Team Warren —
There’s a huge preexisting body of work on these matters of “principal-agent problems” (aka “agency problems”). I have no doubt you’ve consulted a lot of that prior art in drafting your proposal, but there are a few unforced errors in your piece that those artists have specifically warned against. While vertical foreclosure is a problem that needs a surgically precise regulatory solution, the biggest concerns of those who are dedicated to analyzing these dynamics are the (mis)incentives and externalities of less surgical or broader enforcement. (Horizontal foreclosure is a good example of the latter.)
Please take care to get this right, because, given errant implementation, the opportunity cost of progress foregone — gross and net — is much bigger than the gross incremental gains. The following addresses all of those issues, with case studies for Alphabet/Google, Amazon, Apple, Facebook, Netflix, etc:
The Chinese Wall of Worry: App Store Impunity and Vertical Foreclosure
The principal-agent problem, solution, and externalities
Adventures in Consumer Technology
In addition, here are a few examples that directly address some of the specific prescriptions you mentioned…
These companies would be prohibited from owning both the platform utility and any participants on that platform.
Be careful how you implement that so as not to ultimately inflict more harm upon consumers: “The Chinese Wall solution would provide [someone like] Apple the discretion to use 3rd party partnerships like this when it needs to leverage a depth of integration that could otherwise threaten end-user security. And, Google Maps proves that the consumer experience didn’t suffer from such a partnership. Not one iota.”
Platform utilities would be required to meet a standard of fair, reasonable, and nondiscriminatory dealing with users.
It’s impossible to find the bright line between right and wrong here. Just juxtapose Tim Pool vs Kara Swisher’s agitations for change regarding these standards — from the perspective of right vs left, respectively. Those standards (“fair”/“reasonable”/“nondescriminatory”) are political constructs for 99% of cases. You must start with the 1% cross-section, implementing structural solutions (e.g. prohibit data collection or inference of a user’s race), then address the rest ex post.
Platform utilities would not be allowed to transfer or share data with third parties.
It’s impossible to implement this without having an asymmetric downside impact on consumers, internet properties (big and small), and SMBs — for all of whom the rise of data and targeted advertising have leveled the playing field. Unless you have an innovative plan to have your cake and eat it too, it’s important to be up-front about the tradeoffs you’re making to achieve your regulatory proposal’s desired end state.
Part-and-parcel, a data sharing prohibition further cements these giants’ competitive advantage, creating a bigger problem than the one it’s intended to prevent. And the opposite, a data-portability mandate, is no more sound: “Would you rather have one well-resourced/centralized entity controlling everyone’s data or poorly-resourced/decentralized individuals controlling their own? As with most things in life, you’d probably want some balance between the two extremes.”
One last parting word from “The Content Police”:
…notice the juxtaposition: The subjective/manual/human/editorial era has been succeeded by the objective/automatic/algorithm/curation era. These are extremes on the spectrum of information dissemination approaches. Perhaps the social web will settle into a happier medium. Maybe it won’t. I guess my point is that solutions need be careful so as not to encourage wading further into one extreme or another. Regardless, as the printing press and the industrial revolution showed us, it’s still Day 1 of this modern renaissance.
Now, there’s a movement underway that’s fracturing the once stateless, borderless, open web, including…
The US is fighting to standby protections like Section 230 of the Communications Decency Act, which provides digital platforms with safe harbor that shields them from liabilities resulting from user generated content they host. While this amplifies free speech and open information, it makes censorship a prohibitively cumbersome legal process. (e.g. Given the velocity of digital content, the damage of digital distribution is already done by the time analog courts hand-down verdicts.)
The EU is moving to push the burden of censorship back onto the platforms themselves with both Article 13 (a copyright filtering mandate) and Article 11 (a proposed “link tax”). While this amplifies censorship, it not only makes free speech and open information economically cumbersome, but also gives more power to unelected corporations’ to govern free speech — with all the harm that comes from false positives — which neither the left or the right of the political spectrum want.
China is tightening its centralized control over the web, with the heavy-handed government increasingly enlisting private businesses to further the ubiquity of “The Great Firewall”. This is pure censorship, leaving no room for free speech and open information.
Please take the opportunity here to move these systems back to the middle — not deeper into the algorithmic era or all the way back to the editorial, but rather somewhere in between that maintains the growth trajectories of free speech, business, and creativity while chipping-away at the negative byproducts.
You want to be an agent of progress, not regress. Tim O'Reilly once pointed out that the Luddites’ bear case never played-out because tech’s augmentation of manufacturing produced cheaper goods which stimulated more unit consumption that required more service jobs. Please, please, please, let’s focus on accelerating the flywheel toward achieving those new jobs to meet the future’s new demand — as opposed to winding-back-the-clock to an obsolete era.