I was reading MetaFilter this morning, and came to this comment in a thread about the root pwnage hack of the PS3, regarding why people harbour animosity towards Microsoft. The comment is a sincere query, coming from a place of ignorance.
It is interesting to me that enough time has now passed that there are millions of people, presumably in their early twenties (or younger), who missed bits of history that I just take for granted. It’s an odd feeling, since I’m only in my early thirties and don’t feel old.
The Berlin wall came down in 1989, but if you were born in that year, you’d be 21 now. That still amazes me.
jbc has a great summary of why people don’t like Microsoft as a comment in the MetaFilter thread, but there’s a lot more to it, which I’d like to write about.
So let me give you a summary of why a bunch of people harbour a lingering resentment towards Microsoft.
When I was very young, my father bought an Apple IIe computer. This was the mid-‘eighties, and there were many different personal computers around: Spectrum, Amstrad, BBC Micro, Commodore 64, Amiga, the Apple II. Of these, the Apple II series, Commodore 64 and Amiga were dominant, mostly because of the games. It was much like the current console gaming setup, with systems from 3 major companies, mostly running different games, competing for your attention.
IBM was the evil empire back then, and a 1984 ad from Apple may help to illustrate the hold they had on the business computer market. This is a classic ad, and if you haven’t seen it before, you really should click over and watch it. I’ll wait.
But then Microsoft arrived.
This article at Wikipedia gives you some background on the introduction of Microsoft’s MS-DOS operating system for the IBM PC.
By the early ‘nineties, the Commodore 64 had fallen by the wayside, the Macintosh had superseded the Apple II line, and the likes of Spectrum and Amstrad were just a vague memory.
The IBM PC used Intel’s 8088 processor, and a more-or-less open architecture for the computer system. Apple used a Motorola 68000 series chip, and a very, very closed approach to the architecture. Microsoft supplied an operating system to IBM for their IBM PC architecture. IBM’s approach to the architecture (the IBM PC/AT) worked better than Apple’s, as numerous ‘clones’ as they were called, sprang up in the marketplace. Apple firmly crushed any attempts to clone their architecture and sell competing hardware, mostly through lawsuits, a habit they continue today.
Apple computers were easier to use, but expensive, and designed for creative types, not business users. You couldn’t buy an Apple clone. You didn’t have to buy a computer from IBM to run MS-DOS, you could buy a clone, which could still run all the same software. IBM responded with the IBM PS/2 architecture, which couldn’t compete with all the clones, and eventually died (leaving remnants, like the PS/2 mouse interface).
The end result was the the IBM PC beat out everything else for market dominance, which meant MS-DOS.
The Age of FUD Begins
Competition between operating systems was… different, back then. I the early ‘nineties, the choice was really PC architecture (same as now). You were either Apple Macintosh, or IBM PC. On Apple Mac, you got the Macintosh operating system. On IBM PC, you had DOS.
But there was more than one DOS.
MS-DOS was dominant, but not like today. There were competing versions of DOS that were compatible for all the base features, so all your software would run on any of them. But different types of DOS had different features, some of which were pretty neat. This was around MS-DOS 4. Digital Research had DR-DOS, which had extra features not available in MS-DOS, or only as third party add-ons, which cost more.
Now Microsoft was working on a new operating system to take over from MS-DOS, developed in partnership with IBM: OS/2. Never heard of it? Not surprising. It was over-hyped and fairly useless when it eventually arrived, and sold really badly. There was a falling out between IBM and Microsoft partway through the development of OS/2, and Microsoft went off on their own to develop what would become Windows.
But until then, they didn’t actually have a product to compete with DR-DOS. So they did what Microsoft has always been really good at: Marketing. They talked up the features of MS-DOS 5, and people stalled on buying DR-DOS 5 to see what MS-DOS would look like. Which didn’t ship until well after (like, over a year) Microsoft started talking about what was going to be in it. They did it again after DR-DOS 6 came out, and hyped MS-DOS 6.
The issue was that if MS-DOS was incompatible with DR-DOS, the people using DR-DOS wouldn’t be able to run the same software as the MS-DOS folks. Big business was heavily IBM-PC and MS-DOS dominated, so DR-DOS and incompatible software would have fragmented the market, which really concerned a lot of business people. So they’d adopt a ‘wait and see’ approach, which suited Microsoft just fine.
Windows 3 was the first ‘real’ version of Windows as you would know it today. Before then, everything was text based, at least for the majority of users. Apple had had a graphical user interface for years, which is a major reason all the creative types used it. But if you’re an accountant, you don’t care about pretty icons, you just care about columns of numbers.
Windows 2 was interesting, but just abysmal to use, so it never went anywhere. But Windows 3.1 just exploded into the market.
Windows 3.1 introduced the idea of being able to run more than one application at a time… sortof. It wasn’t truly multi-tasking, but it felt a lot more like it. Prior to Windows 3.1, if you wanted to run a different program, you had to completely leave the one you were currently using. In Windows 3.1, you could ‘task switch’.
This had been available on Apple Macs for years, but business people didn’t use Macs. With Windows, though, now business people could change from their accounting application into a word processor, and back again, without all the tedious shutdown and startup time.
Windows actually ran on top of DOS at the time, so you could still run DOS programs, or you could start up Windows and do things in there. Over time, people would just configure .BAT files to automatically run Windows at boot time, and DOS gradually died off, and DR-DOS with it.
IBM’s abortive attempt at launching OS/2 meant Microsoft came to dominate the market with Windows. Apple was a tiny niche player, and there really wasn’t any other choice for a home PC.
Enter Anti-Trust Law
This is a massive topic just on its own, so I’ll only give you a ludicrously simple summary for the purposes of this post.
When a company completely dominates a market, there’s a risk of it becoming a monopoly, i.e.: you have no competition, so you can basically do what you want. What are your customers going to do? Buy from someone else? Hahahaha! Here, let’s just raise the price of drinking water by 8000%.
Because unscrupulous people often do bad things if they have a monopoly, various laws have been introduced to handle this form of market failure. In the United States, this is called anti-trust law, and the government agency responsible for regulation of competition is the Federal Trade Commission.
What they attempt to do is to safeguard consumers by ensuring that companies either a) don’t abuse their market power, or b) don’t get to keep a monopoly.
Much of this law was introduced after poor behaviour of the American Telephone and Telegraph company which had a monopoly on long-distance telephone service in the 1970s.
At this point in our story, around 1990, Microsoft were essentially a monopoly, and the Federal Trade Commission thought that it was worth investigating. This was the beginning of more than a decade of legal wrangling about whether or not Microsoft abused their market power, and multiple anti-trust lawsuits, both in the US and in Europe.
Back in the early 90’s, the Internet was virtually unheard of outside of academia. PCs had no software for connecting to it, and Internet Service Providers were unheard of. Back then, Bulletin Board Systems, or BBSes, were all the rage. The World Wide Web didn’t even exist until December 1990. Everything was text.
Enter Mosaic, the first real ‘web browser’. It was the first mass-appeal browser and introduced the general public to the World Wide Web in 1993. Mosaic begat Netscape Navigator in 1994 and the Internet took off. Windows users joined their mini-computer cousins in academia in enjoying graphical information display from all over the globe. Internet commerce took off, and really, the world hasn’t been the same since.
What’s interesting to me, is that I see preciously little commentary these days about how Microsoft nearly missed the boat on this one. They essentially ignored the Internet for the first half of the 1990s. It took a missile from Bill Gates himself to get the company to start looking at the Internet at all.
Let me impress upon you how big a deal this was. Back then, Windows didn’t have an IP stack. You had to install something called Winsock to be able to connect to the Internet at all. Windows had zero Internet capability. None.
Microsoft, to their credit, turned on a dime and went mental for the Internet. They released Internet Explorer 1.0 in mid-1995 (which sucked) and three months later released IE 2.0 as a completely free download. IE 2.0 also sucked, but Netscape Navigator was only free for non-commercial users. IE 2.0 was free for everyone, even business users.
And so began the Browser Wars.
By 1996, IE 3.0 started to get close to what Netscape could do, but it wasn’t until IE 4.0 in 1997 that Microsoft finally made real headway. By bundling IE 4.0 into Windows for free, Microsoft used their dominant market share in the operating system market as leverage into the browser market. And won. In 1998, the US Department of Justice started an antitrust case against Microsoft that was to go one for more than a decade.
Back in the DOS days, there was a word processing program called WordPerfect. It still exists, but as a shadow of its former self. In the early 1990’s, it was the dominant force in word processing.
But then Windows happened, and WordPerfect was slow to adapt to the new interface. New word processors entered the market once dominated by WordPerfect: Abiword, and the now-ubiquitous Word for Windows.
Once again, Microsoft bundled Word for Windows with the operating system, and when customers saw they already had a word processor installed, why would they shell out for another one unless they had a specific need?
Word for Windows took over, partly due to bundling, but partly because WordPerfect dropped the ball on providing a product that met their changing customer’s needs.
A result of Word for Windows taking over is that now its (proprietary, obtuse) file format became the de-facto standard in the late 1990s. Due to the way it was implemented, each upgrade of Word for Windows would break compatibility with previous versions. If someone sent you a Word 97 file, and you only had Word 95, you couldn’t read the file. This was a major frustration at the time, for both home and business users, to the point that file converters were written to convert from one Word file format to another.
Since then, Microsoft has actively opposed attempts to define an open standard for documents, using their by now classic ‘embrace, extend, extinguish‘ strategy. Seeing that the OpenOffice standard for document formats, the Open Document Format, or ODF, was about to be ratified as an ISO standard in 2006 (which it was), Microsoft began to aggressively push their own, alternate standard: OOXML.
There much much wrangling and controversy, but OOXML was also accepted as an ISO standard for documents. So now we have two. That’s the wonderful thing about standards, there are so many to choose from.
Only the ISO standard isn’t what Microsoft are actually shipping. No. Now that they’ve achieved the urgent goal of positioning themselves well, they can relax and take their time. Don’t expect standardised OOXML for 5 years.
Microsoft introduced a networking server based on the Server Message Block protocol/system in Windows for Workgroups back in 1992. This was a way of sharing files and other information between Windows machines over a network, usually NetBIOS over IPX, which you’ve probably never heard of, because it was killed off a decade ago.
Again, this is back in the day when Microsoft were ignoring the Internet, and so didn’t have an IP stack. They were primarily competing with Novell Netware, which in the early 1990s was the only way to connect multiple DOS or Windows computers together. Netware is also interesting in that it really kicked off the idea of industry certifications with the CNA: Certified Netware Administrator.
Netware pretty much invented the idea of ‘file and print’ for groups of PCs back in the DOS days. You could connect your PC to a Netware server, and share files, or print documents to shared printers. You couldn’t do this directly from Windows until Windows for Workgroups came along. Netware was a big deal in companies small and large.
Windows for Workgroups made some headway, but it wasn’t until Windows NT 3.5.1 that things really started to go Microsoft’s way. Novell had dropped the ball on improving their software, and didn’t adapt to the Windows world fast enough. Netware was also slow to move to IP from IPX/SPX as their transport protocol.
So Microsoft took over again, and immediately consolidated their position to exclude others. This was in the mid-1990s.
In 1991/92, Andrew Tridgell released Samba, a product designed to provide file serving capabilities from a Unix server, using the SMB protocol. By the mid-90s, Microsoft had embraced and extended SMB to include a variety of proprietary extensions and changes. To build something compatible with Microsoft’s implementation, Andrew used a process he dubbed the ‘French Cafe‘ analogy. Compare and contrast with the method used to build the original versions of Samba in this little bit of history.
By 1998, Microsoft were once again dominant in an area dealing with personal computers and in danger of breaching antitrust law. Which they did, as you can see here. 9 years later the European Commission ruled against Microsoft, and the protocol documentation was ordered to be made available.
Well, everyone who uses a computer should, particularly if we consider what might have happened if Microsoft hadn’t abused their market power. When a monopolist abuses their power, customers all lose, because they don’t get to enjoy the more rapid improvements that robust competition provides. It’s one of the key reasons we think competition is a good thing.
We might not have been stuck with IE6 for nearly a decade now. Once Microsoft won the browser wars, they had no incentive to improve their browser. So they didn’t, and IE6, with its buggy, flawed implementation is the bane of web developers the world over. Countless hours have been wasted developing for a product that is provably inferior and should have been replaced with a better one by now. Only it wasn’t until other competitors eventually sprang up, like Firefox, that we finally saw some movement from Microsoft.
We might have had a competitor for file and print services that drove Microsoft towards making their own systems more stable, and admins the world over wouldn’t have had to reboot their print servers every few days for several years. We might have had a few competing Active Directory implementations.
We might not have been stuck using Microsoft Office formats, further reinforcing Microsoft’s dominance of the operating system market.
But lastly, and this is the big one for me, we might not have a monoculture of operating system on the Internet with such a poor security model.
Mainframes and Unix have had multi-user systems for decades, but Windows used the one, all powerful user, for everything up until Vista. Technically, you could have used multiple users since NT or XP, but most user programs assumed a single, all-powerful user, and required Administrator access. Couple this with the integration of IE6 into the operating system (to help with excluding competitors), and you have a gift to virus writers and spammers the world over.
Imagine a world where Symantec didn’t exist, because viruses weren’t so easy to write and spread to all the world’s computers. Imagine a world where spam didn’t constitute 90% of all email because it wasn’t so easy to take over a PC and turn it into a botnet zombie. Imagine not having to do impromptu tech-support for family members who accidentally installed a bunch of spyware.
Imagine all the time and money that has been, and continues to be, spent on fixing all of the issues that a better security model 10-15 years ago might have avoided.
Microsoft have made (or bought) some excellent products, as they continue to do. There are many wise, capable, and perfectly reasonable people who work there, what with it being a big company and all. This is not a company that is an unrestrained force for evil in the world.
Microsoft have a history of abusing market dominance in order to exclude competitors. Many of the top management running the company at the time are still there, running the company today.
Perhaps there will be no repeat performances, but there are very good reasons for greeting rhetoric from Microsoft regarding their openness with some scepticism.
Inflammatory headline aside, let me be clear that I don’t hate Microsoft. But I can understand why there are those who do.
It’s too late now to do anything about what happening the past. But something we can do is to learn the lessons of history and attempt to prevent it again in the future.
Prevention is better than cure, after all.
On the flip side, it’s an excellent set of case studies in How to Be Evil for Fun and Profit, if you were so inclined.