In 1998, the United States Department of Justice filed an antitrust complaint against Microsoft, accusing the company of abusing their monopoly power. Among the litany of examples, one of the most egregious was Microsoft’s Windows licensing policy whereby computer makers who shipped Windows pre-installed on any of their computers had to pay Microsoft a license fee for every computer they shipped, whether Windows was pre-installed on it or not. The complaint first established that Microsoft had achieved monopoly power in the market for PC operating systems. It then detailed the many abuses of that power Microsoft committed, and sought remedies to curb the behavior.
In watching this unfold during the late ‘90s, most of us had grown accustomed to the cut-throat behavior of Microsoft. But looking back over the case now, one is easily left with the impression, “how did these guys think they were going to get away with this stuff?”
The dreadful behavior of dominant market participants is nothing new. In fact, it seems almost inevitable. Companies with dominant positions, especially in markets exhibiting network effects, often take drastic actions to advantage themselves mostly because they can. They do this until one of two things happen: (a) a government steps in and stops the anti-competitive practices or (b) a disruptive force comes along and enables competitors to usurp the company’s power. In the case of Microsoft, it was both (the U.S. DOJ/European Union cases and mobile computing).
We are at a similar point in the mobile/internet ecosystem. We have a handful of dominant tech platform companies, operating in markets with network effects, wielding enormous market power. One difference this time, many of us thought, was that most of these companies were formed after the Microsoft case and vowed never to abuse market power in the way Microsoft did. Google proclaimed its core motto was “Don’t Be Evil” and, for many years, the company behaved admirably, putting users first and being cautious with user data. But we are now faced with an increasingly untenable situation—overwhelming market dominance by platform companies like Google and Facebook, a rather capricious and sometimes arbitrary wielding of their power and responsibility, and new U.S. government FCC policy delivering unchecked power to ISP gatekeepers, to name just a few ominous concerns.
Many of the new gatekeepers have built their businesses largely by selling their users’ attention to advertisers. Their products are free to use, but in exchange, they gather as much data on user behavior and user preference as possible and allow advertisers to use this data to precisely target prospective customers with ads. While it is easy to observe that the overwhelming majority of users are comfortable with this tradeoff (how else to explain the more than 1.5B users who use Google and Facebook daily?), there is some growing frustration with the end effect of being targeted in this way, and some users are seeking alternative models which maintain privacy and control over their data.
As much as I love (and religiously use) the wonderful products from Google, Facebook, Twitter, and Apple for example, we are witnessing disturbing market behavior from many of them. A few recent examples are:
- Facebook’s complete abdication of responsibility for the misuse of its platform by foreign governments, bots, and untrustworthy information sources
- Twitter’s seeming love of inflating usage figures by refusing to rid the platform of bots and it’s arbitrary application of rules governing speech and user behavior
- Facebook’s outright copying of competitor’s features and its arbitrary banning of ads for crypto products and services
- Apple’s decision to ban templated apps from the AppStore and their decision to ban Google Voice.
- Facebook’s uneven application of content censorship
- Google forcing Amazon to remove YouTube from Echo devices and Google’s questionable practices in Google Shopping search results
- Google’s search results demonstrating bias in India by favoring their own flight search results over competitors
- And of course, the last few weeks revelations about Cambridge Analytica’s nexus with Facebook.
I am not claiming that this behavior breaks laws (although in some cases, authorities found it did). I am simply observing the cost to the ecosystem of having successful companies wield enormous market power. I am the first to point out that their success is a direct result of them building beloved products that often improve our lives. This is just the consequences of success. And it can be quite damaging to the overall ecosystem, suppressing innovation and deterring competition.
The recent decision by the U.S. Federal Communications Commission to overturn net neutrality regulations is another massive cause for concern. In the U.S., most of us have very few choices for broadband. In fact, just under half of the U.S. population (46 million households) is served by only one broadband provider for 25 Mbps or higher speeds, and another 10 million homes have no access at this speed. We effectively have little or no choice for internet access in this country. As a result, our ISP’s are our gatekeepers, and wield enormous power over what we can see and the costs others must pay to reach us. There is ample evidence that, left unchecked, ISPs will abuse this power. Until recently, they were prevented from doing so. Now, once the spotlight shifts away from them, they may choose to exercise this new power in harmful ways and we have effectively no recourse and an unsympathetic government. There must be an alternative here.
Welcome, decentralized networks.
When power is concentrated in the hands of the few, we can predict with reasonably certainty that this power will be exercised in ways not in the interest of the many. In my opinion, the most important implication of the arrival of cryptocurrencies and the blockchains on which they are built is the possibility of building sustainable decentralized computing platforms, services and apps. It may finally be possible to build widely-distributed networks without centralized trust or control, and to allow user consensus to govern their future. And, through cryptotokens, it is now possible to vest the majority of economic value creation in the hands of the network participants, rather than in the hands of the centralized provider. In this scenario, “commodity” applications like messaging, social media and application infrastructure like file storage and compute become very much like public utilities — and they are owned and governed by their participants. For many of us, this is the mission behind crypto.
Over the past twenty years or so, we have seen a significant erosion of trust in our institutions—government, religious, media, financial and corporate. Public trust in government is at its lowest level since Pew began measuring it in 1958. The same is generally true of religious institutions, the Supreme Court, organized labor, and of course, the President himself. The social networks don’t fare much better, and much to my surprise, a majority of Americans now support regulating tech companies. In this era of massively eroded trust, open decentralized networks emerge with an alternative structure. They offer us a way to transact without trusting a centralized party to administer the transaction (or having to trust the counter-party). The timing of introducing distributed transaction technology not requiring the users to “trust” an intermediary couldn’t be more perfect. (In fact, Bitcoin was launched on the heels of the 2008 credit crisis and gained traction partly in response to distrust in the financial system.) The scale of this notion is breath-taking in its possibilities, in so much as every traditional marketplace today (from Uber to Seamless to eBay to stock exchanges to Amazon to Etsy to AdWords) is centrally facilitated. A lot of reinvention is possible.
The open source software model has provided a roadmap here. A natural next step beyond open-sourcing infrastructure software is to effectively open source the data itself. As Joel Monegro and Chris Burniske eloquently state in their Placeholder Thesis Summary:
“Following the history of information technology and the massive trend towards open source, we can see that democratizing information is the natural next step in the incessant trend to open source, and thus the next big opportunity for innovation.”
We have arrived at the moment where the global internet needs ways of working around the dominant platforms. Users need mechanisms to transact without banks, ISPs, financial service companies and other middle men getting in the way. Creators need platforms for self-expression that allow them to participate in the economic value creation their continued usage creates.
How Will Decentralized Systems Compete With The Platforms?
As Adam Ludwin artfully explains, “on almost every dimension, decentralized services are worse than their centralized counterparts.” Just because the participants share in the economic value creation doesn’t make decentralized apps any faster. They will be slower, much harder to scale, and more challenging to govern. You can’t just launch decentralized Facebook and sit back and watch it win. Success will likely result from a combination of offering benefits unavailable in centralized networks (such as true data privacy, or a lack of censorship) and/or a better economic model for participants, and likely in some specialized and possibly narrow use cases at first (such as prediction markets or non-fungible goods marketplaces).
In my view, the most interesting projects are the ones which offer developers a better deal. Many of today’s dominant platforms offered, at some point in their rise to prominence, an app platform with built-in distribution for developers. Facebook offered a developer platform for several years (and Zynga rode that to prominence) before shutting it down leaving developers suffering. Twitter was once developer friendly before they became developer ambivalent, destroying significant goodwill in the dev community it worked so hard to cultivate. Apple’s iOS attracts almost three million developers to build apps for their platform and Android more than six million. Developers, in most cases, make or break the success of platforms. But something funny happens on the way to dominance, and as Alex Felix put it to me, “centralized platforms seem to stunt their own growth at critical junctures by attempting to monopolize ancillary business opportunities rather than plotting a course toward long term universality.” This is likely a result of inevitable investor pressure for better monetization. But crypto communities might have the benefit of building for the very long term and could resist these pressures. As Alex puts it, “crypto companies and their user/owner structures can afford to think generationally.”
The real question to ask is, “on which platforms will the apps of tomorrow be built?” My bet is that quite a few decentralized platforms will offer developers a much better deal—far less control and restrictions, long-term platform stability, and more economic gain—and it will be developers that deliver user benefits atop new protocols. Two years from now, an app like HQTrivia might have launched atop Ethereum or PROPS, avoiding Apple’s 30% distribution tax and taking advantage of a native economic model based on tokens. This will happen provided two things become true: the infrastructure for building and scaling decentralized apps massively improves, and decentralized apps can achieve very wide frictionless distribution. I believe both are likely to be true.
Is The New Boss The Same As The Old Boss?
Even if decentralized systems can prove themselves to work as well as, or at very least provide meaningful new benefits when compared to their centralized counterparts, what is to stop successful decentralized systems from turning into dominant and powerful groups? And if they do, what will stop them from acting badly too? Governing decentralized systems (which are typically composed of self-interested actors) has proved challenging. They are, by definition, more difficult to steer. One hopes that the more widely distributed economic value of these systems can act as a counterbalance to the tendencies of centralized actors to exert power in damaging ways. But this remains to be seen. At very least, as Vitalik Buterin points out,
“It is much harder for participants in decentralized systems to collude to act in ways that benefit them at the expense of other participants, whereas the leaderships of corporations and governments collude in ways that benefit themselves but harm less well-coordinated citizens, customers, employees and the general public all the time.”
The design of many of these systems intends to make it easier for good coordination and to make it harder for harmful coordination. But beyond that, through the social networking era, we have learned a lot about the design of online communities and how to achieve coordinated communities with good intent. It is possible to build communities with norms embedded in them over time, whereby the overwhelming number of community participants act positively and work together to purge the community of bad actors. We have seen this purposefully constructed in Tumblr, Amino and to some extent in Wikipedia. In decentralized networks, the users themselves have more power than they do in centralized systems. If the users build communities with particular norms, it becomes harder for the community to act against those norms.
There are tons of unanswered questions in cryptoland. It’s relatively easy to doubt that decentralized apps will be any better than centralized versions. You could argue my point that “developers will build great things” is a cop out and avoids the crux of the “will decentralized apps actually be competitive with the existing platforms?” question. But some of the trends are easy to see here. There are tens of thousands, maybe even hundreds of thousands of developers focused on decentralized protocols, dapps and crypto networks. There remains massive distrust in our institutions and growing anger at the power of the platforms. And the alternative incentive structure is novel and just plain more fair. We sit at an interesting crossroads.
At Venrock we are enlivened by the possibilities here. We are proud investors in the PROPS Project (decentralized network of video applications) and have formed a strategic relationship with CoinFund (an advisory-focused cryptofund). We remain focused on decentralized protocols, services and apps and foresee a future internet with fewer gatekeepers, where users have more control, and where users participate more fairly in both the governance of these networks and the economic value creation that ensues.
Special thanks to Jake Brukhman, Oleg Golubov and Alex Felix at CoinFund, and Jonathan Glick (my wisest and most thoughtful expert on so many topics) for their helpful guidance, ideas, conversation and pushback on this piece.