Decentralization has been the hope, but reality paints another picture
Life imitates art: It’s hard to change how things are done
Fans of the TV series Silicon Valley know that the core plot last season was the company’s attempt to build a “decentralized internet”, free from the control of governments, large corporations, et al. Spoiler alert: Our intrepid coder/entrepreneurs have only mixed success by the time the season is finished. In March, Wired took a look at how much of that fictional future was in our present and came away with similarly mixed results: There are some distributed apps (dApps) out there, but they’re harder to use, not especially popular, and have issues around performance and copyright.
Still, with trust in tech companies running low, the interest in dApps remains high. This is especially true inside the crypto and blockchain communities, where the belief is that the nascent technologies can help realize the vision (see this post for more). The very idea of bitcoin centers on decentralized money, free of central banks, governments, and their whims.
KJ Erickson took a deep look at the prospects for decentralizing other parts of our technology/commerce value chain. She examined the risk faced by four of the internet’s giants: Facebook, Twitter, Apple, and Amazon . Erickson looked at how different economic factors help determine which of them might face a decentralized alternative in the future by measuring how the services are paid for; how hard they are to build/replicate; what new value a new decentralized service might bring; et al.
The conclusions are interesting and also sobering. Given recent events, one might think the highest desire among users would be for a decentralized Facebook alternative: some kind of social network outside the grasp of Mark Zuckerberg and the Menlo Park wizards . But Erickson finds that unlikely, noting that Facebook users are invested in using it to store photos and info about events, making switching away annoying if not difficult. She also points out the platform might err on the side of censorship after the Cambridge Analytica mess. Still, of eight possible factors only two seem to place Facebook at risk. (But keep an eye on Mastodon: It might be the next Ello or Path, but early traction is intriguing.)
On the other hand Amazon is seen as high risk for decentralization, with vulnerabilities seen across the board. Whether or not one agrees with her basic conclusion, it’s easy to argue the opposite. Amazon has invested billions in physical infrastructure that makes it better than anyone at shipping billions of items to millions of users. That’s not the kind of thing that someone is going to replicate easily even if they can build a better website (and, little doubt, the latter is achievable).
Indeed, Todd Hoff over at High Scalability argues in favor of centralization generally. He’s mostly talking about technology in a software/hardware/cloud sense while arguing we’ve moved mostly from the days of decentralized protocols (think FTP, TCP/IP, DNS) to centralized software (e.g. Google, Salesforce, and yes, Facebook). Hoff laundry lists things to like about centralization:
“Centralization is not only possible now, it’s economical, it’s practical, it’s controllable, it’s governable, it’s economies of scalable, it’s reliable….” (there’s much more in his post)
And that list explains a lot of our current world: two mobile OSes, giant SaaS companies, a dominant e-commerce site, the social network. All of this is true even while new kinds of fundraising (ICOs), data tracking (blockchain), and media of exchange (tokens) are all realized inventions. Richard Hendricks and his distributed internet are still mostly trying to get their foot in the door.
