launch its new operating system, codenamed Longhorn, in 2007, we won't have an XML kernel before that point. And now, IBM is also planning to launch a new combined database with SQL and XML functionalities. For that reason, he had been right. It was the right strategy to focus on XML. However, the strategic direction to build it up is dependent on market demand, and the market demand came later because the big players did not really jump on it (at that time). And the big payers have to be in the market to create demand.
We, as a relatively small company on a worldwide basis, cannot make the market as Microsoft and IBM can make the market. The basic technology and standards (for XML) are there. But they have to be implemented in key systems in the market like IBM's database and Microsoft's operating system. And then on that, we can build applications.
It has been said that flimsy software development is the cause of security loopholes in software products. Do you agree? Do you think we need better regulation in software development?
I don't think it should be regulated. Whether it is precise enough or not, well, software development is almost an art because it is a very specific, creative process...like drawing a picture. If you regulate drawing pictures, I don't think that makes any sense.
One way to ensure no errors is certification. We certify all our software guys, such as with ISO certification. And you must have the right deployment process for software and include tests and pilots with customers. The implementation is more important than the coding itself.
I don't think there's any real crisis or a underlining problem in the software industry. However, it is a complex creative process with a lot of dependencies and so on. And for that reason, complex software will never be error-free because there are infinite routes for certain commands where you are never able to test all of them.
The software will become more stable as it gets more mature in its lifecycle, and that's the reason Software AG's Adabas (database management system) and Natural (enterprise application server) are so stable, reliable and highly regarded by customers. Because during their lifecycle, we continuously improve the quality of Adabas and Natural…that's why a certain level of maturity is good for the end users and the software.
There are currently no comments for this post.