Column: The recent resignation of Shai Agassi, president of SAP's product and technology group, took the software world by surprise, causing speculation that his departure would sidetrack SAP's vision and product roadmap. In this guest column, Axel Angeli, SearchSAP.com site expert, asks and answers questions about whether Agassi's departure will actually benefit SAP.
What about SAP in America now?
If there are fears, I can tell those in Palo Alto not to worry. SAP will now, no doubt, take a clearer course, focusing again on application development rather than technology. But there is still a lot to be done; the ESA [enterprise services architecture] framework, especially, needs still to become a reality.
I am convinced developer satisfaction will soar from now on -- both SAP's development teams and customers'. CEO Henning Kagermann, whom I see as the true technical visionary, takes responsibility for all development, and Leo Apothéker, deputy CEO, is the man who drove SAP America's growth.
Wasn't Agassi a major proponent of "globalization," making SAP compliant with international standards?
That is nonsense, and always has been. SAP did not conquer the North American ERP market because it transformed itself into an American company. To the contrary, SAP had its appeal because it introduced the solid and relaxed European style of innovation to America. Customers were intrigued by German "Wertarbeit"
I challenge the allegation that there was a customer demand to transform SAP into a more American company. This alleged demand was created by business analysts, who don't buy or extensively use SAP software.
The stock market reacted negatively, claiming that SAP lost a visionary mastermind.
I would ask: What have Shai Agassi's visions been? In my eyes, it was absolutely insane to bet on Java. Java is the technology of IBM, Sun and Oracle, while SAP's ABAP [Advanced Business Application Programming] runtime machine is by far the best and most reliable execution engine you can find on earth.
The real vision brought forward by SAP, which proved to be the right direction in light of today's SOA [service-oriented architecture] hype, was the BAPI [Business Application Programming Interface] framework. It was a great step forward in direction componentization, a small step for SAP and a giant leap for ERP. When BAPIs appeared in 1998, APIs were already an established standard in Windows but were absolutely new for ERP. But the BAPI concept fell under the aegis of the creator of the ABAP engine, Gerhard Rodé. Under Agassi, the BAPI idea was neglected in favor of some Java-focused new idea, to the horror of customers.
It was also around 1998 when customers started asking SAP to transform ABAP into a genuine object-oriented [OO] language. This led into WebAS, but the result is still half-hearted. ABAP could be a great object language if the money invested in the Java adventure had been put into development of a proper ABAP language parser. Instead, we now have a half-ready ABAP engine and yet another Java engine made by SAP that does not even intrigue the purist Java disciples, who could live with any other alternative from Sun's JEE [Java Enterprise Edition] to the more-or-less free JBoss offer.
Where will SAP go from here?
We need to wait until the new management gets sorted and communicates their strategic goals. But I expect that in the future SAP will concentrate on the essential.
On the application side, the software stack will be broken down into components that allow easy adoption to specialized needs -- a giant challenge that comes with its new A1S strategy to enter the SMB [small and midsized business] market. This will, of course, happen mainly on the ABAP side.
NetWeaver applications like BW, SEM and APO have found their niche, so they will have a life in parallel and will be developed in cooperation with key customers. There is a lot of potential to improve BI, but this will go slowly, as the BI market is fairly closed and highly competitive.
The SOA challenge is even bigger now. ESA is moving too slowly and XI [Exchange Infrastructure] is a weak placeholder for it. I expect that SAP's technology teams can now work quickly on transforming the ABAP engine into a full-featured, message queue-based Enterprise Service Bus. If they succeed in doing this within one year's time, it will be like an afterburner for SAP's business. In combination with the already existing domination in the ERPO market, ESA will make the SAP ABAP engine the premium application platform for nearly everyone.
Axel Angeli is EAI development mentor and SAP project manager for Bruehl, Germany-based Logos! GmbH. He has more than 20 years of experience in the industry as a developer, author, conference speaker and consultant.