CMP -- United Business Media

Intelligent Enterprise

Better Insight for Business Decisions

UBM
Intelligent Enterprise - Better Insight for Business Decisions
Part of the TechWeb Network
Intelligent Enterprise
search Intelligent Enterprise



Email article


One Word: Services


A maturing software industry is losing its product orientation

by David Stodder

Every generation has its coming of age movie. Typically, the main character, just out of high school or college, is sorting through his or her dreams, ambitions, and rebellious instincts. At the same time, he or she must come to terms with the establishment, which runs according to a secret code that the protagonist can't crack - or perhaps doesn't care to. As we head into the new century, the software industry is coming of age. It is turning away from some unrealized dreams of the past - the world according to objects, perhaps even the "New" Economy. It is on the threshold of a mature relationship with the "establishment": a.k.a., a business world that has existed for generations, if not a millennium.

The Graduate is still my favorite coming of age movie. Directed by Mike Nichols, the 1967 hit struck a nerve with the baby boomer generation. A young Dustin Hoffman plays "Ben," a disillusioned college graduate, brooding about his future as Simon and Garfunkel provide the musical leitmotif. Anne Bancroft, as Mrs. Robinson, leads a great cast of character actors who bring to life the 1960s Southern California social scene: booze, cigarettes, swimming pools, dark glasses, and an uptight sensuality moving in and out of the shadows.

Ben - pure of heart -is ready to jump out of his skin at his parents' poolside party when one of the establishment partygoers pulls him aside. In a memorable scene, a grizzled shark guides Ben to a private spot and then stands face-to-face with him. He delivers the secret code of success: "One word: Plastics." Repelled, Ben rushes away. We never find out whether he ultimately adopts this secret; after all, Ben has more burning issues on his mind. But if I could ever so deftly update this scene to some software industry pool party in Atherton or Los Altos Hills, the word delivered by our grizzled wise man would be service.

Chasing romantic dreams of idealized worlds for 20 or 30 years, the software industry largely ignored this essential aspect of its relationships with business customers. Software was first a throw-in to sell iron, then a product, preferably shrink-wrapped. Service usually came with a big price tag, or at least a 24-hour call center number. A few vendors have received high marks for service, but many prefer to point inquiring minds toward more "interesting" subjects.

The New Economy brought to a climax the notion that the rules of business, commerce, and customer relationships would be rewritten by software. The big players in banking, insurance, media, and other industries feared for their lives; software companies seemed capable of anything, including crossing the line to compete in their vertical industry domains. Meanwhile, real service organizations, such as Andersen Consulting, EDS, and more trendy outfits like Computer Technology Partners reaped huge profits from integrating all the software. Full service vendors, namely IBM and Hewlett-Packard, also did well.

But now, with dot-coms across the e-commerce spectrum struggling and the slumping PC industry pinching desktop application sales, the software industry is falling in line behind a new mantra: "software is service." The customers are not just application service providers (ASPs), enterprising IT shops, and tech-savvy line-of-business super users: The idea is to go mano-a-mano with business users themselves by providing portals to online services that require "no software." Working with emerging standards such as XML and SOAP, vendors themselves propose to support drag-and-drop interactions on a massive scale.

Oracle CEO Larry Ellison would have to be considered the pied piper of this movement; he could play the "oracle" of wisdom encountered by "Ben" in my Silicon Valley version of The Graduate. But Ellison is clearly not alone. Recent speeches by IBM's Lou Gerstner, Sun Microsystems' Scott McNealy, Microsoft's Steve Ballmer and Bill Gates, and others demonstrate a shared vision. Ellison has been determined, first, to loosen Microsoft's PC-centric grip on the software industry by presenting an architectural alternative; and second, to position his company to reap the high-margin rewards enjoyed by the established solution providers.

The overall result is a strong move away from viewing software as product. The product-to-service reorientation is a tectonic shift - and one no easier for companies in the high technology arena than for businesses in other industries. The shift will rival all the major technology "discontinuities," such as the moves to client/server architectures and then the Web. And not every company will make it.

At the enterprise level, we need only look at Xerox Corp. to see what can befall a technology provider that misses the shift. Xerox is fighting for its life, not simply because desktop printers and email attachments have rendered its products irrelevant; Xerox met many of these challenges with a rather awesome piece of technology, Document Centre. The company's bigger problem is that it has not effectively reoriented its sales force toward developing full-service relationships with its customers. Ellison, of course, is hoping that Microsoft, reliant on its product-focused hegemony over desktop operating systems and applications, also won't make the shift. But we shouldn't take it for granted that it will be easy for Oracle, either.

Component Services, Not Software

To use Roger Sessions' plain-spoken definition (from COM+ and the Battle for the Middle Tier, Wiley Computer Publishing, 2000): "A software component is just a blob of software that can do something for you." Coming out of the street battles of object-oriented (OO) programming, component technologies have always been more about packaging than implementation. The notion of components offers developers an opportunity to "abstract" their way out of the "partisan bickering," to borrow our new president's favorite expression, which still ripples the OO world. Through components, developers can seek the correct level of integration between "blobs" of software.

Ultimately, the component notion leads to the move toward software as services that "do something for you." The trick, however, is to use this notion to rise above competing software component models based on programming languages (such as Java), application interconnections (such as CORBA), and wiring (such as COM). At the heart of its .Net initiative, Microsoft seems to be trying to resynchronize the frayed software development community around a framework that unifies shared concepts of OO programming and components. My mind is being shaped heavily here by an excellent column written recently by Bertrand Meyer, president of Interactive Software Engineering, in Software Development (November 2000). He writes: "An OO enthusiast like me is likely to see .Net as the latest vindication of the observation that today there is no way to escape using the OO model as a foundation for serious software engineering."

Needless to say, Microsoft's chief rivals are a tad less charitable. Ever playful, Oracle refers to its 9i Dynamic Services as ".Now." Announced in December, Dynamic Services and related tools exploit what Oracle considers standard open-systems technologies, enabling services developers to build data-driven Web applications, portals, and content aggregators right now, not once Microsoft delivers the bulletproof .Net object model and key editions of its software servers. In case you've ever wondered what happened to the developers of Sedona, Oracle's once-imminent OO application development environment shelved by Ellison a few years back - well, they have been hard at work on the component architecture and tools to support the services approach.

Yes, this does sound like we're embarking on yet another chapter in the 100 Years War between Microsoft and Oracle. But we shouldn't forget other significant contestants. Sun Microsystems recently announced a new thrust that positioned its servers, storage, Solaris operating system, and Java-based application development tools for services. IBM has been repositioning its Websphere-centered suite toward services development.



Rate This Article

Comments:

Optional e-mail address:

Services for Services

How will companies manage an increasing portfolio of e-business oriented services? In the view of Candle Corp., IT will need systems management and application integration tools built around a service-oriented architecture (SOA), such as Candle's recently announced CandleNet eBusiness Platform. According to Candle, an SOA is process centric, whereas current enterprise application integration (EAI) solutions, even if component based, are program centric. In other words, Candle, like Oracle and Microsoft, is using the component notion to lift integration to a more facile level of abstraction. In Candle's view, an SOA answers time-to-market challenges by allowing companies to reposition "legacy" systems for new enterprise applications without getting lost in forests of integration programming.

It will be interesting to watch how other tools vendors respond to the software industry's coming of age. IT should view the revolution with healthy suspicion. But organizations shouldn't hesitate if the ROI equation is more in their favor.


David Stodder (dstodder@cmp.com) is Editorial Director of Intelligent Enterprise.

RESOURCES

Candle: www.candle.com
Microsoft: www.microsoft.com
Oracle: www.oracle.com


PRIVACY STATEMENT

Copyright © 2004 CMP Media Inc.
ALL RIGHTS RESERVED
No reproduction without permission