Why I hate Microsoft

"A personal, lengthy, but highly articulate outburst"

Table Of Contents | Previous chapter | Next chapter

5. Bad practice, foul play

"The greatest joy a man can know is to conquer his enemies and drive them before him. To ride their horses and take away their possessions. To see the faces of those who were dear to them bedewed with tears, and to clasp their wives and daughters to his arms."

-- Genghis Khan

Doing business with Microsoft has never been without risk. They have earned a reputation for dirty deals and backstabbing their business partners whenever that happens to suit them. Time and again small but innovative high-tech developers have entered into partnerships with Microsoft, only to find that Microsoft broke agreements, stole their technology, did not deliver, then dumped them on the edge of bankruptcy. It's also not uncommon for Microsoft to take over a small but innovative company and to put it in the trash can right away, just to keep new and competing bits of technology entirely off the market.

Of course there are plenty of other companies who play dirty tricks in order to get ahead in the market. Microsoft is merely one of the biggest companies who have routinely used foul tactics. In fact that's one of the normal risks of doing business. "If you can't stand the heat, stay out of the kitchen", as the saying goes. Still the risks of a partnership with Microsoft must be considered.

The customer on the other hand has not chosen to be part of this particular fight for commercial domination. Microsoft's clients, and the clients of their business partners, are being promised, pay for, and are thus entitled to expect, good products. Instead they get bad products, or none at all, and they end up as pawns in Microsoft's foul play to establish a complete monopoly. They start out with receiving sub-standard products, and eventually they find themselves tied into the deal indefinitely.

Bug fix New product

As has already been discussed, there are many things wrong with Microsoft products. Bugs and design flaws are common. Of course, all 'issues' with MS software will be dealt with in the next release... But not right now. That's the whole point: instead of fixing bugs, Microsoft actually uses these flaws as an excuse for an aggressive update policy. Microsoft doesn't fix bugs, but continuously releases new versions. They call this "innovation", but in truth the only purpose of this strategy is to inflate Microsoft's already obscene profits even further.

Picture this: you buy a newly-built house from a real estate company. As soon as it starts to rain, you discover that the roof leaks. When you complain about it, the real estate company either ignores you or they tell you that this kind of roof is a brand-new innovation; the sort of house they used to sell never had such a beautiful roof. Instead of fixing your roof they promise that the next house they'll build won't leak. Eventually they complete their next house, three years or so behind schedule, and you have to pay a hefty price for it... only to find that it comes with a patched roof, and now the water seeps through the walls instead. The new house has an extra wing added to it that you didn't ask for, but as soon as you enter it the floor collapses, and if you try to save yourself you find door jammed.

Would you accept such nonsense? Of course not! You'd file complaints, you'd sue! But this is what Microsoft has been doing with their software products, and the user community has been taught to accept this.

The "innovation" upgrade treadmill

Of course new releases are necessary if software is going to evolve at all. But are these new versions indeed as innovative as Microsoft would have us believe? Or is it merely a chance to integrate the separate Microsoft products more tightly and to increase the users' vendor dependence?

Let's say that a new version of a Microsoft application comes with documents in a Microsoft proprietary format (e.g. the .CHM format, a Microsoft-proprietary version of HTML, used for help files and such). This is something that has happened many times in the past and will continue to happen. The use of this proprietary file format means that we're now suddenly forced to install (or update to) a recent version of Internet Explorer (which means installing or updating Outlook as well) because no other application will correctly support this file format. Since earlier version of IE under Windows were designed to conflict with other browsers (e.g. Netscape Navigator) something as simple as online documentation in a "product update" could mean to discontinue competing products in favor of a Microsoft-only environment. These days, in the post-IE5 era, such conflicts aren't the problem they used to be, but still IE and Outlook represent additional overhead, additional security vulnerabilities, and additional maintenance. And the user has no choice but to accept that.

And what good is such an update, really? What are the innovations in, say, Office XP over previous versions? The most significant 'improvement' is that new versions of Office produce documents that are incompatible with older versions of the same applications. If I create a document in a current version of MS-Word (e.g. Word 2000 or XP) and mail it to a friend who still uses an older version (say, Word '95), he cannot read it, view it, print it or anything else. He's forced to let Microsoft ram an unwanted and expensive upgrade down his throat before he can use my document.

And it doesn't stop there. Let's take a look at Office 2000. Quoting PC-World, June '99:

"An interesting issue is that Office2000's HTML format is incompatible to some extent with almost any other program, including MS's. So you can create all sorts of groovy Word, XL, or PP documents, and only people with Office2000 can read them. Even IE4 and 5, and Front Page 2000 can't read XL or PP files in HTML format without significant distortions in the display."

New releases of Microsoft products generally don't contain any significant innovations whatsoever. As far as improvements are concerned, a "new" release like Office 2003 has barely made more than a ripple: it doesn't do anything better than versions five years old do. It's the same with Office 2007: the only real difference with Office 2007 is the user interface, and most users don't consider that an improvement anyway. In fact you could take versions of Word and Excel from 10 years ago and they'd do the same job just as well. But still we have to keep buying new versions at steep prices in order to maintain document compatibility with our fellow users. Microsoft calls this "the freedom to innovate", and waxes poetic about "all this exciting new technology that has been invented by Microsoft".

As if the document's version-dependence wasn't enough, Microsoft also discontinues serious support for each product version soon after a new version is released. Although by the end of 2002 Microsoft announced prolonged support in the future for their major products, support for older versions remains limited. Several months after Windows Vista's release the new version proved less than popular, mainly due to its many driver issues, bizarre hardwere requirements and strangely expensive licenses. Microsoft's response to the continuing demand for Windows XP was to announce that by the end of 2007 XP licenses will no longer be available.

If you have problems with any Microsoft product, you are invariably encouraged to buy a new version, usually at a rather steep price. But when you do, you'll find that the bugs haven't been fixed, that new bugs have been introduced, and that all design flaws have been perpetuated.

Proprietary lock-in pushing bad products

Microsoft forces us to buy substandard, proprietary technology along with Windows. Take ADSI for example. The Active Directory capabilities in Windows 2000/XP are much harder to integrate into a multi-platform environment (e.g. in combination with Novell Netware or Unix systems) than the more primitive domain services (which could be taken care of by means of a simple redirection mechanism). Of course Microsoft has done little to facilitate the integration of ADSI with other products; ADSI is engineered to promote Microsoft-only environments; it's an immature product that can't handle multi-vendor or multi-platform environments and scales poorly.

After only a few months of use in the larger corporate environments, ADSI's limitations were already painfully obvious: only 5000 users per group, single points of authentication (which means that remote offices are dependent on the availability of WAN links for local log-on) and most painful of all: the lack of adequate record locking, which means that two simultaneous updates of one record will result in serious loss of data.

ADSI's limited scalability was again confirmed by the Gartner Group in August 2000: large corporations will suffer from excessive overhead and network load when implementing ADSI over, say, 300 offices or more (something that can be, and has been, successfully done with NDS). Microsoft's counterargument was that their own ADSI-based network contains some 39,000 PCs, but they neglected to mention that those PCs are scattered across multiple non-integrated domains. And of course this "metadirectory solution" touts being LDAP and ODBC compliant, but fails to mention that it requires custom meta agents to talk to an X.500 directory or sync engine (other than MS DirSync).

Draw your own conclusion; mine is that Microsoft has again released a half-baked product that hasn't been seriously thought through by qualified networking software architects, and no amount of patching or service packs will be able to remedy all the basic design flaws.

And of course real Active Directory support is available only on the Windows 2000 platform which doesn't have native NDS support, but still users will have to adopt it eventually. (Soon most applications and drivers will demand it, and of course Microsoft has already discontinued serious support and code maintenance for NT versions previous to XP.) Also, Microsoft has shamelessly admitted that the flaws in ADSI mentioned above would not be fixed before Windows XP, thereby effectively forcing large customers to accept another mandatory "upgrade" and of course even more vendor-dependent features.

So much for freedom of choice.


Microsoft has always claimed that the bundling of application software with Windows was only intended to improve quality, and that consumers are better served by the fact that both operating system and applications are produced by the same company. Well, we've seen how that goes. Word Perfect was a better word processor than MS-Word ever was (read: it delivered a better quality of word processing, whereas Word only contains more gadgetry). But when Windows 3 was released, few application developers had caught up with the need to entirely rewrite their application code. Only the Microsoft Applications Group was ready.

As it turned out, the latest release of Word at the time just happened to be fully compatible with Windows as soon as it hit the market, while WP Corp. struggled to get their DOS-version ported to Windows - with an unsurprising lack of success, as they had previously been forced to write DOS-dependency into their program code, due to DOS's lack of decent device support. And WP Corp. wasn't the only one: when Windows was first released most competing software vendors soon discovered that porting their existing DOS applications to Windows looked easier than it was, and that it took a complete re-write to produce efficient and stable code.

It's also common knowledge that MS applications perform much better under Windows than competing products ever can, since MS controls the API (Application Program Interface) and uses undocumented features to enhance their own products. Compare Internet Explorer, for example, with other browsers: IE hooks directly into Windows' internals while others are limited to documented API calls. And since IE and Windows share major chunks of code, firing up IE is much faster since when you start Windows you already preload most of IE. But Microsoft still denies having an unfair advantage over competing developers of application software. Instead they call this bundling "the freedom to innovate".

Can a monopoly innovate?

Let's take a good, hard look at this idea. Are a monopoly and the bundling of products really conducive to innovation?

Imagine for a moment that Standard Oil hadn't been stopped by the Sherman Anti-trust Act at the beginning of the 20th century, and had gone on to seize complete control of the fuel market. Let me stretch your imagination even further: suppose Standard Oil had also bought the Ford Motor Company. Owning virtually every gas station around, SO could have switched to a type of fuel uniquely suited for their own automobile products, and less well suited (and eventually unsuited) for competing cars. Consumers would have had no choice but to switch to SO-powered Ford automobiles. Both competing fuel vendors and automobile manufacturers would have been history.

If this had actually happened, what would car traffic look like these days? Think about it: would we have had a wide choice from affordable, safe and dependable mass-produced cars running 50 miles to a gallon? Or would we be driving a glorified Model T instead, in any color as long as it's black, at the original price or more, corrected for a century of inflation? (Today most people can afford a car. A century ago automobiles were far beyond most peoples' budgets.) And what would we pay for a full tank of gas, with oil prices being whatever the sole supplier says they are?

That is why monopolies and the wholesale bundling of products are bad. Bundling does not lead to innovation. Instead it merely lends power and control over the masses to those who practice it. This is interestingly, and maybe more than incidentally, reflected in the root of the word 'fascism', which is derived from the Latin 'fascio', or 'bundle'.

To illustrate: In November 2003 (during a major slump of the IT market) Microsoft declared a quarterly turnover of $2.81 billion for their Windows division, with a $2.26 billion net profit. Read: an 80.5% net profit margin on a multi-billion dollar turnover. MS-Office did slightly less well, with a mere $1.63 billion net profit, on revenues of $2.29 billion. On 30 September 2003 Microsoft had $51.62 billion on the books. And this at a time when the entire ICT industry couldn't afford to spend any money that can possibly be saved! And as if that wasn't enough, subsequently Microsoft's fourth fiscal quarter showed not only a 15% increase in turnover, but over 80% increase in net profit. In 2005 Microsoft's annual turnover had reached $40 billion with an annual net profit of $12 billion. The first quarter of 2006 showed $2.98 billion on a turnover of $10.9 billion, and the first quarter of 2007 netted a stunning $4.93 billion on a $14.4 billion turnover. That's almost five billion US dollars net profit in three months! The second quarter of 2007 continued the trend with another 13% increase in turnover, and subsequent financial figures proved this growth to be persistent.

This is what happens if a monopolist has been allowed to eliminate all serious competitors. By contrast, other Microsoft divisions such as Home Entertainment still have to compete with other players in the market, and these divisions declared losses up to a few hundred million dollars.

Perception is reality

A central principle in Microsoft's marketing is that it's far more important what the customer thinks he's getting rather than what's actually being delivered. Another major strategy is to hide the fact that Microsoft can never deliver what they promise, so that the customer will keep purchasing new versions over and over again.

Of course hardball sales tactics have never been the exclusive domain of any one corporate software company. And all is fair in love, war and marketing... Or is it? What about foul play? What about the distinction between competition, the prevalence of sales targets over ethics, and illegal practices?

For most of its history Microsoft has been involved in legal actions, the most important of which has been the investigation by the US Department of Justice (DoJ). This has culminated in the late nineteen nineties with the so-called anti-trust trials. The testimony from Microsoft's competitors was especially interesting to hear. Intel Vice-President Steven McGeady, called as a witness, quoted Paul Maritz, a senior Microsoft vice president as having stated an intention to "extinguish" and "smother" rival Netscape Communications Corporation and to "cut off Netscape's air supply" by giving away a clone of Netscape's flagship product for free. IBM representatives testified that IBM had been forced to drop OS/2 when Microsoft threatened to raise their prices for Windows OEM licenses. Digital Research demonstrated how Windows 3.11 was tweaked to crash when running on top of DR-DOS instead of MS-DOS. These are only a few examples; the list goes on and on.

Microsoft claimed in defense that this was all "innovation" and that the integration of Internet Explorer was a technical necessity. The Department of Justice then went on to demonstrate that this was a blatant lie, and made short work of Microsoft's entire defense plea. Microsoft in turn didn't even manage to present credible witnesses; all those who testified either had a significant interest in Microsoft or could be put at a significant disadvantage by Microsoft. Ironically, Microsoft's own witnesses, and even Gates' own testimonies, did their own case more harm than good. When Gates was summoned to testify in the case as the chairman of Microsoft, he was called "evasive and nonresponsive". He argued over the definitions of words such as "compete", "jihad", "concerned", "ask", and "we". BusinessWeek reported, "Early rounds of his deposition show him offering obfuscatory answers and saying "I don't recall" so many times that even the presiding judge had to chuckle. Worse, many of the technology chief's denials and pleas of ignorance have been directly refuted by prosecutors with snippets of E-mail Gates both sent and received."

US DoJ: Findings of Fact

On 5 November 1999 the DoJ published their Findings of Fact, and concludes, to condense the original document into a nutshell, that Microsoft has used foul play, has manipulated the market, has impeded progress, has harmed the IT market, the user community and consumers, and has violated anti-trust regulations:

"The ultimate result is that some innovations that would truly benefit consumers never occur, for the sole reason that they do not coincide with Microsoft's self-interest."

On 3 April 2000, the DoJ went on to state their Conclusions of Law and Final Order, leaving even less to the imagination:

"[Software bundling] cannot truly be explained as an attempt to benefit consumers and improve the efficiency of the software market generally, but rather as part of a larger campaign to quash innovation that threatened [Microsoft's] monopoly position. [...]
In essence, Microsoft mounted a deliberate assault upon entrepreneurial efforts that, left to rise or fall on their own merits, could well have enabled the introduction of competition into the market for Intel-compatible PC operating systems [...] thereby effectively guaranteeing its continued dominance in the relevant market. More broadly, Microsoft's anticompetitive actions trammeled the competitive process through which the computer software industry generally stimulates innovation and conduces to the optimum benefit of consumers."

I won't bore you with the rest of the legalese (which you can read for yourself, if you're so inclined, in the original document) but the bottom line is that Microsoft was found guilty as charged.

This ruling was in part (not in whole) reversed by the U.S. Court of Appeals and sent back to a lower court for reevaluation, not because the facts that led to the initial ruling were invalid (the court unanimously found that Microsoft engaged in unlawful conduct to maintain its dominant position in the operating systems market) but mainly on the grounds of unprofessional conduct by judge Thomas Penfield Jackson, who discussed his personal feelings about the case in the press immediately after the ruling. Although Microsoft claimed victory after this partial reversal of the original ruling, the essence of that ruling still stands, and Microsoft has still been found guilty of illegal monopolist practices and other unlawful conduct, such as:

Microsoft's response to anti-trust: business as usual

Shortly after this ruling Microsoft suddenly lifted the ban on the shipping of competing application software with Windows by PC manufacturers. This gesture was part of their attempts to mend their fences with the DoJ, but it's far too little and much too late to make any real difference as far as the IT market is concerned.

The release version of Windows XP proved to be nothing more than a continuation of Microsoft's monopolistic and anti-competitive practices. After being found guilty of forcing Netscape (later AOL Time Warner) out of the market with Internet Explorer and HTML, they continued the practice against Real Networks, and they are thumbing their nose at the DoJ and at AOL Time Warner by bundling MSN Messenger and MediaPlayer with the OS. You cannot have Windows XP without MSN Messenger, and it is cumbersome to install either of the other Instant Messaging Services into Windows XP. RealPlayer has already lost much of its market share to the bundled MediaPlayer, and is likely to be the next victim of Microsoft's product bundling strategies and suffer the same fate as Netscape did. How long do we have before people totally give up on AIM or ICQ?

Even consumer organizations have become worried now, as shown in a study by four major consumer organizations in the US in September 2001.

In the aftermath of the legal wrangling that followed the antitrust trials in the US, both Gates and Ballmer claimed repeatedly that it is technically impossible to remove application software like Internet Explorer and Media Player from the Windows distribution, and that Microsoft would have to take Windows off the market if a court order forced them to remove it anyway. This is of course nonsense. Microsoft has always claimed to be able to make anything that can be called software, and they have never refrained from doing so. They can put anything into Windows they want, and now they're unable to remove something from it and to implement a workaround to deal with any side effects? That's hard to believe.

The European Commission didn't believe it either, and the European Union's court ruled against Microsoft in 2004. Microsoft complied with the EU ruling and produced a Windows version without Media Player, thereby proving that Gates and Ballmer had been willingly and knowingly lying in their teeth when they said it couldn't be done. Of course by that time such proof was hardly necessary anymore. By the end of May 2002 Microsoft had already announced, in response to earlier legal settlements, that Windows XP Service Pack 1 would incorporate changes to allow consumers and PC makers to override Microsoft's default media products, and replace them with competing products. In other words, after Steve Ballmer's earlier statements of having to take Windows off the market if this ever came to pass, Microsoft fixed it with no more than a service pack. This would make Ballmer's hyperbole rather laughable... if the matter weren't so serious. In fact, it should tell us two things.

First, Microsoft admits having lied about this for years. So what else have they lied about? They said Media Player could not be overridden. They also said, in court, that Internet Explorer couldn't be overridden. Perjury is such a nasty word.

Second, Microsoft's claims about how this puts an end to unfair conduct and anti-competitive monopolist practices should be seen against a long history of consistent lying. We should not be surprised to discover other fraudulous practices. For example, much of Microsoft's removal of offending components could be limited to hiding the associated icons. Or code could be moved from application executables and hidden in DLLs or other obscure modules. After proof of the blatant lies we've seen recently, anything is possible. Whatever the case, we'd better not expect miracles.

At least not if Windows Media Player is any indication: after installing an update of WMP in Windows ME or XP, the application cannot be removed since it replaces parts of the OS. Even a service pack can be uninstalled, but WMP can't. Uh-huh. Furthermore, Windows XP Service Pack 1 made other changes to the system as well, to achieve 'further compliance' with several court rulings. This apparently included changes made to Outlook Express, that caused Outlook to label Microsoft's competitors' documents as dangerous, in particular Adobe Acrobat documents. However, the number one virus carriers in the world --Microsoft Office Documents with macros-- were not blocked.

Meanwhile the EU continues to take exception to Windows Media Player, which Microsoft still bundles with all major versions of Windows. The EU holds that this gives Microsoft an unfair advantage over competing media formats and is essentially a continuation of the same anti-competitive strategies that the DoJ objected to. The EU demands that Microsoft allow fair competition from other content providers and software manufacturers, but at this time of writing Microsoft's usual obstructive and delaying tactics have continued to prevent the EU from enforcing any legislation upon the company.

In short, it's still business as usual. The lies go on and on and on. After having testified in April 2002 that too many versions of Windows would be bad for consumers and for competition, Microsoft essentially doubled --to about two dozen-- the number of "current" versions of the operating system software. Between November 2002 and April 2003, for example, Microsoft released three new versions of Windows XP alone. This, and everything else, should should tell us a lot about the trustworthiness of Microsoft's testimonies.

Little to fear

Let's face it: even after officially being found guilty, Microsoft has little to fear. The sad truth is that the company has grown too big to be seriously affected by something as trivial as law and order. Microsoft could buy large portions of the United States if they wished to. Microsoft could buy several small countries. Microsoft knows the inside of the software that powers much of the world's economy; software that runs on government computers, including those used by the DoJ, the CIA, FBI and KGB...

But I digress. Suffice it to note that, according to a study by Common Cause, Microsoft has spent millions on political lobbying, doling out large sums across a wide spectrum of political activities since 1997. These millions are of course a mere pittance for Microsoft. Should the political climate turn unfavorable, I'm sure a few hundred million (which is only a small percentage of Microsoft's annual net profit) can easily be reallocated from the marketing budget to politics.

As much as I applaud the efforts of the DoJ to expose the practices of Microsoft for what they really are, I have to admit it's always been unlikely that effective action against Microsoft would ever be taken. The DoJ's ideas on how to make Microsoft cease and desist (e.g. the proposal that the company be broken up into different business units) could not have had the desired result, even if such measures could actually be enforced. So when the dust settled, Microsoft was still standing, grinning and raising a finger at the world, the user community and a powerless DoJ.

In fact, Microsoft has managed to appropriate some of the legal aftermath and turn it to their own advantage. For example, in January 2003 Microsoft settled one of their big anti-trust cases in California. Under the terms of the settlement, Microsoft agreed to pay back $1.1 billion to their customers. If part of that sum is not claimed, Microsoft will donate 2/3 of the remainder to schools, under the condition that at least half of the donation be spent on Microsoft products.

Hang on -- how's that again? Microsoft promises not to do it again, as they have done so many times before, never keeping their promise even once. Then they'll spend 1.1 billion dollars-- a sum that won't make a big dent in their annual revenues. But since usually less than 25% of the money is claimed in cases like this, they'll end up giving most of it to schools, half of it in the form of free Microsoft products, thereby eliminating the competition in just about the only market that Microsoft hasn't managed to monopolize yet. And they get away with it!

Something's very wrong here.

Expect no change

So far nothing has changed. And nothing will change. Things will just continue to get worse. Even after the DoJ's ruling, all available evidence suggests that Microsoft persists in practices that have been found unlawful. Next to R&D and marketing, Microsoft has now taken to budgetting major money (over $700 million in 2005 and more in subsequent years) for antitrust claims, rather than to clean up their act.

No matter how spectacular the innovations by competing vendors may be, the chances that they will be able to offer us these innovations so that we may benefit from them are practically zero. After all, Microsoft versions of similar products will already have been forced upon us with the installation of Windows, and Microsoft products will generally conflict with competing products. For example, when Windows XP was released, the media players from Apple (Quicktime) and RealNetworks (RealPlayer) wouldn't work any more, for no apparent reason. Users had to download patches or updates from Apple or RealNetworks in order to get these players to work again. And Windows XP came bundled with tons of multimedia applications to start with.

Even during many trials, Microsoft showed no signs of cleaning up their act. Arguing that conduct remedies were insufficient to stop Microsoft's anti-competitive and unlawful conduct, the DoJ reported that on July 11, 1999, "Bill Gates wrote an E-mail directing that Microsoft redesign its software to harm competitors" who make personal digital appliances. It indicated "a willingness to change the details of its Office applications to favor devices that run on Windows, even if doing so would disadvantage other customers who now rely on the Palm Pilot", officials said. The department noted that this was less than 30 days after the company's 78-day trial ended, in which it was accused of using similar tactics against Netscape and others. Microsoft went on to release PocketPC, the successor to the Windows-CE operating system for hand-held devices. Initially the above allegation of unlawful conduct was denied, but then Microsoft requested that Gates' E-mail be placed under court's seal.

Moving right along

The future doesn't look very promising either. Leaked-out beta versions of Internet Explorer contained hardcoded links to Microsoft websites, and have increasingly been designed as an integral part of Windows. And this will only become worse as more and more Microsoft products become Internet-based.

Microsoft's long-term strategy will target (read: attempt to appropriate) Internet Services, and promises to introduce even more proprietary standards than before. Microsoft still controls all major proprietary API's. The documentation they released as part of a legal settlement in August 2002 was incomplete and virtually worthless, not to mention full of errors that were obviously the result of sloppiness rather than of malicious intent. No adequate API documentation has been released since. That means that, with a shift to Internet-based computing, Microsoft essentially controls what will work or not for third-party software. During several early announcements of Microsoft's new Internet-based strategy, Bill Gates conceded that "while all .Net devices will have access to Microsoft's .Net infrastructure, those based on the Microsoft Windows platform will work better". Sounds familiar, doesn't it?

If you haven't got the full picture by now: Gates told the audience at the MS Developer Conference on 13 July 2000 that, quote, "the next two releases of Windows is where you'll see .Net built into the user interfaces.", unquote. He went on to outline the most profound changes in the User Interface that can be expected in the foreseeable future. One of Microsoft's software partners, who requested to remain anonymous, said:

Remember it's Microsoft we're talking about. Microsoft's number one priority in the post-Windows-2000 era is the same -- to make sure all devices are Microsoft-based.

As a taste of things to come, MSN users noted as early as October 2001 that MSN suddenly required the use of Internet Explorer. Users of other (competing) browsers were redirected to a webpage where they could download IE. After considerable public pressure Microsoft dropped this requirement, thereby proving that there was no technical necessity for such a browser-dependency, but that it was a commercial issue only. However, a Microsoft spokesman warned that users of competing browsers would have a "slightly diminished experience" because non-MS browsers "do not support MSN's HTML standard". Go figure.

Now, several years later, many of Microsoft's earlier attempts to bind the user community to proprietary Internet-based technology have not yet panned out. .Net was eventually released as a development framework for network applications. Their attempts to seize control over third party authentication services have crumbled under public pressure, and the announced changes in XP and Vista have failed to materialize. However the long-term strategy to shift from desktop-based to Internet-based computing is still in effect, and will prove to be yet another attempt at exerting control over the user community.

MS marketing: the anti-truth

Some examples of untruth in Microsoft marketing are almost funny. In the first months of 2002, Microsoft (along with partner Unisys) put up a website with the title "We have the way out". This website was part of a campaign that used slogans such as "Unix makes you feel boxed in. It ties you to an inflexible system." The ICT community was vastly amused: this website ran on Apache and Free BSD Unix. Then, in August 2003, Microsoft changed their DNS so that requests for www.microsoft.com no longer resolved to machines on Microsoft's own network, but instead were handled by Akamai's caching system... which ran Linux.

But most of all, Microsoft continues to, how shall I put it, adhere to rather peculiar ideas of what's true and what isn't. Spreading FUD (Fear, Uncertainty and Doubt) and other forms of misinformation has always been normal business routine for Microsoft. For example, on 22 December 1999 the Microsoft website blandly stated that:

"These are just a few of the features Windows 2000 Server offers that aren't found in [Novell] NetWare:
Integrated namespace support, file compression, configurable block size, mirroring, duplexing, striping with or without parity, removable device support, link tracking, integrated content indexing, user-definable file properties and a tracking log to audit storage services utilization."

They went on to state that NDS (Netware Directory Services) is known for its poor scalability, then they emphasized that Active Directory supports LDAP and DNS (yes, for Microsoft that's a novelty all right) and to cap it all they called Active Directory "secure", I kid you not.

While I'm not sure what they mean with buzzwords like "link tracking" or "integrated content indexing", I've found practically all the above features in Netware 4 since 1993 (!) while NDS has already been scaled to handle billions of objects. On the other hand, Microsoft failed to mention important weaknesses in Active Directory, such as the inability to adequately protect sensitive data from the Administrator account. Granted, Administrator's access rights to an Organizational Unit can be revoked, but the Administrator account can retake those access rights at any time. In other words, it's not possible to adequately shield sensitive data from the Administrator. OK, it is possible to detect unauthorized access (e.g. through a security audit) after the fact. But that's about it. This weakness was also present in Netware's earlier bindery-based architecture, which is one of the reasons why Novell abandoned in 1993 with the release of Netware 4.0 and the switch to NDS.

In April 2001, Microsoft spread the rumor that Novell was moving out of the software business and even managed to get it published by third parties. Microsoft eventually modified the statement in response to demands from Novell. However they repeated this nonsense on 1 October 2001 in a direct mailing to Novell customers. This marketing piece suggested that Novell server products would "expire" at some unknown date in the near future. This is not true; there is no "expiration date" on Novell products, they keep working indefinitely. They also claimed that Novell, after its merger with Cambridge Technology Partners, would discontinue software development and shift to consultancy. They implied that Novell customers would soon be left with a server platform without the full support of its manufacturer. Novell of course filed suit, but much of the damage had been done.

Even in their own certifications Microsoft attempts to misinform their audience. A reader of this paper reports:

I have recently been forced to acquire A+ certification. The content of the exam was weak at best and tested a minimum of skills (after using their prep materials I STILL hadn't found a good explanation of memory timings, but that didn't stop me from getting a perfect score on the exam). It wasn't the sloppiness of the exam that bothered me, though... it was the fact that the whole curriculum is used to peddle MS as the champion of computing (or as the only existing option) and Windows as the only OS a "professional" would consider.
MS through their puppet companies (CompTIA, PrepLogic, etc.) use these exams as written infomercials, and they LIE to do it. For example: The PrepLogic Network+ practice exam had (has?) a question on which Network Operating Systems are X.500 (LDAP) compliant-- apparently Windows NT 4's NTDS and Windows 2000'S Active Directory are considered FULLY X.500 compliant, while "Linux in any flavor is supposed not to have a directory service of this type"... Go figure-- the LDAP daemon I run daily doesn't exist!!!

Rigged tests, distorted reality

In November 2001 Microsoft spread more lies when they published a whitepaper on their website that compared Embedded Windows XP with embedded Linux. Among other inaccuracies, the paper touted the superiority of embedded XP, called it "proven performance and reliability" It blithely ignored the fact that XP is Windows and therefore known for its unreliability, and that XP was brand new and barely tested at the time. It claimed that Linux is "a follower, not an innovator", based on the fact that Microsoft continues to integrate support for new "standards" in their products that the Open Source community struggles to keep up with. The opposite is true, and we all know it.

Nor is this the only example Microsoft's attempts to distort reality. In February 2001 Microsoft's Windows Operating System chief, Jim Allchin, stated that freely distributed software code such as Linux "could stifle innovation" and that "legislators need to understand the threat". The result of Open Source initiatives will be the demise of both intellectual property rights and the incentive to spend on research and development, Allchin claimed. He went on to call Open Source an intellectual-property destroyer, and stated that nothing could be worse than this for the software business and the intellectual-property business.

And it goes on and on: in October 2004 Steve Ballmer wrote an edition of 'Executive E-mail' titled "Comparing Windows with Linux and Unix" in which he stated, among other things, that an MS customer who ran Linux "migrated to Microsoft Windows Server System, and reduced Total Cost of Ownership by 25 percent, consolidated the server population by 50 percent, reduced maintenance time by 50 percent, and boosted developer productivity by 200 percent." I suppose it is possible to replace a freely available, robust and independent Open Source environment with a proprietary, expensive and unreliable product from a vendor known for a sales-driven development strategy... but I can't see how. In any case you'd spend more money rigging the comparison than you'd eventually save.

Ballmer also writes that "A number of third-party reports have questioned how safe the Linux platform really is" and he continues to suggest that Windows is at least as secure as Unix, quoting another success story in which "the core reason for selecting Microsoft was the increase in network security, complemented by the ability to reduce patch-deployment time by 50 percent while cutting unsolicited e-mail by half."

I have to admit that Ballmer (or his ghost writer) has worded it brilliantly. He even manages to pass off the ridiculous amount of security patches released by Microsoft as an indication of how well Microsoft's products are being kept secure. Rather than, say, as an indication of how many security holes there are still being discovered on a daily basis in products that have been in maintenance mode for a long time.

After this nonsense, Ballmer's conclusion comes almost naturally: "it's pretty clear that the facts show that Windows provides a lower total cost of ownership than Linux; the number of security vulnerabilities is lower on Windows, and Windows responsiveness on security is better than Linux; and Microsoft provides uncapped IP indemnification of their products, while no such comprehensive offering is available for Linux or open source." However the actual truth is different. There are lies, damn lies, and Microsoft "facts". Unfortunately whatever nonsense Microsoft publishes is generally repeated indiscriminately in the press.

Microsoft's marketing machine continues to make groundless promises. Windows Server 2003 is being marketed as a huge cost saver. Advertising campaigns promise (without qualification) that you'll "save a nickel on every transaction" just by switching to Windows Server 2003, or that you'll reduce the complexity of your infrastructure by implementing Active Directory and "save two million dollars a year". Of course such unqualified promises are meaningless without being put into any context (such as what kind of infrastructure and ICT environment you have) but that's all irrelevant when Microsoft salesmen step into the boardroom.

Another distorted aspect of the whole Windows Server 2003 campaign to "do more with less" is Microsoft's announcement that their latest and greatest will save you "millions of dollars" because now users can restore their own accidentally deleted documents. Wow. What a great and innovative feature! Users of Novell Netware especially will appreciate it. After all, they've been using Netware's SALVAGE command for exactly this purpose since the 1980's. And this simple feature, long present in a competing product, will save millions on user support now that it's finally been introduced in Windows? Microsoft would have us believe that this makes Windows 2003 a must-have... rather than recognize it as proof that the competing product has been the better one for over 20 years, or as proof that Windows users have wasted millions on user support for lack of such a basic feature. In fact they'd rather not mention this at all. Neither do they mention the fact that no 64-bit version of Windows Server 2003 or XP Professional existed until May 2005, which made Linux the only operating system available on PC-grade hardware (e.g. LAN servers) that fully utilized the power of 64-bit CPUs. Even today, the 64-bit version of Windows consists mostly of 32-bits code, just like earlier 32-bits versions of Windows were mostly 16-bits code under the hood. Yet Microsoft would have us believe that, in comparison to true 64-bit products with a proven track record, Windows is the superior one.

While all this is going on we keep seeing Microsoft-financed "research" that pronounces Windows both superior to and cheaper than Linux, in spite of independent research, daily experience and common sense proving the opposite time and again. Why is it that Microsoft has to pay researchers and analysts before they'll come up with conclusions that favor Microsoft?


Releasing biased or distorted reports on the ostensible benefits of Microsoft products is not the only tactic by which Microsoft attempts to manipulate mass opinion. Occasionally an objective (and highly damning) report on the real state of affairs is being released, in which case Microsoft's preferred response is to have it quietly removed from public view.

For example, Bloor Research once compared Microsoft's database engine with its main competitors, and tested DB2 on AIX, DB2 on NT and MS SQL Server 6.5 on Windows NT. The report was published under the title "The Realities of Scalability" in March 1997. It is still, today, an impressive body of work. Over 130 detailed pages of complex tests really put the three database engines through the wringer. Bloor tested their performance under many different conditions and performed a rigid statistical analysis on the results to determine their significance. The resulting conclusion was highly critical of MS SQL Server 6.5, especially in comparison with the other two contestants. It found SQL Server to be seriously lacking both in scaleability and reliability, cited a number of repeatable failure states, and used the words "dramatically worse" when comparing it to the alternatives.

The report was quietly suppressed. The acid test is looking on Bloor's own website. You will find an archive there that does, indeed, go back to 1997, but there is no record of any database scalability report.

Unfounded accusations

In May 2001 Microsoft CEO Steve Ballmer stated in an interview that "Linux is a cancer that attaches itself in an intellectual property sense to everything it touches". Even better; the user license for the second beta version of Microsoft's Mobile Internet Toolkit came with a condition that the product not be used "in combination with potentially viral software". The document went on to name examples of what Microsoft considers "potentially viral software": any software distributed under the GNU Public License (the most common license for Open Source software) and also the Lesser General Public License, the Mozilla Public License and the Sun Industry Standards License.

Spreading FUD (Fear, Uncertainty and Doubt) is a staple ingredient of Microsoft's ongoing attempts to damage competing products. A good example is the rather shameful affair of Microsoft hiring SCO for over $60 million to make bogus claims of copyright infringement by Linux and threaten Open Source and the Gnu Public License in general. Of course SCO proved unable to produce a single shred of evidence to back up their accusations, and the whole matter was swept under the carpet in the usual fashion... until 2007, when Microsoft did it again, when Steve Ballmer accused Linux of violating hundreds of Microsoft patents, and urged Novell users to purchase a license from Microsoft in order to prevent legal repercussions. If at first you don't succeed, spread some more FUD...

The truth emerges

Microsoft's track record speaks for itself. Decades of non-innovation and monopolist practices. Legal procedures, lies that border on perjury (and perhaps even cross that line) and finally a ruling by the DoJ that Microsoft mostly ignored. Nothing but rewrapped old technology from competing products, touted as the hottest thing since sliced bread and even marketed as a cost saver. Nothing but misinformation, FUD and outright lies on web pages aimed at the ignorant. Suggestions that TCP/IP is a Microsoft protocol, claims that the integration of Internet Explorer and Media Player in Windows is a technical necessity, unrealistic promises about cost savings and reliability, slanderous untruths about competing companies and products, and attempts to paint the Open Source community as being fascist and full of copyright infringements.

Interestingly, though, Microsoft's own actions show the truth. Internet Explorer 6 was released years ago, and has been in maintenance mode ever since. Only in response to the hugely popular Mozilla Firefox webbrowser (an Open Source project) Microsoft decided to start development of IE7, with the deliberate intention of adding many features that IE6 lacked and Firefox has. Not only does this prove that freedom of competition (and not a monopoly) drives innovation, but it also clearly shows that Open Source drives innovation and even causes Microsoft to follow these innovations. It proves Open Source to be not a cancer but rather a cure. The fact that IE7 only runs on XP and Vista and not on earlier Windows versions or competing operating systems, on the other hand, merely points out one of the ailments in need of such a cure.

Meanwhile the lies continue. In fact Microsoft has been proven guilty of most of the things they accuse their rivals of, and then some. Can you say "dishonest"? It's spelled M-I-C-R-O-S-O-F-T.

Table Of Contents | Previous chapter | Next chapter

Comments? E-mail me!


Contents copyright © 1997-2007 F.W. van Wensveen - all rights reserved.