This article skips over the direct role that United States v. Microsoft had in creating space for innovative companies like Apple to succeed. Without the US anti-trust intervention it's quite possible Microsoft would have totally embraced, extended, and extinguished open Web technologies and subsumed the Web into a Microsoft proprietary platform. Macs would not be nearly as viable a product if Internet Explorer were the only browser that worked with most web sites.

Another important part of the history is Microsoft's June 1997 $150m investment lifeline to Apple when Apple was worth less than $4B. That investment predates US v Microsoft but at the time it was widely believed that part of Microsoft's motivation was ensuring Apple didn't go completely broke and thus invite even more monopoly scrutiny to the PC business.

In what way did the anti-trust case help other companies? Or how did the bundling of IE hinder other companies like Apple? I was too young and in a country without internet access at the time to remember what the deal exactly was. But it seems to me that having IE or not on my PC would in no way have affected other companies. I mean, with Firefox and later with Chrome, people switched to it mainly because they were better browsers, not because IE wasn't on their PCs already.

The problem was that Microsoft introduced proprietary extensions to standard HTML that would only work properly in IE. They first tried to create a monopoly on browser market share. By bundling IE with Windows 98 as the default browser, they succeeded in gaining 90%+ of marketshare just by default, which pretty much destroyed Netscape overnight. Then, they started to release proprietary extensions to HTML like ActiveX that conveniently would only run on a Windows computer running IE.

If they would have succeeded in this strategy, we would have a rich web experience with "web 2.0" applications, however, the protocols would all be proprietary and you would only be able to access them on an x86 computer running Windows and IE. Their strategy of embrace and extend was hugely successful, and you can see the aftermath all around you. Look at the extraordinary lengths developers have to take to support older versions of IE. Up until the justice department rulings in the early part of the 2000s Microsoft was still making subtle tweaks to try and break standard IETF protocols. They had a willful disregard for RFCs and always tried to add Microsoft only extensions on top.

The article is interesting in that he highlights the end result is that innovation trumps monopoly, however, he ignores the damage this caused to both users and entire companies that were literally wiped out by these anti-competitive practices. WordPerfect, Novell, Netscape, and countless other innovative companies were extinguished by this.

> The problem was that Microsoft introduced proprietary extensions to standard HTML that would only work properly in IE.

Microsoft wasn't doing anything that Netscape wasn't doing. Both of them were introducing their own HTML extensions left and right. Each was free to implement the extensions the other introduced.

> They first tried to create a monopoly on browser market share. By bundling IE with Windows 98 as the default browser, they succeeded in gaining 90%+ of marketshare just by default, which pretty much destroyed Netscape overnight.

That's not correct. What destroyed Netscape was the fact that IE was a better browser. IE did a better job of implementing standard HTML, and it did a good job of implementing Netscape's non-standard HTML extensions, and it performed better.

It's easy to say Microsoft won by bundling, but if you look at sales numbers from before bundling, when the competition between IE and Netscape boxes on the shelves at Egghead and CompUSA and the like, you'll find that IE dominated. I don't remember the exact numbers, but I remember them being massively in IE's favor. (My employer had retail software for sale, so we subscribed to a report from a company that had an arrangement with all the major software retailers to get sales numbers, and they produced a nifty monthly report showing pretty much ever item available at those stores and how many units were sold every month. Browsers were one of the items I watched monthly on those reports).

Sales numbers for boxed copies of Netscape/IE are irrelevant because in 1998 everyone just used the browser that came with Windows. You had to intentionally try to download Netscape. In fact, you couldn't even download Netscape without using IE to do so.

I will admit that when IE 4 finally came out it was better than Netscape 4 (minus the active desktop crap that bogged down the 64MB RAM systems of the era), however, previous versions of IE were far inferior to Netscape.

Proprietary tags in Netscape? Yeah, I guess if you consider the <blink> tag proprietary, sure. They can be counted on both hands. IE had entire proprietary APIs, not just a few tags. There's a huge difference.

Netscape, in addition to <blink>, added <font>, <basefont>, changing the text color via an attribute of the <body> tag, attributes to change numbering style for <ol>, the height and width attributes of <img>, the background attribute of <body>, <center>, a bunch of attributes to <hr>, <ilayer>, <keygen>, <layer>, <multicol>, <nolayer>, <server>, <spacer>, <nobr>, <noembed>, <wbr>. <big>, <small>, <sub>, <sup>. I'm not sure if this is an exhaustive list.

I recall that Microsoft became very cautious about executing embrace and extend tactics to decrease their profile for potential anti-trust cases.

My guess after working at organisations under enhanced regulatory scrutiny, the distraction was probably felt most seriously in the upper cadres of the company as they would have had to critically evaluate acquisitions, new products as well as coordinate the legal and PR.

The decrease in ravenous embrace and extend activities left large numbers of smaller businesses with viable business models, and ensured that the Internet remained fairly neutral for non-Microsoft software (although I still struggle to this day with intranets coded with VBScript for basic things like timewriting: if this had happened on the free Internet, it would have been significantly harder for the Firefox and Web 2.0 people to get any kind of traction...)

It's impossible to conclude from the evidence presented that the antitrust action had no effect on the breaking down of Microsoft's monopoly. It's entirely plausible that the enforcement left them unable to respond as quickly to new markets as their competitors, leading to them beginning to get undermined a few years later. Abusing their monopoly in the PC OS market to get advantages in unrelated markets was in many complaints against Microsoft at the time.

It's entirely within reason to believe that had the Justice Department not been breathing down their necks that Microsoft would have used its monopolies to illegally squash the original iPod. There are plenty of examples of products that were well-made, innovative, and desired but didn't make a major market impact for poorly-understood reasons (WebOS comes to mind). It's not at all obvious that the iPod was bound to succeed; if all those amorphous outside factors hadn't gone its way, we could be sitting around discussing how the iPod was a repeat of the Newton, and how inevitable it was that Microsoft finally cracked the the MP3 market with Plays4Sure.

> It's impossible to conclude from the evidence presented that the antitrust action had no effect on the breaking down of Microsoft's monopoly. It's entirely plausible that the enforcement left them unable to respond as quickly to new markets as their competitors, leading to them beginning to get undermined a few years later.

In fact, companies targeted by antitrust (I think it was IBM) have stated that antitrust regulation, by itself, shifted their focus. You are no longer rushing ahead, instead you must be cautious about everything. So it is indeed quite plausible that antitrust actions against Microsoft made possible the current non-Microsoft-dominated era.

Apple was on life support during the mid-90's. Microsoft nearly killed the Mac with its unimaginative imitator, Windows, but probably held off from killing Apple completely because of the Antitrust case hovering over it. Many if not most of the early personal computer software companies were driven out of business by Microsoft.

This article misses the forest for the trees. During the height of Microsoft's monopoly, consumers were very much limited in their choices. MS did everything they could to keep the likes of OS/2 and BeOS off of new PCs, and that means no one ran other OS's, so no one wrote apps for them. Had that not been the case, the PCs, phones, and tablets we use today would be very different. I daresay they'd be better, for having arrived at many current innovations sooner due to the more robust level of competition and potentially having more viable options in the market today.

Just because MS lost its monopoly on its own does not mean that we, as consumers, would not have benefitted--and would still be benefitting--from breaking up that monopoly when it was more dominant.

Except Microsoft's leverage in operating systems never really carried over to other hardware platforms, like phones and PDAs.

I'm not claiming it did. I'm saying that a very different PC desktop / laptop OS and software ecosystem back in the mid-nineties would have ripple effects on PCs, phones, and tablets today.

Yea, the entire MS OS/2 2.0 fiasco is one of my favorite topics. Why did it take ten years after Intel introduced the 386 before 32-bit programming became popular? And remember what MS did to kill DR-DOS? Guess what, OS/2 never depended on DOS!

University of Chicago economics professor says government regulation not needed? Why am I not surprised?

I know what you're getting at, but one could argue that the market indeed took care of the problem (if you consider it to be a problem) of Microsoft having such a large market share. Apple innovated, and Linux found a place in Android. Suddenly Microsoft is playing catchup.

The market took care of the problem only because the antitrust injunction prevented Microsoft from dominating the market in its customary manner. Without that injunction, Apple would be dead today. (Microsoft's investment in Apple during the antitrust proceedings is the only reason Apple survived long enough to push out the iPod.)

Indeed, a lot of what Apple is lauded for today (i.e., integration) is stuff that Microsoft was explicity prevented from doing under the terms of the antitrust injunction.

Microsoft's investment in Apple wasn't mere charity, if we're to believe this anecdote: http://www.forbes.com/sites/ericjackson/2012/03/01/steve-job...

Things like the iPhone weren't going to come from Microsoft. Rather than beating them in the desktop market, which Apple still hasn't done, it won on a different battlefield.

That was my point -- without the antitrust injunction, Microsoft would not have invested in Apple. Ergo, Apple owes its existence today to the injunction.

The article I linked never mentioned the injunction. Apple was suing Microsoft, and the investment by MS was part of how Steve Jobs offered to settle the matter.

I wonder how many people who argued about Linux vs BSD in early 90s would have predicted that 20 years later, in the post-PC-era, these would be the two most popular operating systems by far, backed by the two biggest post-Microsoft tech companies, Google and Apple.

I remember it being kind of the opposite to how you describe it. In the 80s and early 90s, unix enthusiasts could be smug about the rise of unix being inevitable. Maybe that wouldn't have been true of linux ppl. But unix systems didn't meet user needs, and unix got sent into the wilderness. Twenty years on, X-windows still shits me for the same reasons it did back then. Why is configuration so awkward? Why is the clipboard system so awful?

I think plenty of people in those segments believed those operating systems would be huge, but they missed by a mile when it came to the form.

I remember arguing with slashdot Linux fanatics at the turn of the century, and they stubbornly believed that somehow magically Linux as it was at the time would go mainstream on the desktop and vanquish Windows. It was always an absurd premise. It almost always takes a paradigm shift to leave a big enough opening, and mobile provided that (the desktop still being ruled by Windows).

What's interesting is they're not really competing.

BSD for desktops, Linux for servers. Not the split I would've guessed 10-13 years ago.

My point was that Android is based on Linux and iOS on BSD. So the split really is BSD & Windows for desktops, Linux for servers, and BSD & Linux for phones. Who would have predicted this?

I'm pretty sure the only thing Windows can compete on, on the desktop, is price. That has to really suck.

There are very few niches/users left that actually want Windows > Apple, and most of them begin with "gam" and end with "ing".

Windows still owns the enterprise. Active Directory / Group Policies still beat anything you can do with OS X, and IT groups have 10 years of experience they don't want to walk away from.

But, users are pulling these organizations, kicking and screaming into a world in which everyone wants a MacBook Air/Pro.

I really don't envy Microsoft.

Could you elaborate? I'm not particularly familiar with Active Directory, so what makes it so much better than using Open Directory on OS X? Wikipedia says Open Directory even has "Active Directory domain member support", whatever that is.

There's almost no comparison.

Open Directory is basically a ldap with MIT kerberos that can sort of integrate with AD. Most folks setup a hybrid (referred to as the "magic triangle") where you use AD for user auth and OD for computer policy stuff. It's not very robust, and Apple isn't exactly committed to the notion of server operating systems, so it doesn't get much love.

You really need to buy a client app like ADmit Mac, or a server-side solution like Centrify to really work with Active Directory.

AD itself is a really brilliant collection of software. Think of it as a fully integrated, smoothly operation bundle of LDAP, Kerberos and DNS -- it "just works".

Coming from a Unix background, AD was amazing to me as there isn't any Unix equivalent at all, except maybe for the ancient NIS and NFS integrations. You get single sign on. Robust Access Controls. The ability to trust the identities of external organizations. A policy engine to do all sorts of management activities at various levels. Oh, and it allows you to easily build fault-tolerant deployments without much forethought.

The reason that AD is so needed is the ease of implementation with EVERYTHING, as long as its MS(ish) based. One AD account can auth a user to any service running on the company systems. For example, I work for a university as an applications dev; My local box login is the same for every VPN and additional university service that we use.

Now, you can do this with OD, but it's not as easy as "built into all windows installations, set it once and forget it".

Except there it isn't so much Windows competing on price, as it is the hardware, which is a commodity.

The way Microsoft prices Windows, at least at retail, and the way they segment their OS features for different markets, are really annoying and somewhat painful.

I'm surprised by the large Android figures there. For a time OSX became the most widely distributed unix-like consumer system and that was exciting. But despite a strong play, it's being overtaken by android. Looking at these numbers, if I was starting a consulting business at the moment I'd be tempted to focus at android.

Fractured, commodity market, lots of competition, lots of dead ends (failed devices, companies)

Vs

Unified, Premium market, no hordes of java devs jumping bandwagon, less porting, less chance of backing loser.

Tell Samsung Android is a dead end. They're generating extraordinary profits off of Android right now.

900,000 Android device activations per day

Seems likely that the Android party will continue to be a big fractured mess, but at that kind of scale there's always opportunity.

The rebirth of Netscape navigator, JQuery and XMLHttpRequest happened to the Microsoft Monopoly.

Microsoft created the XMLHttpRequest standard; it was first featured in Internet Explorer. Microsoft was also one of the first backers of jQuery.

Mozilla owes much of its market share to the EU, which forced Microsoft to bundle competing browers with Windows, and to the collapse of Netscape in the first dotcom bubble.

Mozilla owes almost none of its market share to the EU. We earned almost all of our marketshare well before the EU decision requiring microsoft to advertise (not bundle) alternative choices and since that change in the EU a couple of years ago, we've seen very little movement in our user numbers. It may prevent IE from getting as many users, but it splits those users among Firefox, Chrome, Opera, Maxthon and others.

Yeah MS fucked up enormously when they created XMLHttpRequest - that is what is so special about it.

So the interesting question is, why didn't Microsoft hop into the Web with both feet in the late 1990s back when it owned personal computing via Windows monopoly? Why did MSIE stall out at IE6 for so long?

Hindsight is always 20-20.

We can only speculate at this point, but if we take a look at the timeline for the browser wars, we get some interesting insight.

From the mid 90's through 2000, we had Netscape as a serious source of competition for MS. MS released IE 1-6 between 1995 and 2001. That's 5 versions inside 6 years. Netscape was bought out by AOL in 1998, thus ending the need for MS to continue improving IE6. Now, there were other browsers at that time, as Netscape opensourced the browser code and gave it to Mozilla, but they all failed to achieve traction. Throughout the 5 year period of IE stagnation, Mozilla had the time to start the war machine up again and start pushing out new code. MS knew that FF was coming, and they knew the feature set that FF supported. The answer to this was IE7 in 2005, however it was pretty much too little too late. FF2.0 was released in 2006, only about a half year after IE7. It had all the same features (plus a few additional) but guess what? It had an intuitive UI and worked well, unlike the MS counterpart that still reeked of R&D stagnation. Not to say that MS didn't do good things for the web, just that, without another company to light the fire under their ass, they failed to innovate. And that's pretty much where we are now: Microsoft is trying to build a browser that actually works (IE9 isn't a total failure), and Google/ Mozilla keep them on their toes. It's a win win for the consumer.

I have to admit that graphs using logarithmic scaling can come quite handy when trying to make a certain point.

We should be happy Microsoft is not a monopoly anymore, and instead of trying to encourage them to come back to that position, we should encourage the competition to even things out some more.

The only way Microsoft, like IBM before it, could have held onto its market position was if the government stepped in and specifically protected it.