Microsoft apps all business

Its Business Application Developer's conference aims to attract top developers to become an enterprise-class systems supplier.

Mike Ricciuti
Mike Ricciuti Staff writer, CNET News
Mike Ricciuti joined CNET in 1996. He is now CNET News' Boston-based executive editor and east coast bureau chief, serving as department editor for business technology and software covered by CNET News, Reviews, and Download.com. E-mail Mike.
4 min read
LAS VEGAS--The only way for
Microsoft to extend its reach into the upper levels of corporate computerdom is through the front lines of business application developers.

The company today delivered its message of business application development and integration to more than 3,000 mid- and high-level business system developers here at a first-of-its-kind Business Application Developer's conference.

Microsoft also used the event as a venue for disclosing additional details on its overall development strategy, namely new and evolving Component Object Model (COM)-based frameworks, including COM+ and two new frameworks, Storage+ and Forms+, for making its underlying technological plumbing easier to use and understand.

The conference was conceived as a way for Microsoft to catch the attention of the top developers and systems integration specialists that it needs win over to become an enterprise-class systems supplier on the order of IBM, Oracle, and other vendors long trusted by the Fortune 1,000. Roughly half of the audience here has never attended Microsoft's other annual developer conferences.

While Microsoft has been wildly successful selling Windows development tools, company executives admit the giant has not equaled that success in the upper echelons of corporate IS, where the decisions on long-term technology adoption are made.

The central reason for this is deceptively simple: "Historically, we have not done a good job telling people how to use our technology," said James Utzschneider, an evangelist in the company's development tool division.

Basically, Microsoft got its start building and selling end-user applications. So, as it has moved into the business server software area, the company took the same tried-and-true approach--outdo competitors on feature lists and individual product releases.

Now Microsoft will spend much more time explaining to customers the types of applications that they can build with its software. That's a big win for time- and budget-constrained customers who would rather build business systems than back-end technology, and for third-party vendors that Microsoft is using to showcase applications that make the best use of its products.

The new message from Microsoft will sound familiar to IBM customers: "We're taking much more of a solutions focus," said Utzschneider.

At the conference, Microsoft is handing out CD-ROMs packed with example application source code and other information to give developers a jump-start on system design.

The other main event here is the unveiling of several new pieces of the company's Windows DNA architecture.

Windows Distributed interNet Applications (DNA) architecture, first laid out by Microsoft a year ago, has been an ill-defined scheme intended to move Windows from the realm of operating system into full-fledged Internet-aware development platform.

But the company has offered little detail about its DNA plans until today.

At a late-afternoon keynote, David Vaskevitch, Microsoft's chief technology architect, will announce the new COM frameworks, and demonstrate for the first time COM+, the next generation of Microsoft's distributed component development technology that forms the base of its most important products, including Windows NT and BackOffice.

COM+, along with newcomers Storage+ and Forms+, are the most tangible pieces of DNA yet offered.

Like its biological namesake, Microsoft's DNA is meant to provide some basic building blocks. Windows DNA is intended to allow business application developers to spend less time poring over the details of software development and more time building business systems.

COM+ includes all of the company's core component technologies, and is beginning to subsume many related technologies, such as transaction processing and message queuing. The first beta-level COM+ technologies are included with the latest beta release of Microsoft's Windows NT Server 5.0, said Vaskevitch. "Developers feel they have to decide between technologies like message-oriented middleware and procedure calls--one or the other. COM+ figures out which is best to use for a given application."

Storage+ builds on low-level data access APIs, such as OLE DB, ODBC, ADO, and others. It simplifies the task of writing business applications that can pull data from a variety of sources--relational databases and data typically stored in file systems, such as Office documents and other client applications.

And Forms+ is an attempt to simplify client-side application development and deployment. "Right now, developers must decide whether to build a client user interface as either Web pages or as [Visual Basic] code. It's an either-or situation now and it doesn't need to be," said Vaskevitch.

Vaskevitch said some of the DNA technologies are already available, and more are coming soon with the impending releases of SQL Server 7.0 and Office 2000. He would not place specific time frames on Storage+ and Forms+ deliveries.

When the three DNA "engines"--COM+, Storage+, and Forms+--debut, they could radically simplify the lives of business programmers, who must learn ever-more complex APIs for each new project, said Vaskevitch.

"If you asked Microsoft what its API strategy was, we'd say Win32 [the underlying Windows programming interface]. But that's for systems developers, not business developers. It's complicated--more than 8,500 APIs--and hard to learn. DNA is aimed at business developers."

Analysts in the past have cautioned that DNA may be no more than the latest marketing spin on the company's existing COM technology. Other marketing schemes that attempted to make COM more palatable to business developers have failed. Most notable is the company's ActiveX thrust, which attempted to define a development environment to encompass corporate client-server and Internet deployment. The company has since scaled back ActiveX, and it still exists as a moniker for the company's Windows-specific components.