Bloomberg Anywhere Remote Login Bloomberg Terminal Demo Request


Connecting decision makers to a dynamic network of information, people and ideas, Bloomberg quickly and accurately delivers business and financial information, news and insight around the world.


Financial Products

Enterprise Products


Customer Support

  • Americas

    +1 212 318 2000

  • Europe, Middle East, & Africa

    +44 20 7330 7500

  • Asia Pacific

    +65 6212 1000


Industry Products

Media Services

Follow Us

Bloomberg Customers

Businessweek Archives

Ibm Rides Into Microsoft Country

Bits & Bytes


Bill Gates thought he had a golden opportunity. Seeing IBM's software strategy in apparent disarray, Microsoft Corp.'s chairman set up a May 26 meeting with James A Cannavino, now a key technical adviser to IBM Chairman Louis V. Gerstner Jr. The Microsoft chairman was hoping that Cannavino, who as head of IBM'S personal-computer division terminated IBM's rocky 10-year relationship with Microsoft in 1990, would be ready to forget the bad blood. IBM's internal software efforts were so disorganized and ineffective, Gates surmised, that he might just persuade Cannavino to once again embrace key Microsoft products, including Windows.

At the last minute, Cannavino canceled. But the meeting might not have gone as Gates planned anyway. What he didn't know was that after months of soul-searching, IBM is set to embark on a fresh bid for the desktop software market that Microsoft now dominates. The decision sprang from recommendations of a task force that Gerstner set up to benchmark IBM's lagging PC software business against Microsoft. Tellingly, the task force, one of several to assess IBM units, became known as the "seminal task force," say company insiders.

The team, headed by John M. Thompson, an IBM senior vice-president and group executive, reported to Gerstner in early May and outlined several options. They ranged from simply giving up on the critical but frustrating PC-software market to "playing to win"--getting superaggressive and trying to unseat Microsoft. Gerstner's choice: a modified play-to-win strategy that will rely on internal IBM efforts, the help of partners, and possibly acquisitions to get into applications software.

The cornerstone of the PC software effort--and for shoring up IBM's entire software business--is a high-stakes gamble on what the company is calling Workplace technology. The idea is to consolidate all software development around a common "microkernel," a form of software even more basic than an operating system. A single microkernel would be used across different product lines and, with the addition of "personality" software, would mimic existing operating systems, say the OS/400 of IBM's AS/400 minicomputers. Workplace would also include a common graphical interface and a shared set of object-oriented programming languages.

The Workplace strategy makes a full-scale rapprochement with Microsoft unlikely. Indeed, it reflects an opposite impulse: Gerstner's urge to create a more cohesive IBM that reintegrates nearly autonomous business units and encourages them to rely more on IBM technology. He has already brought the renegade IBM PC Co. back into the fold. And he has pushed all IBM units to build new designs around the PowerPC chip and send them out the door with more IBM software.

DUPLICATIONS. Gerstner's long-range strategy is now crystallizing: Up and down the line, he wants IBM to standardize on the PowerPC as basic hardware and Workplace as the core software. By doing so, IBM may solve a persistent problem: eliminating the huge incompatibilities across its product lines. To that end, Thompson is now prepared to reallocate resources from IBM's cash-cow mainframe software business and other software projects to Workplace.

Right off the bat, sharing Workplace technology could save IBM lots of money. Today, IBM creates each feature, say a file system, several times--once for each operating system. That duplication contributes to the $1 billion that IBM, by some estimates, spends on operating-systems development each year. And the Workplace strategy could give the company a badly needed, coherent software marketing message.

To marshal forces behind Workplace, IBM on May 27 was expected to announce internally a major reorganization of its software-development groups. Says one executive: "You can expect to see the same kind of focus on Workplace as we've put on the System 360," the mainframe architecture that fueled more than two decades of enormous profits for IBM.

IBM even has hopes of making Workplace popular with other computer and software makers. The company is offering to license Workplace technology and says it has some nibbles. Meanwhile, it plans to ship the first products with the Workplace microkernel--test copies of OS/2 for PowerPC--in June.

While Workplace makes a lot of sense, it will be difficult to pull off. "It sounds good on paper, but let's wait and see what it looks like in the flesh," says Stephen Smith, a PaineWebber Inc. analyst. First, it could be years before IBM can create a single microkernel that works on everything from PCs to mainframes. And even if it can, there are questions about how fast the resulting programs will run. In the meantime, IBM could lose vast chunks of its $11 billion software empire, still tied largely to mainframes. With mainframe sales declining, "those revenues are at risk," says Smith.

JUMPING SHIP. The idea of getting the PC industry to embrace Workplace also might be a bit of a stretch. Ever since it handed over the PC operating system to then-partner Microsoft in 1980, IBM has been in a catch-up mode. In recent years, Microsoft's Windows has sold 50 million copies, compared with 5 million for IBM's OS/2. And Microsoft is tenaciously going after the market for server operating systems, software that allows powerful PCs to serve up programs and data to networks of desktops. OS/2 has its firmest foothold in servers, and Microsoft's Windows NT has not been a big hit. But industry watchers expect NT to catch on eventually, perhaps beginning with an update due later this year.

In the near term, IBM plans to counter with new versions of OS/2. Some will be scaled-down versions for portable PCs and for buyers who don't need all the bells and whistles. One will be aimed at large-scale servers, including parallel processing machines using many microprocessors.

None, however, will solve OS/2's biggest handicap: There aren't many applications programs, such as word processors and databases, that work with OS/2. Without them, few customers will ever discover the nifty features, such as pen and speech input, that are built into the operating system. "We need to show the end user the value of the plumbing," says Leland Reiswig, president of IBM's Personal Software Products Div. The lack of applications has led some key customers to jump ship. Sears, Roebuck & Co., one of IBM's biggest accounts, had been a big OS/2 user. Now it plans to switch all of the OS/2 computers in its merchandising unit to Windows starting this summer.

To get the applications it needs, IBM is rolling its own. The company's Software Solutions Div., run by Steven A. Mills, in the past year has come out with a half-dozen groupware applications-such as Ultimail, a multimedia mail program, Time & Place/2, a scheduling program for groups, and Visual Age, for building client/server programs. IBM also plans a database to compete with Microsoft's Access and FoxPro databases.

Partners can help, too. IBM is expected to seek more allies, such as Lotus Development Corp., which sells versions of its popular Notes groupware and SmartSuite office programs for OS/2. And Mills says IBM is not ruling out acquisitions. He won't comment on specifics, but the company is believed to have Borland International Inc. on a list of possible targets.

Another new effort is aimed at consumers. In a software store you have to look pretty hard to find OS/2, much less applications for it. "Customers turn to us to knit their departments together, but on one-by-one sales," IBM has less presence, says Mills. So IBM is gearing up for a big push into stores, he says. And by next year, PSP hopes to have hot new products, such as software for interactive TV.

For now, IBM's decision to focus on Workplace raises questions about other efforts. The fate of AIX, IBM's version of Unix, is now up in the air. No AIX "personality" is planned for the Workplace microkernel. Another unknown: the role of Taligent, IBM's pricey joint venture with Apple Computer Inc. Taligent's original mission was to create a new object-oriented software environment for mid-1995. Now, IBM plans to add bits of the Taligent technology to programs such as OS/2 and create a Taligent "personality" on the Workplace microkernel.

LATE HOUR. But Taligent is late for a critical IBM deadline: The computer maker plans to release a version of OS/2 based on the Workplace microkernel in early 1995 and wants to include some Taligent object-oriented programming. Taligent says it won't be ready. "If we miss that window, we'd have to question the Taligent investment," says one IBM exec. The answer could be painful: IBM just ordered sharp cuts at Kaleida Labs, its unproductive multimedia software venture with Apple.

Ultimately, IBM's biggest problems come back to Microsoft. By yearend, it plans to ship Chicago, a major update of Windows. OS/2 runs current Windows applications. But because its technology-sharing agreement with Microsoft expired September, IBM must do its own programming to make OS/2 compatible with Chicago. What's more, Chicago will nullify many of the advantages OS/2 now has. Says Rick Sherlund, an analyst with Goldman, Sachs & Co.: "When Chicago ships, OS/2 becomes superfluous.

Despite Gerstner's continuing efforts, some factions within IBM would agree. The IBM PC Co., for example, ships millions of copies of Windows-and relatively few copies of OS/2. "Some groups are putting more resources into supporting Microsoft than IBM," grouses one development manager. "Now, Gerstner is trying to say: 'Meet IBM's goals, not just your own." But if IBM's top goal is selling what the market wants and quickly restoring profits, maybe Bill Gates can find a receptive ear after all.Amy Cortese in New York

blog comments powered by Disqus