Attracting Competitors
By the 1990s, Microsoft had achieved monopoly in the operating system market for personal computers. Microsoft was making big profits. Microsoft was publicly traded. The whole world knew how lucrative its operating software business had become. So, one would expect that Microsoft would naturally attract challengers.
After all, a fundamental idea behind capitalism is that if one dominant firm takes full advantage of its monopoly, it will be quickly disciplined by the market. New challengers will spot an opportunity to steal away unhappy customers, and healthy industrial relationships will be restored. As Justice Scalia once put it, “The opportunity to charge monopoly prices–at least for a short period–is what attracts “business acumen” in the first place; it induces risk taking that produces innovation and economic growth.”
How an Applications Ecosystem Can Protect Incumbents
But this didn’t happen for Microsoft in the 1990s. A serious challenger in operating systems wasn’t emerging. This was because, as the DOJ explained, “Microsoft's monopoly power is protected, and has been protected for years, by high barriers to entry into the operating systems market, the most important of which is the applications barrier. The applications barrier to entry exists because applications written to Windows will not run on other operating systems and other operating systems cannot effectively compete against Microsoft unless they can offer PC users a wide array of applications similar, in depth and breadth, to the vast set of applications that exists for Windows.”
Importantly, no software application developer would want to undertake the costly and laborious process of recreating an app for an operating system where there are going to be no sales to customers. Because its not in the interests of any one developer to write an app for New OS, none of the other developers want to do it either. Because there are no applications on New OS, no customers would want to buy it. The applications network effect is self-reinforcing. Everyone wants to write code for the platform where people already are, which makes that platform stronger over time.
Netscape + Middleware
But technological shifts can erode the applications barrier to entry. This certainly happened in the 1990s with the rise of internet browsers. All of a sudden, middleware, or applications that could live on the web became real. Suddenly, Microsoft sprung into action.
As the DOJ explained, “Microsoft has engaged in a broad pattern of unlawful conduct with the purpose and effect of thwarting emerging threats to its powerful and well-entrenched operating system monopoly. Most prominent among these was the threat posed by competing Internet browsers, particularly Netscape's Navigator. Non-Microsoft browsers, if widely used, promised to form the center of an emerging middleware platform that could have helped to erode the high applications barrier to entry that protects Microsoft's monopoly.”
In other words, what the DOJ argument boiled down to was that web-based applications, like what we now know as Google Docs, were starting to become possible with the rise of internet browsers. The DOJ alleged that Microsoft realized that if something like Google Docs could exist on the web, then the lock-in for Windows OS could be eroded over time.
The erosion would happen in steps. People who used Word might come to realize that they can get similar functionality on Google Docs. Then users might realize that they can get access to Google Docs on a different operating system. After all, Apple computers could access the web too. So could Linux and others. The Windows OS monopoly could entirely erode!
Bill Gates’ Deposition from the Microsoft Monopolization Case
So, Microsoft underwent a series of actions to quash Netscape and middleware to protect its operating system monopoly. In the end, the DOJ challenged Microsoft and won in court. Microsoft was ordered to stop, and Apple, Google, and others emerged as massive companies from that quasi-breakup.
Mobile OS Duopoly
All of that brings us to today. The applications barrier to entry again protects a newer type of OS: mobile OS. Namely, iOS and Android.
For example, Windows Phone tried to go head to head with those two, but developers gave up on Windows’ Mobile OS. As one reporter put it, “One of the key problems for Microsoft was the dearth of apps for its mobile platform. Belfiore said Microsoft had tried hard to attract developers, even to the point of building apps for them, but that there too few Windows mobile users for it to be viable.”
The applications barrier to entry in mobile OS has resulted in a situation where Apple and Google both charge developers extreme transaction fees of 30 percent, and these developers don’t have anywhere else to turn because the costs of developing for a new platform from scratch simply aren’t worth it.
But is that changing? Could it?
The Rise of Generative AI + the Erosion of the Applications Barrier to Entry
In theory, at least, if generative AI tools can understand and produce code. Then they can be used to cheaply reproduce app functionality for an entirely new OS, at a much lower cost.
Importantly, many apps already store user data and can port that to a new ecosystem. So for example, if it was incredibly cheap to develop the Twitter app for New OS, well then users could sign in and all their tweets would already exist on the app because those tweets are stored on Twitter’s servers. The Twitter feed would be indistinguishable on New OS versus iOS.
Imagine that at scale. App developers could use a generative AI tool to scan a code base in one OS and recreate that functionality for another OS. Maybe Microsoft could resurrect the Windows Phone. Maybe an entirely new mobile OS provider could emerge. The 30 percent VIG that the mobileOS duopoly charges would necessarily come down. Instead, app developers would thrive!
Unique Moment for Entry
It’s not clear that any publicly announced Generative AI tools can clone app functionality yet. But, if I was a VC, this is certainly a bet I would want to make. There are a few different tools that likely need to be created here. For example, one is a Generative AI tool that can clone apps for a new OS at virtually no cost. Another is a Generative AI tool that can integrate those new apps with the backend data infrastructure. Further, would be a Generative AI tool that can bring down the cost of building out an OS itself.
I imagine the tools currently being built can likely do a lot of these functions, even if they’re not explicitly being built for that. I can also imagine that the cost of computing to do these functions might still be too expensive to really allow for the applications barrier to erode too much. Still, as the price of computing continues to decline, surely this will become affordable at some point?
The lower the cost of porting an application onto a new OS, the more viable a new OS becomes. If the costs go down to zero or even net negative (assuming VC subsidies to developers), then an entire application ecosystem can emerge anew.
Creative Destruction and Redistribution
As CNBC estimates, “If all developers paid a 30% cut to Apple, Apple’s App Store grossed more than $85 billion in 2022…If Apple’s commissions were all 15%, the App Store’s estimated gross would come in lower, around $70 billion.”
In a world where the applications barrier to entry evaporates, how will end-users choose which OS to use? Likely quality competition will restart. OS competition will become more about internally developing functionality to differentiate yourself. In this sort of world, we would be more likely to see Apple create more of its own unique apps at high cost to stand out. We would be more likely to see Google do the same thing. Maybe more hardware innovation would happen too. This would be pretty good. I’m tired of the same marginal hardware changes with each new iPhone.
Importantly, more quality competition wouldn’t really hinder anyone else’s ability to also innovate on hardware or on internal software or OS functionality. We’d see an explosive race for new features. We’d see vigorous competition on prices to continue to attract developers. There would be a massive redistribution of income away from the platform monopolies that Google and iOS have and into the hands of developers like Epic Games and users of those apps.
Ultimately, this could be a very good thing for app developers and users. For Google and Apple, maybe not so much.