In-Depth

Decoding Bill

As the mantle of leadership at Microsoft shifts, what is Gates' legacy as a developer?

"Nerdy Computer Geek Writes Code, Makes Billion$!" Thus runs the stereotype of Microsoft's soon-to-retire founder Bill Gates. Like all clichés, it has a grain of truth but falls short of the reality that is Bill. Candid conversations with colleagues, associates and competitors leave no doubt that Gates is a more versatile, complex, interesting and, yes, even well-rounded person than the conventional caricatures of him.

As he prepares to transition from his day-to-day role as Microsoft's leader, much will be said in the coming weeks as Gates exits the stage. Some programmers have conflicting recollections of Gates' contributions to modern-day software development, but no one disputes his legacy as an engineer, technologist and architect of today's computing frameworks. Gates gave developers his final perspectives at last week's Tech-Ed Developers 2008 conference in Orlando, Fla.

Gates' roots in development date back to his teenage years, when he started out writing a fair amount of assembly language code. As a young adult, Gates and his childhood friend, Microsoft co-founder Paul Allen, wrote a BASIC interpreter and system tools in the late 1970s. But by his own account, Gates hasn't written code for any shipping product since IBM PC BASIC (1981) and the Tandy Model 100 (1983). As a result, hundreds of thousands of programmers have significantly more direct programming experience than Gates does.

An Engineer's Instincts
Dan Bricklin, president of Software Garden Inc. and co-creator of VisiCalc, the spreadsheet program that became the first "killer app" for personal computers starting in 1979 on the Apple II, remembers talking with Gates at a trade show at the time. Bill told him that he didn't really code much anymore. "He seemed sad about it," Bricklin says.

Gates' conventional reputation as a crack coder is not easy to verify directly. "The thing is," Bricklin recalls, "a lot of the people I know that were close to Bill didn't program with him or weren't programmers enough to evaluate him." Bricklin says he never studied any of Gates' code himself except for a few brief samples, not enough to form an opinion on it. But he has high regard for Gates as a technologist.

Dan Bricklin "Because of his technical background, he has instincts that are an engineer's instincts about the technology. He knows where to look, he understands the medium that he was working in, and he has the business instincts to put them all together."
Dan Bricklin, President of Software Garden Inc. & Co-Creator of VisiCalc

"I don't really know what he worked on, but I know that he's looked at a lot of code," says Bricklin. "He's been involved in an awful lot of projects and technologies and hardware systems. So, while his experience as a coder may not be as extensive, because he hasn't put the time in, given the way he managed his developers, he's had to get very deep in the weeds. So he's exposed to a lot of issues that you normally would only know if you were programming a lot."

Bricklin adds: "Because of his technical background, he has instincts that are an engineer's instincts about the technology. He knows where to look, he understands the medium that he was working in, and he has the business instincts to put them all together."

This view is shared by many of Gates' early competitors. C. Wayne Ratliff, author of dBASE -- the first successful database program for personal computers -- is among those who has high regard for Gates' technical prowess, though he met Gates in 1983 and was not particularly impressed at the time.

C. Wayne Ratliff, Author of dBASE "It sure seemed to me that Microsoft and Bill Gates had a good, firm understanding about the business they were really in -- the people, the market, the product and the equipment."
C. Wayne Ratliff, Author of dBASE

"It's not my encounters with him that give me my greatest impression of his technical acumen," says Ratliff. "It's the way he runs the company -- or ran the company -- with the understanding of the business he was in. He understood software and he understood computers, and he understood the people that use computers."

Today, Ratliff, who is mostly retired, putters around programming microprocessors for racing his sailboat. He recalled his own experiences working at Ashton-Tate Corp., the company that acquired dBASE. "There was a relatively rapid progression from the early guys like George Tate and Hal Lashlee and Dave Cole who understood computers to the Harvard MBAs. They tried to run the company by the book-some MBA book.

"I felt I had a much better appreciation for where software was going than the Ashton-Tate managers did," he says. "But it sure seemed to me that Microsoft and Bill Gates had a good, firm understanding about the business they were really in -- the people, the market, the product and the equipment."

Overlapping Windows
Robert Carr, the founder and CEO of KeepandShare, a Web productivity startup, concurs. The former chief scientist for Ashton-Tate, creator of the Framework-integrated office suite and architect of Go Corp.'s PenPoint OS, Carr competed with Microsoft over several decades. A veteran, too, of the Xerox Palo Alto Research Center (PARC) Advanced Systems Department in the late 1970s, Carr is eminently suited to assess Gates' use of the PARC ideas at Microsoft.

"He offered better products to the market sooner than anyone else," says Carr. "Back in the '80s he won a lot of application market share because they produced a superb product and they had the vision to bring Xerox PARC technology to the market as soon as they could."

Robert Carr "Bill combined the notion of copying the Xerox PARC overlapping windows idea but implementing it on a business model that licensed the operating system to a hundred different hardware companies."
Robert Carr, Founder and CEO of KeepandShare; Former Chief Scientist at Ashton-Tate Corp.; Creator of the Framework Office Suite

Gates hired Charles Simonyi away from Xerox PARC, where he had created Bravo for the Alto PC, in 1981. "Charles started building WYSIWYG word processing on top of MS DOS," recalls Carr. "That's how Microsoft Word started out. Not an easy task to do. But they did it, they did a pretty good job, and they kept iterating and improving it. That turned into the Microsoft Word franchise. Then, because Lotus was slow to implement its [1-2-3] spreadsheet program on the Macintosh, Microsoft was able to get a beachhead by having the premier spreadsheet on the Macintosh in the '80s. So when Windows came out, they were able to port that technology to Windows, beat Lotus to the punch and start grabbing a lot of spreadsheet market share on the PC."

Carr also believes that Microsoft did a better job than its competitors of bringing a windowing OS to the masses. "Bill combined the notion of copying the Xerox PARC overlapping windows idea, but implementing it on a business model that licensed the operating system to a hundred different hardware companies," Carr says. "That's what the market needed and wanted. So I think that Bill's legacy really was bringing the graphical user interface to the mass market both at the operating system level and at the application level."

Sport Coders
Ironically, one of few people who actually did have an opportunity to see and to work with Gates' code is not as enthusiastic about Gates as a programmer as some Microsoft competitors. Jeff Angus, now a management consultant and baseball writer, was once known as "Mr. BASIC" at Microsoft where he was hired in 1983 to write business programming examples and documentation for MS BASIC. He wrote 13 programming manuals for Microsoft in 15 months. Some of the earliest material he worked with was the collection of BASIC programming examples that had been written by Gates and Allen. Angus was not impressed.

"Bill is capable of hyper-focus," says Angus. "He was very good at a very narrow, intensive lone-wolf sport. Kids brought up today who have that set of personality traits and who are very intelligent the way Bill is end up playing video games instead. But the game for Bill, I believe, was coding -- faster, better, smaller -- in a world where smaller and faster really meant something. He was a sport coder."

According to Angus, documenting code and what people did with the end product weren't Gates forte back then. "The original BASIC documentation and programming examples written by Paul and Bill were crappy," he recalls. "Good code is not as small as you can make it, but as self-documenting as you can make it. They had no ethic of self-documenting variable names or commenting or doing anything that clarified the code for a maintenance programmer, for someone who did this professionally. So in that sense, Bill and Paul were not professional programmers. They were monstrously gifted, clever, sport coders. But they weren't the kind of people that a Redmond Developer reader would hire to be on their team."

Angus' view, though, is clearly in the minority. Anders Hejlsberg is a Microsoft technical fellow, distinguished engineer, key architect of the C# language and major contributor to the development of the .NET Framework. He is also the key architect of Microsoft's Language Integrated Query (LINQ) and one of the most respected developers anywhere today. Asked about Gates' coding skills and whether he thought Gates would do well coding in C# today, Anders replies: "I'm sure he could do just fine in either VB or C#. He still writes lines of code. I mean, he doesn't write a lot of them, he doesn't write products or anything, but you'd be surprised on how up-to-date he is on [programming]. Bill is formidable when it comes to that," says Hejlsberg.

No BS
Perhaps the most compelling evidence that Gates is not only a good coder and technologist, but has broader capabilities as well, comes from the founder and CEO of WildTangent Inc., Alex St. John, who remembers his first encounter with Gates. St. John was hired by Microsoft in 1992 at the age of 24 because he'd already acquired a great deal of knowledge and experience with digital printing technology. Just three months after joining Microsoft he was told to handle a press query from InfoWorld. He recalls:

"I was naive at that point. And they put me on the phone with the InfoWorld guy and he asked a lot of questions about the Windows printing architecture. One of the things I said, very openly, was 'Windows print architecture is not that great and needs a lot of enhancements, a lot of work, and there's going to be dramatic improvements to it in Windows 95 and NT ...' And apparently the next day they printed their front-page article saying, 'Microsoft executive, Alex St. John' (I was just middle management really), 'admits that Windows is an inferior publishing platform to the Mac, but claims that maybe it will be fixed in Windows 95.'"

This earned him a scathing e-mail from Gates saying, in effect, "What the heck are you doing, have you never heard of PR?" Convinced that he was about to be fired anyway, St. John replied to Gates via e-mail.

"I wrote an e-mail back to Bill saying, 'You know Bill, I'm sorry about [messing] up the PR thing. You're right about that. I don't have any experience with it and so excuse me but, to be honest, I don't know what you're talking about. Your print architecture is a disaster! Here's why: This is broken, that is broken ... here's why it's inferior."

Gates responded with an e-mail to Microsoft senior management saying, in effect, according to St. John: "This new kid that we hired who, I'm told, is an expert on publishing, tells me you've been lying to me all these years and that our print architecture is a disaster and inferior to the Mac -- which would certainly explain why we get our butts kicked by them. So I want to know why the heck you guys have been BSing me all this time about having a great print architecture when the expert we just hired tells me you guys don't know what you're talking about."

The end result was that, after a series of meetings in which St. John stuck to his guns despite major heat from some senior managers, he wound up with broad responsibilities for re-making Microsoft's printing architecture. This incident shows, as well as any might, that Gates' abilities to astutely manage people to get results are equal to his widely recognized technical skills. And St. John sums up as well as anyone might the essence of Bill Gates' long-term legacy:

"It's absolutely fair to say the guy is a genius. The guy is absolutely brilliant. History may be inclined to remember Bill as a brilliant technologist, but in some sense I think he may be somewhere between Thomas Edison and Henry Ford. History remembers Thomas Edison as a great inventor. They don't necessarily recall that he created the most valuable company in history, General Electric. Henry Ford didn't invent the car, but he's the guy who made the automobile industry. So though Gates was certainly a programmer at some time in his career, I'm not sure the extent to which he was a brilliant software engineer. But he is a brilliant technologist and businessman."

CEO 2.0: Who's on Deck to Run Microsoft?

With Gates retiring and CEO Steve Ballmer stating publicly that he plans to stick around another nine years -- max -- it's not too early ask: Who will be running Microsoft in the next decade? The obvious CEO candidates would seem to be the three Microsoft presidents: Kevin Johnson, head of platforms and services; Robbie Bach, head of entertainment and devices; or Stephen Elop, the successor to Jeff Raikes, who will run Microsoft's business division. I've heard a few folks throw Chief Operating Officer and former Wal-Mart Stores Inc. exec Kevin Turner's name onto the Microsoft CEO 2.0 list. Chief Software Architect Ray Ozzie seems to have little desire to be in the limelight and run the company based on his actions to date. I've also heard a number of Microsoft watchers speculate that Redmond is likely to look outside -- rather than inside -- the company when choosing its next CEO.

Still, there are a number of rising stars inside the company who could rise to the top. If I were to choose the 10 'Softies who are on track to become increasingly prominent and powerful in Redmond, this would be my list, in alphabetical order:

Jon DeVaan, Senior Vice President, Windows Core Operating System Division: A protégé of Windows Chief Steven Sinofsky, DeVaan is in charge of the engineering team that develops the core components and architecture of Windows.

Alexander Gounares, Corporate Vice President, adCenter and Commerce Platforms: Gounares has been leading the engineering efforts for Microsoft's core advertising and commerce platforms, helping to oversee payments, points, subscriptions and more. Given the increasing importance that adCenter and online advertising in general have to Microsoft's future, Gounares should be a key player.

Scott Guthrie, Corporate Vice President, .NET Developer Division: Here's a name we all know well. Guthrie runs the dev teams that build Silverlight, Windows Presentation Foundation, Windows Forms, the Common Language Runtime, ASP.NET, IIS 7.0, Commerce Server, the .NET Compact Framework, and Visual Studio Web and Visual Studio Client dev tools.

Bill Hilf, General Manager of Platform Strategy: From his start as head of Microsoft's Linux Lab, Hilf has had a tremendous amount of influence inside and outside Microsoft on how the 'Softies relate to open source vendors and customers.

Chris Jones, Corporate Vice President, Windows Live Experience Program Management: One of the original "Baby Bills," Jones is still seen as a golden boy at Microsoft. A Windows and IE Group veteran, Jones now does planning, design, usability research and more for the amorphous entity known as the Windows Live Experience.

Don Mattrick, Senior Vice President of Interactive Entertainment Business: Mattrick joined Microsoft at a key juncture for its gaming unit. Days before Mattrick joined the company, longtime Microsoft Xbox veteran Peter Moore announced he was leaving to join Mattrick's alma mater, Electronic Arts Inc. So far, Mattrick has been Ray Ozzie-like in his silence and absence as a visible Microsoft champion.

Brian McAndrews, Senior Vice President, Advertiser and Publisher Solutions Group: McAndrews came to Microsoft as part of the 2007 acquisition of aQuantive, where he was CEO. He leads the newly minted group tasked with building and marketing all of Microsoft's myriad ad platforms.

Marshall C. Phelps, Jr., Corporate Vice President for IP Policy and Strategy: Phelps oversees all of Microsoft's intellectual property (IP) groups -- trademarks, trade secrets, patents, licensing, standards and copyrights. He's been building for Microsoft an IP portfolio for licensing, as he did for IBM Corp. during his 28 years there.

Eric Rudder, Senior Vice President of Technical Strategy: Just a few years ago, Rudder seemed destined to become the new Bill Gates. As senior vice president of Microsoft's lucrative Server and Tools unit, Rudder was in the catbird seat. Rudder is said by sources to now be working on a project under Craig Mundie to design and develop a distributed OS that might or might not be built on the Windows core.

David Treadwell, Corporate Vice President, Live Platform Services: A 19-year Microsoft veteran, Treadwell has been focused on honing Microsoft's various "platforms" throughout his career. Treadwell also helped start the "Windows Live Core" incubation project, which is now known as Live Mesh.

-- Mary Jo Foley

comments powered by Disqus

Featured

Subscribe on YouTube