Committee Of Correspondence

Does Microsoft Play Fair?

The committee: Steve Ballmer, James Fallows, James Gleick, Peter Huber, and Roger McNamee.

Herb Stein Wednesday 6/19/96—8:03 a.m.

The computer is rapidly carrying us into a whole new world of personal and business life. In this new world the Microsoft Corp. is a dominant player. Almost all of the personal computers now in operation or being built run on operating systems produced by Microsoft–MS-DOS, Windows or Windows 95. Almost all new software applications–word processing programs, spreadsheets, CD-ROMs, Internet access programs and others–are made to run on Microsoft operating systems. A large but not dominant part of this software is made by Microsoft itself–Microsoft Word, Excel, and Microsoft Internet Explorer, for example–and the rest is made on license by Microsoft.

This pervasiveness of the Microsoft product has many advantages. It helps the novice to learn the system, it permits easy communication between computers and it assures the availability of a large, varied supply of usable software. But it raises many questions.

Can and does Microsoft use its present dominance to prevent the development of new, and possibly better, operating systems and applications?

Does the method in which Microsoft licenses its operating systems to manufacturers of computers keep other producers of operating systems from competing? In 1995 Microsoft entered into an agreement with the Department of Justice to refrain from certain actions that were alleged to have that effect, but that has not quieted complaints that the methods still in use have the same anti-competitive result.

Does the bundling of Microsoft applications with the Microsoft operating systems give these applications an advantage and inhibit the development of superior applications?

Does Microsoft’s dominance as provider of operating systems enable it to neglect the interest of consumers in the most efficient and reliable use of their equipment?

Is there a possibility that Microsoft’s present position and resources will enable it to dominate the future world of communications as AT&T once did?

On the other hand, supporters of Microsoft point out that it has been a leader in a process that has given businesses and individuals power to write, to calculate, to communicate and to obtain information that was unimaginable even a few years ago. That power has been provided at dramatically-falling prices. New firms are entering the software market every day. Technological advance already in sight may undermine whatever dominance Microsoft now has. The ability to bundle applications with the operating system may lose its advantage when applications may be easily downloaded from the Internet. In fact, the disk operating system may be bypassed altogether, with the user getting everything over the Internet.

How valid are these complaints and claims? What, if anything, needs to be done? During this next week these questions will be discussed by five observers with strong opinions about Microsoft, from every point on the spectrum.

Steve Ballmer Wednesday 6/19/96—8:45 p.m.

It’s ironic in 1996 to see all these old questions about whether Microsoft is “too” successful being trotted out once again. We’ve spent much of the past year answering the opposite question: “Can Microsoft even survive The Internet Tidal Wave?”

Make no mistake about it—we are determined to lead the way in providing the exciting Internet technologies that computer users want. But the mere fact that some people are questioning Microsoft’s ability to survive goes a long way toward answering Slate’s questions.

Microsoft software is very, very popular with consumers. We’re very proud of that, and we intend to build upon it for the future.

Some of our competitors say Microsoft is “too” successful, and somehow that hurts the industry. Yet I can’t think of any sector of the economy—not one—that is more competitive, dynamic and innovative than software for PCs. Look at all the companies cashing in on the Internet gold rush. Twenty-three of the nation’s 100 hottest growth companies are software and computer-service providers, far and away the most of any industry. Look at Netscape—two years ago it didn’t exist, and today it has a market capitalization of nearly $5 billion and tens of millions of customers. Not bad. And not possible in practically any business other than PC software.

All of Slate’s questions boil down to one: “Who should determine success in the software industry?” Millions of consumers making individual purchasing decisions? Or government regulators?

A handful of Microsoft’s competitors would prefer the latter. But that’s not going to happen. Why? Because the software industry is doing a great job for consumers. Innovative products and services are being delivered at such a dazzling pace that some computer users (or IS managers, anyway!) wish we’d all slow down. But there’s no time for that because any software company that stands still for a minute is going to get run over. That’s why Microsoft—and lots of other software companies — are racing like mad to integrate Internet capabilities into their products.

If anyone thinks that Microsoft ought to slow down, stop innovating, or stop competing so hard, they’re going to be sorely disappointed. We are in the midst of the most exciting period in the history of the software industry. We have a mission: to make the Internet (and, yes, intranets) rich, active, rewarding, and easy to use for our customers. Our competitors may not like it, but it’s consumers, not competitors, that we aim to please. We’ve done a pretty good job of it in the past, and we’re sure not going to stop now.

James Gleick Thursday 6/20/96— 7:42 a.m.

Just for fun, let’s start with a single data point—me.

I’ve used software from scores of different companies. Until a few years ago, anyway.

Today my word processor is Microsoft Word. I am composing this e-mail in Microsoft Exchange. I browse the Web with Microsoft Internet Explorer. My spreadsheet: Microsoft Excel. I do have a few non-Microsoft programs-Intuit’s Quicken, for example, hasn’t been swallowed up, thanks to Government intervention. Still, the software I use for viewing graphics, backing up my hard drives, sending faxes, creating Web pages, and just about everything else that matters comes from Microsoft. Quite a testimonial, eh?

I’ve given up software from companies with fine toll-free support so that I can put on my blood-pressure cuffs and enjoy Microsoft’s notorious customer service-the (206) numbers, the labyrinthine phone menus, the Orwellian-trained staff forbidden to admit the existence of Bug Fun.

Why have I lain down before the steamroller from Redmond?

Because life is short. Because I need all my software to work together. Because I need it to work with the software everyone else is using. Because there’s really no choice. No company but Microsoft has the inherent power to make sure its products will function with all the various hardware and all the various standards (Microsoft-owned): for interpreting mail messages or dialing telephones or manipulating graphics.

I’m not alone, naturally. The numbers show that Microsoft now owns (with astounding 90 percent shares) software categories that just a few years ago it had barely entered. And often with lackluster products that won purely because of synergy and muscle.

I’m a true believer in free-market economics. Once in a while, the Government has to find ways to protect free markets-that’s what antitrust law is for. You may not have noticed, but competition has vanished in huge areas of the software business. There hasn’t been a significant entry into (for example) word processing in years-not because the word processor has been perfected, but because Microsoft has locked up the space. In 1996 America, it’s far easier to start an airline than a word-processing software company.

Luckily, there are new categories. We’ll be talking this week about Netscape, I hope, and Microsoft’s strategy for the Internet. …

“Does Microsoft Play Fair?” No, of course not. It is known throughout the industry-and has been cited definitively by the Justice Department-for its unethical and sometimes illegal gamesmanship. But that doesn’t even matter. The company could be staffed exclusively by saints, and the outcome would be much the same. If you own the operating system that runs the world’s PC’s, and if you are allowed to leverage that power in entering new businesses, you have an advantage that the great monopolists of the 19th century could only dream of.

Peter Huber Thursday 6/20/96—3:11 p.m.

Pundits get paid to write all sorts of nonsense about Microsoft, just as Rikki Lake gets paid to interview lesbian headhunters from New Guinea. Antitrust litigants have to put real money on the line. When it comes to Microsoft, none of them do.

Manufacturers of telephone equipment sued Bell a decade ago for closing up interfaces to the telephone network. They won a $277 million verdict. MCI sued Bell for refusing to provide open connections to Bell’s local exchanges. A Chicago jury awarded it $600 million, trebled to $1.8 billion. For every dollar Microsoft’s nefarious conduct cost a competitor or consumer, the victim can recover $3. If he can win.

So why aren’t any litigants lining up to collect? Microsoft has had two brushes with the Department of Justice. Both legally trivial. Both quickly resolved. Apart from that, nothing.

It can’t be that Microsoft is just too big and scary. General Motors, 10 times the size of Microsoft, gets sued by contingency-fee lawyers every day. IBM is far richer than Microsoft, and a direct competitor in the sale of both operating systems and applications software. IBM also knows its antitrust law, having successfully fought off the Department of Justice a decade ago. But IBM hasn’t sued.

The Department of Justice has limitless resources. Anne Bingaman, Assistant Attorney General for Antitrust, is known to be obsessed about Microsoft. She’s dying to bring a “tying” case against the company. But she probably won’t. And if she’s foolish enough to try, she’ll lose.

I’m an antitrust lawyer. Most of the legal work I do concerns “tying” — the legal key to Microsoft’s alleged criminality. I wouldn’t dream of taking an antitrust case against Microsoft on contingency. I’ll bet any member of this panel $1000 that no U.S. jury will award antitrust damages against Microsoft in the next two years. Any takers?

Whatever else it may have monopolized, Microsoft hasn’t yet monopolized plaintiffs’ lawyers. But with only trivial exceptions, the antitrust case against Microsoft remains pure vaporware. New products are announced every day in press releases. But not in court.

Why not?

Roger McNamee Friday 6/21/96—1:06 a.m.

Microsoft plays to win. Bill Gates has populated his company with thousands of smart, hard-working and intensively competitive people. They do not like to lose. They exploit whatever edge they can find. This is normal behavior in a market economy. As a consequence of more than 15 years of such behavior, Microsoft is now the most powerful company in the most important growth industry in our economy.

The questions before us–I believe–are two: has Microsoft done anything that violates the antitrust law and if not, does its current degree of industry control violate some other standard which justifies legislative or legal action? As far as I can tell, the antitrust law never foresaw a situation like the one that produced Microsoft, which is just as well. Our economy is better off for Microsoft’s success.

As to whether Microsoft’s domination justifies legislative or legal action, the answer hinges on a subjective assessment of the “public interest.” Has the public been harmed by Microsoft’s behavior? Is it likely to be harmed in the future?

The company has guided a technology revolution that appears to having an extraordinarily positive impact on our economy and society. The American companies in the industry which Microsoft leads have revenues in excess of $150 billion. The market value of publicly traded U.S. companies is greater than $200 billion. (This latter figure excludes companies such as IBM and Hewlett-Packard, which are major players in the PC business, but whose success predates the industry.) The industry has created hundreds of thousands of jobs. It is hard to see from these figures how Microsoft’s success to date has harmed the U.S. economy or society. There are clearly a number of companies which have competed with Microsoft–and lost–which are unhappy, but such is the nature of our economy. All major industries eventually consolidate. That is how the process works.

Microsoft’s Windows operating system now has 85 percent or more of the market. Its core suite of applications has about a 90 percent share. As far as I can tell, Microsoft got to these market share levels because customers made choices in the open market. Many people argue that such market share levels represent a monopoly. There are still alternatives, however, including the Macintosh operating system on which I am composing this message. It also happens that the electronic mail package I am using is produced by CE Software, not Microsoft.

Looking forward, the answer is less clear. There is no evidence that Microsoft’s domination is limiting innovation in the computer industry. In fact, quite the opposite is true. The Internet, a new paradigm which seemingly treats Windows as just another legacy system, has emerged as a huge wave of innovation and opportunity with little help–or hindrance–from Microsoft. Nor is there a shortage of capital for new technology ideas.

Publicly traded Internet companies have a market value in excess of $20 billion dollars, reflecting Wall Street’s favorable expectations for that new market. Technology-oriented venture capital firms are investing record amounts of capital.

There is no question in my mind that Microsoft wants to dominate the Internet just as it does the desktop. Unlike some others, I see no reason not to let Microsoft compete aggressively for that opportunity. Microsoft is giving customers a choice about how to implement Internet technology that is fundamentally different than the one proposed by Netscape and others. It is a choice that is tightly linked to Windows and Microsoft Office. For all the value that customers have gotten from Windows and Office, they have had to bear huge support costs and administrative burdens. If Microsoft does not convince customers that the benefits of linking Internet technology to the Windows environment justify the incremental costs and administration, those customers are going to buy the Netscape solution.

Those who want to restrain Microsoft through legal or legislative action need to justify their actions with evidence that there is some loss to society from the status quo. As one who uses a Mac–and invests in emerging technology companies for a living–I cannot see it.

Steve Ballmer Friday 6/21/96 12:43 p.m.

Microsoft stands accused of making all its software products “work together” and “work with the software everyone else is using.” Guilty as charged!

Last time I checked, that’s what we’re supposed to do.

I want to pick up on Jim Gleick’s point about customer support. Microsoft’s customer support is the best in the industry. For two years running, Microsoft has won PC World’s World Class Award for Best Service and Support in the software industry, based on a survey of independent analysts, press, etc. Also, a survey of more than 8,000 readers of a popular computer magazine recently ranked Microsoft first place for product support. No wonder, since wait times for customers are on average under a minute, and we spend more than $500 million on support to make sure we get you the right answers right away!

If you don’t want to hold a minute, you can get product support from our forums on MSN, from our fax-back service, from our Knowledge Base postings on our Web site, and many other ways. And if you want to beat up on any Microsoft product, let us know when you call or post your complaint to our Web site–we’ve got a place for that too. We want to hear customer complaints and concerns so we can keep making our products better.

Jim says he has “no choice” but to buy Microsoft products. What is he talking about? Tell that to the millions of Apple Mac users. Tell that to the millions of WordPerfect users. Jim says he uses our Web browser, Microsoft Internet Explorer (glad to hear it). But Netscape daily trumpets the popularity of its browser.

Day in and day out I’m meeting with customers all over the world. I can assure you there are lots of people who aren’t using Microsoft products. (It keeps me busy!) GM, for example, has 55,000 desktops that use a mix of Microsoft, Lotus and Novell products.

Customers have choice precisely because Microsoft works so hard to make sure its products work well with software from other vendors. In fact, that’s one of the main reasons for the success of Windows and other Microsoft products–customers like Windows because there’s lot of great applications for Windows. And there’s lot of great applications for Windows because Windows is an open platform, based on published specifications, and Microsoft provides a lot of support for developers writing Windows applications.

James Gleick 2:48 p.m. Friday 6/21/96

On Friday, June 21, Steve Ballmer wrote, “Windows is an open platform, based on published specifications, and Microsoft provides a lot of support for developers writing Windows applications.”

I’m glad Steve Ballmer said this. There won’t be a more important sentence written during this discussion.

Here’s a tiny case in point–one of hundreds of such cases. Microsoft “owns” the operating-system standards for messaging (mail and so forth), known as MAPI. MAPI is public–but now Microsoft is about to release an altered version, with a dozen or so new or changed items. These changes were made to suit the needs of Microsoft’s own messaging program, Exchange–and by the time competitors get access to the changes, Microsoft’s programmers will have a big head start. No big deal–except that this is how it always works, and in a fast-moving industry, these head starts add up.

In fact, while Microsoft does publish huge amounts of information about how to make programs work with Windows, it also withholds much essential information, making it available only to inside programmers and to friends of Microsoft. It uses the information to make deals, doling it out in exchange for other specific benefits and services.

Steve, when I interviewed you for my article last fall, Making Microsoft Safe for Capitalism you made clear that Windows was “open” only as and when that suits Microsoft. You said: “We could say, hey, we’re not publishing any API’s to our operating system. Or we could pick five guys and tell them what’s in the operations system–we’re not going to tell other people.”

Do you stand by that view now? Or will Microsoft commit to a policy of true openness where the Windows operating system is concerned: agree to publish and document all its APIs, and to make the information freely available to outside programmers as soon as it is available to Microsoft’s own programmers? Not just sometimes–always?

James Gleick 6:30 p.m. Friday 6/21/96

On Friday, June 21, Peter Huber wrote, “Manufacturers of telephone equipment sued Bell a decade ago for closing up interfaces to the telephone network. They won a $277 million verdict. MCI sued Bell for refusing to provide open connections to Bell’s local exchanges. A Chicago jury awarded it $600 million, trebled to $1.8 billion. For every dollar Microsoft’s nefarious conduct costs a competitor or consumer, the victim can recover $3. If he can win.”

Notice these plaintiffs are big telecommunications companies. In the software arena, typical small companies can never afford the multiyear, multi-multimillion-dollar route of civil antitrust litigation. Especially against a company with Microsoft’s tenaciousness and deep pockets. (Read Jerry Kaplan’s Startup for a typical case study of a company destroyed by Microsoft.) That’s why we have a Department of Justice.

That doesn’t mean Microsoft can’t be sued occasionally. Stac caught Microsoft flagrantly violating its disk-compression patents (after Stac refused to roll over for the usual hardball negotiating tactics); sued; hung on, betting the whole company; and won a huge judgment. So Microsoft just bought what it needed anyway, including Stac’s silence (where is the disk-compression software industry today, eh? it is Microsoft)–it just cost more.

“IBM is far richer than Microsoft, and a direct competitor in the sale of both operating systems and applications software. IBM also knows its antitrust law, having successfully fought off the Department of Justice a decade ago. But IBM hasn’t sued.”

Special case, of course. IBM is hardly in a position to be an antitrust plaintiff.

“The Department of Justice has limitless resources. Anne Bingaman, Assistant Attorney General for Antitrust, is known to be obsessed about Microsoft. She’s dying to bring a “tying” case against the company. But she probably won’t. And if she’s foolish enough to try, she’ll lose.”

This is exactly backward. No administration wants to go after Microsoft, the paragon of American success stories–it’s a political disaster. The issues are too technical and fast-changing for DOJ lawyers anyway. (By the time they proved the illegal tactics by which Microsoft killed the last remaining operating-system competition, it was too late–all they could do was close the barn door.)

Are we debating whether the Government will take effective action against Microsoft? If so, I’m with you. They won’t. They don’t want to. They can’t.

Or are we debating whether they should?

James Fallows 10:21 p.m. Friday 6/21/96

To Steve Ballmer: As you point out, it is droll that 1996 should be both the year when the Internet was going to crush Microsoft, and the year when MS was going to crush everyone else. None of us has a clue what the operating-system market is going to look like a generation from now. But I hope that the points you start with are official posturing rather than what you really think. To wit: “Yet I can’t think of any sector of the economy–not one–that is more competitive, dynamic and innovative than software for PCs.”

What the fuss is about, of course, is not “software for PCs” in general but “operating system software for PCs.” When it comes to operating systems, it is hard to think of another sector of the economy as dominated by one company’s products as OS’s are by Microsoft.

Let us assume for purposes of argument that the current dominance is *entirely* due to the excellence of Microsoft’s products and the devotion of its employees. We’ll assume that Win3.1 and Win95 are so much more intuitive and natural than the Mac OSes, and so much more technically impressive and robust than OS2, that customers inevitably preferred them on their own merits. We’ll assume that the build-up for these products in the computer press for the last few years and the general press before last Aug. 24 reflected shrewd, sound news judgment. The industry benefits from having a standard, and we are assuming that MS came to set and control that standard for reasons that are entirely to its credit. The question from this point forward is: given the fact that one company controls the standard, does it make sense in the long run for every other entrant to have to compete against different divisions of that same standard-controlling firm?

“All of Slate’s questions boil down to one: ‘Who should determine success in the software industry?’ Millions of consumers making individual purchasing decisions? Or government regulators? A handful of Microsoft’s competitors would prefer the latter.”

Donnez-moi un break, Steve. A few weeks ago I sat with a bunch of other journalists in Washington and listened to Bill Gates, along with half a dozen other execs from software firms, beg for stiffer government action to penalize software pirates based in China. Suppose some Chinese buccaneer had stood up at that meeting to say: “All your complaints boil down to one question: Who should determine success in the software industry? Billions of consumers making individual purchasing decisions, buying their CDs for the equivalent of $4 U.S.? Or government regulators trying to enforce unfair trade laws?”

If you heard that you would have stood up and said, “But, but, but … the trade regulators aren’t TRYING to make purchasing decision. They are trying to enforce the rules so the competition will be fair.” And that is what is going on in the U.S. market too. NO ONE is suggesting that “government regulators” should or will make purchasing decisions. I can’t believe that someone as sophisticated as you can really believe this. From Dick Armey, yes … or maybe a character in an Ayn Rand novel, but not someone who knows how the agreed-on rules of competition affect corporate strategy.

Civilian competition of every sort involves rules. Adam Smith knew it. Teddy Roosevelt knew it. You know it. The question is not whether rules will exist, and it is of course not whether “government regulators” are going to make purchasing decisions. It is whether the rules of the game now are the best ones for the vigor of the business in the long run.

Teddy Roosevelt is one of two turn-of-the-century figures whose views are relevant in this discussion. (The other is Andrew Carnegie, coiner of the phrase “the man who dies rich dies disgraced.” His name lives now because he spent the second half of his life unloading the assets he built up in the first half.) Roosevelt obviously loved the muscularity of competitive capitalism–but he also loved it best when those animal spirits were constrained by certain rules. These rules preserved certain land for national parks (even though market forces would have dictated otherwise), they forbade companies to compete by hiring children, and so on. TR would no doubt say “Bully! Bully!” about the business you have created. But he would not have thought, as you are suggesting here, that therefore it was vulgar even to talk about competitive rules.

James Fallows 11:23 p.m. Friday 6/2196

To Roger McNamee: Although you’re making a calm and lucid case for not worrying about Msoft, your argument contains a nicely distilled expression of what the exact problem is. You say: “Microsoft plays to win. Bill Gates has populated his company with thousands of smart, hard working and intensively competitive people. They do not like to lose. They exploit whatever edge they can find. This is normal behavior in a market economy.”

This rings true to my own observations of the company, from a far greater distance than yours. In normal circumstances it would be nothing to complain about. Indeed it would be the kind of corporate culture many people would admire and study.

The problem is the coexistence of this give-no-quarter behavior with control of the defacto standard for the industry as a whole. In reality, the Windows API is the standard for PC software. You don’t use it because you have a Mac. I don’t use it because, in addition to a Macs, I do my daily work on several OS2 machines. (I originally got them because the software was so much more robust than Win 3.1, and is still far faster as multi-tasking than Win95. But I recognize that this has become a dotty eccentricity, like wearing a bow tie [which I don’t do].) But we are the exceptions and Windows is the rule.

That is fine EXCEPT that Microsoft officials seem to view this standard as part of the “whatever edge they can find.” Anyone can think of nickle-and-dime examples. I recall the charming way that, whenever I load some kind of Microsoft app on my OS2 computer, it detects an alien operating system, warns me about the grave danger this creates to my computer, and helpfully offers to reformat my disk. (I can substantiate this exact sequence if someone wants–it happened about two years ago, and I told Bill Neukom, the MS counsel, about it at the time.)

The glossy magazines of the computer press have on the whole been shameless log-rollers for Microsoft, starting to build a “We Can’t Wait! It’s Almost Here!” atmosphere for Win95 about two years before its second-coming-of-Jesus unveiling last summer. Like second-coming predictors of the normal religious sort, they soon started saying, “Well, the REAL excitement is still ahead, and it’s NT 4.0 and Win96!” Although the editors of these publications squawk that they are deeply offended at the idea that are being boosterish or partisan, it looks from the outside as if Microsoft considers the computer press part of the “edge they can find.” Jerry Kaplan’s book “Start Up”–one of the most enjoyable tech books I’ve read in a long time–is full of retail-level illustrations of “edge.”

Again this would be fine EXCEPT that this anything-goes philosophy is applied by the people who also happen to control the standard. Let’s assume that Roone Arledge, of ABC News and Sports, also “plays to win.” He has been the most successful innovator in modern TV news, and has the biggest rating. But suppose that a man with his tremendous competitive edge *also* had some control over the broadcasting technology that CBS, NBC, and CNN had to use. His nature being what it is, he’d want to use that edge. Bill Clinton “plays to win” in politics. Suppose, in addition to all the powers he has from incumbency, he were in charge of distributing money from the Federal Election Commission. Meaning no disrespect to his honesty, , he might be tempted to see this as an edge–or at least his ambitious underlings would.

And so with MS. A company this seething with vital forces is bound to create problems if it controls a standard that really should not be used as an “edge.” It is possible that some sudden tech change will make the issue moot, by removing WIndows from its “standard” niche. But as long as the same protocol that is a standard for everyone else in the industry is a potential “edge” for MS, the situation is unstable. One solution would be to dampen MS’s general will-to-win, but that would be terrible. A better idea would be to separate control of the industry standard from some of the industry’s most effective competitors.

Peter Huber 10:54 a.m. Saturday 6/22/96

It’s hard to get a real economic grip on how much competition Microsoft really faces. Software is plastic. Microsoft can lose half its revenues without losing a single customer, or a single computer. Or it can double its revenues without gaining one.

Microsoft can fold more features into Windows, and increase price commensurately. Or it can try to capture a larger share of the downstream market for applications software. To take revenues away from Microsoft, other companies execute the same strategy in reverse.

Consumers can favor either side. Yes, they can buy richer, more bundled operating systems from Microsoft. But they can also stop (or delay) upgrading their MS operating system–they can instead buy things like Corel SCSI controllers, or Norton utilities, or device drivers from hardware manufacturers, and load these above or below Microsoft’s OS software. These same consumers can–in principle, at least–simply favor applications sold by Intuit or Lotus.

From a software engineering perspective, competitors can build underneath, alongside, or above Microsoft’s structures.

Neither DOS nor Windows 3.x, for example, supported SCSI devices properly. Corel built the drivers that would load underneath Windows, and packaged them with well-designed software that would run above. Corel SCSI made a lot of money. Windows 95 finally provided proper SCSI support, and that wiped out Corel–in that particular niche. Except for all the users who remained content with Windows 3.x–content because Corel had already sold them the one extra piece they really needed. Now Corel is selling the best software for writeable CD-ROM drives–which Windows 95 doesn’t yet support properly. Windows 97 undoubtedly will. By then, Corel will be on to something else. And once again, some buyers will postpone buying Win97, or perhaps skip it entirely, because Corel has already filled in the one missing tooth they wanted.

This endless process by which Microsoft plays catch-up to the likes of Corel isn’t anti-competitive. All software features start life as “fancy gizmos” and–if they’re really good–end up as “basic needs.” Initially they’re add-ons for the few. Then they become standard equipment for the many. Radios used to be optional in car dashboards. When car companies had the temerity to begin building radios into every dashboard, radio manufacturers sued–and for a while they won. But those cases seem silly today. Today’s consumer expects a car radio to be part of the standard “operating system,” not an applications-software “option” to be bought later, from a different vendor. We aren’t worse off because we’ve lost the chance to install a GM radio in a Ford car. The extra flexibility would be more trouble than it’s worth.

Microsoft’s competitors can displace MS features from above, too. Nobody who does serious work with files uses the capabilities built into Windows–Norton Navigator is much better. Once you’ve loaded Navigator, you never click “Find” in your Start list again. Few serious users of e-mail rely on Microsoft Exchange; I much prefer ConnectSoft’s e-mail connection. Windows 95 includes a fax-printer driver, but serious faxers favor Delrina’s WinFax Pro. Nobody who does serious tape back-up relies on Windows–Arcada is a much better program. And the thought that compression technology has been swallowed whole by Microsoft is laughable. Every manufacturer of tape drives, every manufacturer of video-conferencing software and hardware, most manufacturers of e-mail software (any program with zipping/unzipping capabilities), most modem manufacturers–in short, anyone at all who’s trying to squeeze more into less–is building compression hardware and software, and deploying it wherever needed.

In sum, Microsoft OS capabilities can be displaced, at the margin, from all sides. No single add-on from Norton or Corel eliminates the need for an MS operating system entirely. But together, add-ons like these do dictate how quickly you’ll update to the next release of an MS operating system, or indeed whether you’ll update at all.

This is why MS is so desperately concerned about the Internet, and Web browsers. These are “add ons” that threaten to grow and improve (with the help of the network beyond) much faster than the MS platform on which they stand. If that happens, the platform itself, and upgrades to it, will become less and less important. The platform itself probably won’t disappear for a long time. But its importance, and the incentive to upgrade it, can shrink almost indefinitely. Even if every last computer in the universe still contains a tiny kernel of MS value a century from now, it’s quite possible that no one will know or care. Market power over something as fluid and flexible as software requires much more than a “Windows Inside” logo.

James Gleick 1:08 p.m. Saturday 6/22/96

On Saturday, June 22, Peter Huber wrote, “Microsoft can fold more features into Windows, and increase price commensurately. Or it can try to capture a larger share of the downstream market for applications software.”

Exactly. It has done both of these, inexorably.

“To take revenues away from Microsoft, other companies execute the same strategy in reverse.”

They do? In what universe? If we’re trying to understand the Microsoft phenomenon realistically, one fact that’s important to note is: Microsoft has never lost a franchise. Sure, it doesn’t control the entire world yet. It has stumbled from time to time–sometimes embarrassingly. But there is NO piece of the software landscape that Microsoft has gained and then lost. None.

In any category you care to name, Microsoft’s market share goes up–never down.

That alone shows that, in ever greater segments on the industry, competition is a chimera.

“From a software engineering perspective, competitors can build underneath, alongside, or above Microsoft’s structures.”

Precisely. They can find tiny niches that Microsoft has not yet closed off. They do this knowing that the clock is ticking . . .

“Now Corel is selling the best software for writeable CD-ROM drives–which Windows 95 doesn’t yet support properly. Windows 97 undoubtedly will. By then, Corel will be on to something else.”

By necessity. Or it will be dying, like so many of its predecessors.

“This endless process by which Microsoft plays catch-up to the likes of Corel isn’t anti-competitive.”

Just a couple of years ago, Microsoft owned only the operating-system space. It had no leading application in any major category. Some party-pooping Cassandras warned that Microsoft was using its operating system dominance to take over the major applications-software domains, and indeed, today, things like word-processing, spreadsheet, presentation graphics, etc., etc.–formerly major competitive battlegrounds–are totally dominated by Microsoft. Again: 90 percent shares. No one entering these markets. The old leaders fading away to nothing.

Peter, if this is not “anti-competitive,” what would be?

And by the way, I see not even you are pretending that programs like WordPerfect and Harvard Graphics and Lotus 1-2-3 have been blown away because they were inferior products, less beloved by consumers. Of course, Ballmer says they were. The Microsoft syllogism is: our beloved customers are buying Word instead of WordPerfect; therefore Word must be a superior product. Everyone outside of Redmond has noticed that the dynamics of the industry make a Microsoft blitzkrieg possible with products that are no better, and sometimes significantly worse.

“All software features start life as “fancy gizmos” and–if they’re really good–end up as “basic needs.” Initially they’re add-ons for the few. Then they become standard equipment for the many.”

This shows that you are BEGINNING with the assumption that Microsoft owns the world. In the automobile industry, the fact that something is standard equipment doesn’t mean that it is manufactured by the One True Automobile Company. There is still such a thing as competition. Just not in the software business, where if something is standard equipment, that means Microsoft.

“Microsoft’s competitors can displace MS features from above, too. Nobody who does serious work with files uses the capabilities built into Windows–Norton Navigator is much better. Once you’ve loaded Navigator, you never click “Find” in your Start list again. Few serious users of e-mail rely on Microsoft Exchange; I much prefer ConnectSoft’s e-mail connection.”

Wow–you’re even more of a software junkie than me! But seriously, when “Find” is built into the operating system, and so is Exchange, how many users will buy, install, learn, and maintain these add-ins? For that matter, in the case of Exchange, where Microsoft is playing the game of manipulating the MAPI standards as we speak, third-party e-mail software companies are now fighting a rear-guard action. Go ahead and use ConnectSoft if you like it–but I wouldn’t buy their stock for your retirement account.

“Windows 95 includes a fax-printer driver, but serious faxers favor Delrina’s WinFax Pro.”

Another perfect example. A few fanatical faxers may want to pay for WinFax’s extra functionality. But what do you think is happening to Delrina as Win95 gains market share? Don’t buy their stock for your retirement account.

“Nobody who does serious tape back-up relies on Windows–Arcada is a much better program. And the thought that compression technology has been swallowed whole by Microsoft is laughable.”

No, sir. The disk-compression technology invented by Stac is now a purely Microsoft business. There’s no competition. It’s over.

“This is why MS is so desperately concerned about the Internet, and Web browsers.”

Yes. My comment on the Internet and Netscape is still to come….

Herb Stein 4:55 p.m. Saturday 6/22/96

We now have a number of balls in the air and I would like to try to focus on what seem the most pervasive issues in the discussion so far. I will start with the most elementary one.

Everyone reading this magazine has on his desk a more powerful and capable computer than he had or dreamt of, say, fifteen years ago. That is due to scientific and technological advances in which Microsoft did not have a major part. The question now is not whether there has been great progress in computers but whether the present organization of the industry, with Microsoft dominating operating systems, helps or hinders the translation of scientific and technological advance into the hardware and software we all work with. Would we now have better equipment at our finger tips if instead of Microsoft there had been several more nearly equal competitors? Is the very fact of dominance by one company evidence that it is the most efficient way to organize the industry from the standpoint of computer users? Was it ever possible, and is it possible now, to have more competition in the operating system end of the business?

I would like to hear the panelists’ views on all that . Later I would like to turn to questions about the relations between Microsoft and producers of applications, to the anti-trust or other public policy implications of the way the industry is organized, and about the possible effects of the Internet on industry structure.

James Fallows 12:38 p.m. Sunday 6/23/96

1. “CHOICE” AND “QUALITY” IN MICROSOFT’S MARKET DOMINANCE.

Steve Ballmer says, laudably, that Msf has gotten as far as it has by meeting customer’s needs. Roger McNamee says “As far as I can tell, Microsoft got to these market share levels because customers made choices in the open market.”

None of these exactly rings true to the software world I’ve seen as a purchaser since buying my first PC, a handsome Processor Technology SOL-20, in 1979. (’Jeopardy’ Hint: the SOL-20 is the answer to the question, “What is the only computer ever sold with mahogany siding?” Its initial operating system was NorthStar DOS, and its first word-processing software was The Electric Pencil.)

No one would dispute the claim that Msf is a great company or that it has truly great leadership and talent. But I don’t think I have ever heard anyone outside Msf say that it has succeeded because it offers such unusually great software. What it offers is software that will INEVITABLY BECOME THE STANDARD, and that people therefore get accustomed to rather than considering other, theoretically more-promising alternatives.

To pre-empt angry outbursts: yes, Msf makes software that is a million times greater than anything I could produce. The question is what has accounted for its ratchet-like increase in market share. Is it the ratchet-like increase in the brilliance and convenience of Msf products? Or is it the company’s very shrewd use of “every edge they can get,” including control of the OS standard?

I offer the buildup (Huzzah!) to Win95 last year as a prime illustration. Can anyone seriously contend that success of this software is mainly due to its technical superiority over alternatives (Mac, Unix, OS2, etc.)? Anyone who thinks that has not tried these other systems. It’s better than Win31, sure – but the reason it has sold sixty zillion copies is that everyone assumes that for better or worse it IS the standard. Here is why one of its sales was to me: a week ago, the motherboard finally failed in the trusty old laptop I had used for years. I bought a new laptop, and the model I wanted offered no choice. The ONLY way it came was with Win95 installed. Yes, this is consumer choice in the open market–sort of. But it mainly shows the brilliance of Msf’s VHS-style, McDonald’s-style emphasis on *becoming the standard*, rather than producing what is in theory the most innovative, convenient software. Nothing unjust – or unwise! – about this approach, but it should keep us from falling into Milton Friedman-like panegryics about the eloquence of the consumer’s voice speaking through the market.

I should make clear to my friend Steve B, though, that if there is a company I’m really on the warpath against, it’s not Msf but the dreaded Lotus. Lotus took just the opposite strategy. During the Mitch Kapor era it produced what are, nearly a decade after their introduction, still two of the most brilliantly useful pieces of PC software ever designed: Magellan and Agenda. Then it let them die. Msf, like the great Japanese industrial empires it resembles in other ways as well, has never made that error–and this persistence with its products is part of their air of inevitability.

2) FATALISM FROM J. GLEICK

After Steve Ballmer wrote: “….Windows is an open platform, based on published specifications, and Microsoft provides a lot of support for developers writing Windows applications.”

Jim Gleick then replied:”I’m glad Steve Ballmer said this. There won’t be a more important sentence written during this discussion.”

Aw, shucks, Jim, who knows how important we’ll get during this next week? But you’re entirely right that if Steve B’s statement reflects thought-out Microsoft policy – and how can in not, since there is only one person who could conceivably countermand him – it is very significant. This leads to….

3) ASSIGNED ESSAY FROM OUR MODERATOR

A Mr. Herbert Stein of Washington DC writes:

“Would we now have better equipment at our finger tips if instead of Microsoft there had been several more nearly equal competitors? Is the very fact of dominance by one company evidence that it is the most efficient way to organize the industry from the standpoint of computer users? Was it ever possible, and is it possible now, to have more competition in the operating system end of the business?”

This very nicely clarifies a main issue. It is good for the industry to have a standard. It may be inevitable that one company would end up in control of that standard. It is not necessarily healthy, natural, inevitable, or efficient for that same company to retain proprietary control of that standard while also competing against other companies that rely on the standard. This is all the more so if the company is famous for “competing to win” and “taking any edge they can get.”

Has Steve Ballmer promised that Microsoft will henceforth treat its OSes as a kind of public trust, whose full, published, transparent nature is necessary for the health of the whole industry? If so, then not only will the industry prosper and our citizens cheer, but also Slate may have done the near-impossible: producing news that exceeds the advance-hype billing!

Cordially, Jim Fallows

James Gleick 12:51 p.m. Monday 6/24/96

To tackle Herb Stein’s questions: yes, this monopoly may well have been inevitable–not because it is good for consumers, but because that’s the way the dynamics of this market work. The computer software industry is clearly governed by the law of increasing returns. The more people use an operating system, the more hardware manufacturers will ensure compatibility with their products and the more software makers will support it … and vice versa. It’s a powerfully self-reinforcing process.

Real competition in operating systems might have been better for consumers in the long run. Unquestionably the operating system losers–Macintosh, OS/2, Next–had various real advantages over Windows, some of which Windows has now borrowed. What other operating systems might have been, we’ll never know now. Microsoft’s increasing dominance in all the major applications categories also bodes ill for innovation and technological progress. Have an idea for a new kind of word processor? (I do.) Forget about it (as we say in Brooklyn).

However, an admission: There really are some benefits for consumers in having all this software provided by a single company. Software is complex enough. Microsoft forestalls disorganization among hardware companies who cannot agree on standards; Redmond just tells them what to do. And Microsoft is correct to say that consumers get value for their dollar in a rich, feature-laden product like Windows 95. These benefits are genuine, if slightly distasteful, like the advantages that come from totalitarian governments. (In Singapore, the trains do run on time.)

It is not possible now to envision real competition in the operating-system market. Some people speak metaphorically of Internet applications as a new sort of “operating system,” but that stretches the term beyond its breaking point. Internet applications still need a real operating system underneath. No one could start now and create a complex of software standards on the scale of Windows, supporting the broad range of hardware devices necessary as well as the full panoply of applications software. That moment in history is past.

Instead, as a matter of public policy, it would be productive to recognize that the PC operating system is an essential facility, like the dial tone and the 60-hertz alternating current. Its accessibility to all software designers–the openness that Steve Ballmer referred to–must be made real and then protected.

All this would be much easier to understand, by the way, if the monopoly were in hardware rather than in the esoteric collection of standards that makes up an operating system. Suppose one manufacturer–call it “IBM”–produced 90 percent of the world’s PC’s and maintained ownership of this product through its patent rights. Suppose this monopolist had used its advantage to gain dominance, successively, in one major category of software after another.

Well, we wouldn’t be having this debate today. We would be scared and saddened, no matter how well our PC’s worked.

Steve Ballmer 4:00 p.m. Monday 6/24/96

I must take issue with Herb’s comment that “Microsoft did not have a major part” in the advances that have fueled the PC revolution. I don’t think even our harshest critics would discount the technological efforts and achievements of Microsoft software developers during the last 15 years. We’re working on powerful, capable PCs today precisely because of the synergy of innovations from chip vendors like Intel, hardware vendors like HP and Compaq and software companies like Microsoft, who together have fueled massive penetration of PCs into corporate settings and the home. For the past 20 years, our mission has been to build software that enables consumers to take advantage of the myriad of innovations in hardware, peripherals, etc.

We’re spending more than a billion dollars each year in R&D working to continue this trend of faster, more capable, more affordable personal computing. We formed Microsoft Research five years ago out of a commitment to, in Herb’s words, help “the translation of scientific and technological advance into the hardware and software we all work with.” We have over 100 researchers conducting cutting-edge research in a variety of areas ranging from speech technology to data mining to speech recognition to 3D graphics. The goal? Take unsolved problems and convert them into tangible form so that software developers, within and outside Microsoft, can take the next step and transform them into valuable products that meet customers’ needs and enhance their lives.

A few professional Microsoft critics like to say other products are consistently superior to Microsoft’s. Well, millions of consumers around the world don’t agree. Neither do a host of independent reviewers. The fact is, Microsoft products are among the most honored in the industry, winning dozens and dozens of awards from independent software reviewers who pride themselves on their independence. I asked our product groups to send me a few examples of these awards and reviews. The bottom line is, you either constantly improve your products and technology in this industry or the consumer will choose something else. …

PC Computing MVP awards 1995.

PC Computing

PC Computing

PC Computing Computer Life Best of Everything Award: Encarta 95. (Named best of best of everything awards.)

C/Net Microsoft Delivers Stunning Blow to Netscape With IE 3.0 (5/30/96), by Daniel Will-Harris.

Steve Ballmer 6:42 p.m. Monday 6/24/96

Jim G., you need to go back and check your facts on the outcome of the DOJ investigation of Microsoft, because every time you refer to it, you get it wrong.

Between 1990 and 1994, not one but two federal agencies exhaustively investigated the raft of competitor complaints that you and Jim F. are dredging up once again. In the end, the FTC made no claim against Microsoft. The DOJ determined to challenge only certain narrow aspects of the way we licensed operating systems software to computer manufacturers. In fact, the DOJ specifically told the courts that there were no grounds to file suit against Microsoft other than those contained in the resulting settlement. You can argue for a change in the rules of competition, but under the existing rules, that was the outcome.

Your claim that the DOJ “proved the illegal tactics by which Microsoft killed the last remaining operating-system competition” is just flat out false.

First off, the DOJ didn’t prove a thing. All it did was allege. Microsoft denied that it had committed any wrongdoing, and the case was settled, as civil disputes so often are.

Second, the DOJ didn’t even allege that the initial success of MS-DOS and Windows was due to any unlawful act (much less that Microsoft killed any operating system competitor). The DOJ said in court that there was no basis for an antitrust challenge to the way that Microsoft made MS-DOS and Windows a great success.

Go back and read the DOJ’s briefs. The DOJ’s argument was this–the challenged licensing practices “may have contributed” “to some degree” to the continuing success of these products. How much did they contribute? The DOJ argued that the challenged practices “had only a minor and nonmaterial effect on Microsoft’s current position.”

So what was all the fuss about? It’s hard to say. We always said that the reason computer manufacturers pre-installed Windows on most of their machines was because most of their customers wanted to buy Windows machines (not because of particular contract language). It isn’t that complicated. So we made a business decision to settle with the DOJ, and modify our licensing practices in certain ways. We’ve been operating under that settlement for two years now. And guess what–consumers still want Windows machines, and computer manufacturers are still making them.

James Fallows 7:32 p.m. Monday 6/24/96

To butter up Steve B for a moment, he is a good sport to participate in this kind of debate–you don’t see comparable big shots from, say, RJ Reynolds or AT&T going online to answer questions about their company. Also, he has a point when saying: “I must take issue with Herb’s comment that ‘Microsoft did not have a major part’ in the advances that have fueled the PC revolution.”

My understanding is that Microsoft did indeed play a major role in making the PC business what it is today, but less for technical reasons than business-strategy ones. I don’t think that Steveb@microsoft.com or even Billg@microsoft.com would contend that the technical brilliance of Microsoft programmers was the *crucial* factor in making first DOS and then Windows the standards of the industry. (It would be a little hard to contend that, considering that the original DOS code was bought from outside suppliers, as is well known.) Instead, what made the company powerful, and what should impress observers, is the *business* vision of its leaders–in particular, the idea that becoming the standard mattered more than anything else. Even before there was a VHS/Beta drama, Gates, Ballmer, et al understood the importance of being VHS. They clearly did this for their company’s benefit, but it had a historic effect on the industry by creating a standard for apps-makers to use when programming.

Weirdly, the public still seems not to get this. When they see Bill Gates, they think of some Albert Einstein/ Gary Kasparov counterpart. No doubt the guy is smart, but I contend that there are more people as smart as he is (in the IQ sense) than there are business strategists as capable and relentless.

But back to the theme: Steve, it is good to hear that: “We’re spending more than a billion dollars each year in R&D working to continue this trend of faster, more capable, more affordable personal computing.”

My question to you would be: APART FROM ESTABLISHING A STANDARD, in what areas would you say that Microsoft products have taken the industry significant steps forward? I mentioned earlier that the doomed Lotus products Agenda and Magellan offered capabilities no other products had. Everyone knows about the innovations Apple produced with its Mac OS. NeXT has had significant cutesying-up steps forward. OS2 has had robust multitasking for several years. ALL of these products have failed or floundered in the market, compared to what Msoft has produced. Very many Microsoft products have succeeded. But again, my question: in which products, exactly, has Microsoft pushed the technical or convenience envelope beyond what smaller/weaker vendors had already provided. (I am not implying that there is no answer to this question. I just am curious which answers you would give.)

Now we come to the place where we just see the world differently. You say: “A few professional Microsoft critics like to say other products are consistently superior to Microsoft’s. Well, millions of consumers around the world don’t agree.”

I am not claiming that somebody else’s products are “consistently” superior to yours. But I draw different conclusions from the “millions of consumers” than you do.

For example, at this moment I am typing these words on a laptop using Win95. Did I choose this interface because I was convinced of its superiority over all alternatives? No. I “chose” it because I HAD NO ALTERNATIVE. My old laptop blew up; I needed a new one; the best available hardware at the moment–the new Thinkpad 560–came with pre-installed Win95 as the ONLY available configuration.

That IBM, one of whose divisions is pushing OS2, should offer ONLY Win95 on its hottest new machine says something about the flatliners running that company. But it also shows that simply becoming the standard makes a lot of other “quality” comparisons meaningless. Do millions of customers “choose” the QWERTY keyboard above all alternatives? No–it’s just there. Do I “choose” to fly USAir when I go from DC to Boston? No. There is literally no alternative. And so win DOS/Windows. You can rightly use the “millions of customers” as evidence of the brilliance of your business strategy. As evidence of the brilliance of the products themselves….

This brings us back to Jim Gleick’s crucial point. He says: “Instead, as a matter of public policy, it would be productive to recognize that the PC operating system is an essential facility, like the dial tone and

the 60-hertz alternating current. Its accessibility to all software designers–the openness that Steve Ballmer referred to–must be made real and then protected.”

Yes. Whatever the historical reasons for the Dos/Win ascendancy, whatever the pluses and minuses might have been of a multi-OS world, our real world now is the world of Windows-as-dial-tone. And the question is: how can customers and competitors be sure that Microsoft now sees the OS as a dial tone, and not as part of its “edge”?

Signing off from our nation’s capital, Jim Fallows

James Gleick 10:25 p.m. Monday 6/24/96

Steve Ballmer wrote: “Jim G., you need to go back and check your facts on the outcome of the DOJ investigation of Microsoft, because every time you refer to it, you get it wrong.”

Oh, no. You’re not really going to dispute these facts, are you, Steve? OK, let’s see: “Between 1990 and 1994, not one but two federal agencies exhaustively investigated the raft of competitor complaints that you and Jim F. are dredging up once again. In the end, the FTC made no claim against Microsoft.”

To be precise, the commissioners deadlocked on pursuing the FTC staff’s recommendations, and the Justice Department took over the investigation.

“The DOJ determined to challenge only certain narrow aspects of the way we licensed operating systems software to computer manufacturers.”

Your word is “narrow”? Let’s see. … The DOJ determined, among other things, that Microsoft over a period of years had violated Sections 1 and 2 of the Sherman Antitrust Act, using (to quote from court papers) “exclusionary and anti-competitive contracts to market its personal computer operating system software. By these contracts, Microsoft has unlawfully maintained its monopoly of personal computer operating systems and has unreasonable restrained trade.” In summary: “illegal monopolization and unlawful restraint of trade.”

“First off, the DOJ didn’t prove a thing. All it did was allege. Microsoft denied that it had committed any wrongdoing, and the case was settled, as civil disputes so often are.”

“Prove” is both a plain English word and a legal term. Of course there was no “proof” in the sense of a judicial verdict: Microsoft was permitted to settle in a polite pas de deux, agreeing to end a set of illegal practices without making any admissions of wrongdoing. As I’m sure you recall, the U. S. District Court that should have approved this settlement found it distasteful and refused, stating: “The picture that emerges from these proceedings is that the U.S. Government is either incapable or unwilling to deal effectively with a potential threat to this nation’s economic well being. It is clear to this Court that if it signs the decree presented to it, the message will be that Microsoft is so powerful that neither the market nor the Government is capable of dealing with all of its monopolistic practices.” An appellate court then approved the settlement.

I won’t waste more space on this. It’s history, and it can’t be undone. I hope you aren’t giving us a preview of the Encarta version.

Peter Huber 10:18 a.m. Tuesday 6/25/96

Herb Stein asked for a 200-word response to the following:

“Would we now have better equipment at our finger tips if instead of Microsoft there had been several more nearly equal competitors? Is the very fact of dominance by one company evidence that it is the most efficient way to organize the industry from the standpoint of computer users? Was it ever possible, and is it possible now, to have more competition in the operating system end of the business?”

Languages are built in layers: WordPerfect on top of Windows on top of an Intel processor. Demand for the whole is determined by what runs on top. Supply at the top depends not only on the quality but also on the ubiquity of the linguistic layers beneath. That’s why a community always converges on a common language sooner or later.

We might well have a better foundation today if writers and users of applications software had not committed themselves so early, or in such numbers, to one supplier of processors (Intel) and operating systems (Microsoft).

But delay in converging on a common foundation would have postponed development of the applications above. Word processors, Web browsers and microprocessors would be weaker than they are if Intel and Motorola, Apple, Windows, and OS/2 were all still contending to build the infrastructure.

How long to keep open a contest among competing builders of foundation is a choice best left to the market. To this day, PC users can still switch to another, perhaps more robust foundation–Apple or an Intel-OS/2 system. Few do. The shortcomings in the Intel-Windows foundation just aren’t serious enough to outweigh the advantages of a superstructure already in place above it.

Herb Stein 10:33 a.m. Tuesday 6/25/96

I am greatly impressed by the expert knowledge shown by all our panelists. In the time we have left I would hope to get further illumination on three subjects:

1. Software There seems to be a big difference among the panelists about whether the Microsoft operating system standards are freely and equally open to all manufacturers who want to make applications that would run on the MS operating system. I would have thought that Microsoft would want to have as many good applications as possible, no matter where they came from, available to run on its operating system because that would enhance the value of the operating system. I suppose there is something I don’t understand about that.

2. Antitrust and other policy implications: Economists love the story of the two economists who are walking down the street together. One says, “Look! There’s a twenty dollar bill on the sidewalk.” The other says, “No, there can’t be, because if there was someone would have picked it up.” That is a joke because it makes fun of the economists’ usual assumption that everyone has perfect knowledge and acts instantly and rationally on the basis of it, an assumption that is known to be unrealistic. Huber’s argument that Microsoft cannot be indulging in monopolistic practices because if they had been they would have been convicted reminds me of that story. But suppose that he is correct. As McNamee said, the antitrust laws were not enacted with Microsoft in mind. We may still be left with a situation in which for perfectly natural and innocent reasons one company comes into control of an essential facility of modern life. Is there anything to be done about that? In there any other non-regulated private company that controls so important a product? As I understand him, Gleick wants to declare the operating system a free, common standard, but there seem to many difficulties with that, such as how to effect the transition and what that would do to incentives for further development. Any suggestions?

3. The Internet: The Internet seems to be both a potential solution and a potential problem. It has the capacity to weaken Microsoft’s position as a supplier of applications and, perhaps at some time, as a supplier of the operating system. On the other hand, monopolization of access to the Internet could be much more worrisome than monopolization of operating systems. We see intense competition for position in the Internet market today. Is there a possibility that once one supplier, whether Microsoft, Netscape or some other, gets a lead there will be an irresistible tendency for all users to sign on to it, creating the same kind of situation that exists with operating systems today? Would it matter whether that dominant supplier of Internet access was Microsoft?

Peter Huber 8:41 a.m. Tuesday 6/25/96

Herb Stein wrote: “Economists love the story of the two economists who are walking down the street together. One says, “Look! There’s a twenty dollar bill on the sidewalk.” The other says, “No, there can’t be, because if there was someone would have picked it up.” That is a joke because it makes fun of the economists’ usual assumption that everyone has perfect knowledge and acts instantly and rationally on the basis of it, an assumption that is known to be unrealistic. Huber’s argument that Microsoft cannot be indulging in monopolistic practices because if they had been they would have been convicted reminds me of that story.”

I would doubt the sanity of anyone who told me that $20 had been lying on the ground right outside Grand Central Station for the last five years. But that’s what Jim Gleick is saying. Microsoft’s anti-competitive conduct is obvious and beyond serious dispute. It has been for years.

The best sanity check we have for antitrust “facts” like these is the complete absence of any attempt to present these same facts to a neutral judge and jury. When facts are plain to everyone except those who could extract treble damages out of them in court, we have good reason to doubt their existence.