1. Break up Facebook under anti-trust law. Social network share is as dangerous as, if not more dangerous than, market share.
2. Pass an American GDPR. Consumers get an absolute right to audit and delete their data. Explicit consent is required for each instance of third-party sharing. Companies are liable to their users for breaches, with a minimum amount claimable through an easily-accessibly regulator.
What would breaking up Facebook look like? Most features of Facebook couldn't operate as standalone businesses without creating their own massive ad networks.
Crack it along its social graph(s) or separate the platform (which would become a regulated utility) from the advertising business.
Mergers offer four broad categories of synergies [1]: demand-side, supply-side, political and administrative. Airline mergers are usually about demand-side synergies. Having more market share gives you better data about future trends. It also gives you pricing power. Media/telecom mergers tend to rely on supply-side synergies: prices won't be raised, but content suppliers can be squeezed for distribution. When Lockheed Martin bought Sikorsky it acquired lots of employees in Connecticut [2]; that bought them some bulldogs in the Congress. (Administrative synergies are mostly B.S.; they fall into the "you have an accounting department, we have an accounting department, combined we can have just one" category and were mostly disproven in the post-WWII conglomerate boom.)
Facebook is a super aggregator [3]; it has both demand-side and supply-side monopolistic synergies. Cracking the demand side means users go to sleep with a Facebook account and wake up with a Facebook No. 3 account. (Or to sleep with an Instagram account and wake up with an Instagram account.) This worked for AT&T and could be done today (at significant expense to Facebook's shareholders as it was to Standard Oil's, AT&T's and Microsoft's). Cleaving social graphs is complicated. But we understand graph theory better today than we did in AT&T's days. Bonus: require an open inter-operability standard if these networks want to interconnect.
Cracking the supply side would involve taking related advertisers and hiving them off into a separate advertising company. The Facebook platform left would be a natural monopoly, regulated like a utility, serving dual profit and public roles. (Messenger, Instagram, M, and other relatively-discrete products would similarly be spun off.)
TL; DR The novelty of breaking up a Facebook is superficial and finds precedent in antitrust enforcement history.
Sure, but neither of those have anything to do with what this thread is talking about. Data ownership and privacy are important, but this thread is about GP's claim that (paraphrased) the fundamental problem is that advertising is the ability to buy influence at scale, and that "bad actors" aren't prevented from doing so.
I'm still not so sure it's true though. Is it materially different to go buy from 5 ad brokers instead of 1? Or from ad exchanges that automate it for your? It's hard for me to see how this would have much of an effect, beyond mildly increasing the cost.
> 1. Break up Facebook under anti-trust law. Social network share is as dangerous as, if not more dangerous than, market share.
You should better ask why the mass of people registered on it to let this happen. It is not that privacy activists have not been warning all the time...
To give a little background: In Germany (where there is a lot more concern for privacy) Facebook never got that deeply ingrained into the daily habits of most people. It is also not uncommon that at a social gathering people explicitly tell you that they don't desire that any photos that were taken here to be put on Facebook.
> You should better ask why the mass of people registered on it to let this happen.
Social networks gravitate towards oligopolies(+), because people go wherever everybody else is. Even in Germany, there is a social price to pay for not having a Facebook account.
(+) Oligopolies, because there seems to be a age generation effect: children don't use the network where their parents are.
This is one of the areas where self-regulation simply doesn't work.
> Even in Germany, there is a social price to pay for not having a Facebook account.
I often write here at HN that there is no "objective" status in society, but rather status in various, often vaguely defined groups in the society, where the individual status symbols can even contradict each other.
It is not my impression that all in all I had to pay a social price for not having a Facebook account at any time (quite the opposite). Rather in some groups this lead to a social malus, but in other groups it gave a large social bonus.
I fully agree that this highly depends on the groups to which you belong. Personally, I have not experienced a malus for not having a Facebook account, but I know of occurrences (e.g., a student learning group organizing their communication via Facebook). For myself, I have faced a potential social price for not having a WhatsApp account that was large enough to let me create an account.
Well, 1 could also be taken care of by Congress passing a new law. But I think the more interesting alternative is Zuck deciding to do it on his own. There's no reason he couldn't create a nonprofit and give it the "natural monopoly" pieces. That would certainly include identity and the social graph, and maybe a publishing interchange.
Then Facebook could step back and compete on the non-monopoly pieces. The client, of course. But all the add-ons. Groups and events and marketplace and whatever else. Google, Twitter, Amazon, and probably Apple would probably quickly create their own social clients. And would also join some nonprofit advisory board, which would help to set protocols and standards.
1. Break up Facebook under anti-trust law. Social network share is as dangerous as, if not more dangerous than, market share.
2. Pass an American GDPR. Consumers get an absolute right to audit and delete their data. Explicit consent is required for each instance of third-party sharing. Companies are liable to their users for breaches, with a minimum amount claimable through an easily-accessibly regulator.