Hacker News new | ask | show | jobs
by paulojreis 4197 days ago
Well... you know what a browser is (and you also know that there are alternatives). Most people don't. Most people don't open the browser, they "go to the Internet"; the browser is the "Internet".

If a dominant player bundles their browser (as a component) in their OS, people become less and less aware of the browser as a possible and separate "product", and will of course use the dominant one as default - this is (and effectively was!) bad for competition and evolution(any web developer can tell you the pain this caused and still causes today).

2 comments

But the same is true for the calculator app, or the file manager. Granted, the web is used for a lot more than those... but I don't see why the browser is singled out as something needing special legal treatment.
> but I don't see why the browser is singled out as something needing special legal treatment

Does your calculator interact with anything else but your computer? Same with file manager (SMB excluded)?

Now lets look at the browser, it interacts with different servers, via the HTTP protocol, and the documents they serve via many different technologies such as HTML, Javascript, Java, XML, Flash, along with others. 'Back in the day' Microsoft did a very strong push to tie IE in with Windows, then push developers to use ActiveX on websites so they would only work with Microsoft products. There was no HTML5 multi-platform interoperability, if you wanted to visit a huge number of sites (many government ones too), you used Windows on X86, and you used IE. Even worse, after the early versions of IE, they dug its hooks deep in to the operating system so it became WindowsIE, you didn't have one without the other.

I don't quite see what interaction over the Internet has to do with the issue. Microsoft Office is regularly bundled with new Windows computers as a trial, or basic license. Could that not be seen as anti competitive?

I could argue that the TCP/IP stack is "bundled" with the operating system, and it's something that's very difficult to change (if not impossible without recompiling the kernel). Network engineers have probably had to put up with different TCP/IP stacks using different congestion control algorithms, impacting fairness of bandwidth usage on their networks. So by the same pretense, shouldn't TCP/IP stacks require the same choice that Microsoft were required to offer with browsers?

There wasn't a whole lot of evidence that MS was trying to explicitly kill for-profit TCP stack vendors.
Well that's true, my straw man doesn't quite stand up in this case. However it still doesn't explain why Internet Explorer was the target when there is so much other software that was bundled with Windows, which too had for profit competitors? Those people who exclusively used Internet Explorer as "the Internet" would too have used the bundled mail client (I can't remember what it was called back in the day) exclusively too and seen that as "the email"?
I don't think there was anything else bundled with Windows that explicitly targeted other for-profit competitors at that time.

Dr-DOS and the Berkley screen saver stuff are two other situations that spring to mind, both of which resulted in legal cases, but the size/scope of MS by the time the Netscape thing came around was probably the tipping point (considering their previous behavior already demonstrated a pattern which showed no signs of stopping).

The original complaint for the US case remains available online from the DoJ: http://www.justice.gov/atr/cases/f1700/1763.htm

There are four bases for the complaint, two under Sec. 1 of the Sherman Anti-Trust Act and two under Sec. 2 of the Act.

1. Unlawful Exclusive Dealing and Other Exclusionary Agreements in Violation of Section 1 of the Sherman Act (deals with ISPs, Internet Content Providers, and others that traded favorable promotional placement in the then-monopoly Windows OS for commitments not to license, promote, or distribute non-Microsoft products, particularly browsers.)

2. Unlawful Tying in Violation of Section 1 of the Sherman Act (distributing the browser, a separate product based on market demand, etc., with its monopoly Windows OS system.)

3. Monopolization of the PC Operating Systems Market in Violation of Section 2 of the Sherman Act (willfully maintaining monopoly power over that market by "by anticompetitive and unreasonably exclusionary conduct.")

4. Attempted Monopolization of the Internet Browser Market in Violation of Section 2 of the Sherman Act (targetting "software products that have the potential to compete with or facilitate the development of products to compete with PC operating systems and thereby to erode Microsoft's Windows operating system monopoly" by way of "including tying and unreasonably exclusionary agreements, in order to obtain a monopoly in the Internet browser market".)

Note that a key issue in all of these is Microsoft's then-existing monopoly power in the PC OS market and how it was defending or leveraging that in different markets -- without that monopoly power, much of what they were doing would not have even arguably been an antitrust issue.

Browser+OS wasn't the issue, it was one of the mechanisms -- unreasonable and anticompetitive action to maintain an existing monopoly, and to leverage that monopoly into a monopoly in another market, was the issue.

> However it still doesn't explain why Internet Explorer was the target when there is so much other software that was bundled with Windows, which too had for profit competitors?

Because the browser was the area where the government felt they had the best case for two key features: (1) that the tying, etc., was intended to leverage the existing monopoly to monopolize another market, and, (2) that the tying, etc., was intended to neutralize a threat to Microsoft's existing monopoly.

Bundling two products together isn't generally illegal. Doing so as a means to leverage a monopoly in one market to monopolize another, or to defuse a threat to an existing monopoly (and, a fortiori, to serve both purposes simultaneously) is a different story.

> Those people who exclusively used Internet Explorer as "the Internet" would too have used the bundled mail client (I can't remember what it was called back in the day) exclusively too and seen that as "the email"?

The bundled email client wasn't an issue because it didn't exist yet -- Outlook Express wasn't bundled with Windows until Windows 98, which was still pending at the time the lawsuit was drafted (the complaint notes at para. 57: "Beginning in or around June 1998, Microsoft will introduce to the market the latest version of its operating system for Intel-based PCs, Windows 98.")

As I understand it, Microsoft spent a lot of effort integrating their desktop environment with the browser. Blurring the lines between files on the machine and hyperlinks, making the desktop a webpage, etc. The idea was that "using Windows" and "using IE" were designed to be synonymous -- so much so that the popularity of Windows could be used to monopolize the internet. Nobody would even be able to use another browser without crippling their ability to use Windows, and you couldn't use another OS without crippling your ability to use the internet.

At least, that was the fear.

Under anti-trust law, it is illegal for a company to use their monopoly power in one business as a competitive edge in another separate market. Therefore, the entire argument turns on two things: (1) was Windows a monopoly (2) was IE an integral part of Windows (hence not a product competing in another market).

Since Windows, for all intents and purposes, was a monopoly, a huge part of the discussion was about whether IE was an integral part of the operating system and therefore MS actions were not to be considered as competition in a different market but simply innovation in the existing OS market. Applications like Calculator had been part of operating systems for well over a decade by then and no-one was shipping any OS without one so there really was not a basis for comparison. Browsers, however, were completely new. Netscape, the first widely adopted consumer grade browser, seemed to be doing very well as a stand-alone application. From today's point of view, you could argue that no operating system would ever ship without a browser but that was not clear at the time. A similar idea would be, in say 20 years, BitCoin lockers becoming an integral part of operating systems. Not obvious today, at least not to anyone outside the niche community of enthusiasts.

Clearly, the courts got it wrong, in that browsers are part of the OS today (just as much as the Calculator). However, MS' ability to put Netscape out of business really did slow innovation down overall for a really long time. It took a long time for the market to correct itself and have the better products win. In fact, it took another monopolist (Google) trying to protect its business in another market to come in and accelerate the pace of innovation.

TL,DR: Courts shouldn't be in the position of deciding technology vision. Not because they do not have noble intentions but because they do not have the expertise.

But courts should be in the position of deciding whether specific actions constitute abuse of monopoly.
Sure, that is the role of the courts. However, in this case, that distinction rested on the definition of what constituted a certain peice of technology and the ability to predict how that technology was going to evolve. The courts had no way of knowing the answer to this question. To their credit, I honestly doubt the companies themselves even knew the answer at that time.

This situation was very different than when an oil company uses their monopolistic position in refineries to choke competing exploration and retail companies. Or when a manufacturer of telephone equipment uses the profits from its monopoly in that business to drive out competitors in its telephone service business. Clearly, in these scenarios, it was clear (and remains clear) that the monopolists were using resources from one business where they were clearly dominant to out-compete other companies in separate (even if adjacent) businesses.

Technology is a very tricky area for the law largely because it is fast changing. Situations are cropping up far faster than the laws can be modified to address them and so the courts are pretty much forced to sand-off the edges of existing statutes that were designed for other situations and hope for the best.

You seem to be saying that the evolution of technology over the next 20 years had something to say about whether Microsoft was abusing a monopoly situation at the time of the trial. I don't think that's a valid way of applying the law.

Furthermore, a big part of Microsoft's motivation for making the browser "part of the OS" was exactly to try to blur that line for the court. I am therefore highly skeptical of using "it's now part of the OS" as a reason that the court got it wrong.

Well, I don't know about the basis for a different "legal treatment", but I know that today the browser is the effective plaftorm/container/"VM" for the most popular applications. The browser capabilities and features will actually affect users and web applications - besides the ability to use this as leverage for other products, a single player with browser monopoly will have an unhealthy preponderance in the shaping of the web. I don't believe we can make an intellectually honest comparison with the calculator or the file manager...
Microsoft have come under fire over other software they've bundled free with Windows as well.

IIRC the latest was Windows Defender.

Adding to that, there's an argument to be made that the browser ballot helped make the Web a much richer place.

The reason I say that is that it allowed Chrome to make serious inroads and become a much more dominant browser. Chrome comes from Google, a company whose entire business model revolves around getting people to spend more time on the Web. So they have a vested interest in pushing the Web toward being as rich a content platform as possible so that people spend more time in their Web browser and not in other applications.

So they did that. They made Javascript more performant, they pushed new standards for producing more attractive and responsive pages, etc. And makers of other browsers, who might not have done the same because they're not driven by similar incentives, had to start hurrying to keep up.