The Zingales and Rolnik approach to social network reform

It is sufficient to reassign to each customer the ownership of all the digital connections that she creates — what is known as a “social graph.” If we owned our own social graph, we could sign into a Facebook competitor — call it MyBook — and, through that network, instantly reroute all our Facebook friends’ messages to MyBook, as we reroute a phone call.

If I can reach my Facebook friends through a different social network and vice versa, I am more likely to try new social networks. Knowing they can attract existing Facebook customers, new social networks will emerge, restoring the benefit of competition.

Today Facebook provides developers with application-program interfaces that give them access to its customers’ social graph, Facebook Connect and Graph A.P.I. Facebook controls these gates, retaining the right to cut off any developer who poses a competitive threat. Anticipating this outcome, very few developers invest seriously in creating alternatives, eliminating even the threat of competition.

By guaranteeing access to new customers’ data and contacts, a Social Graph Portability Act would reduce the network externality dimension of the existing digital platforms and ensure the benefits of competition.

Here is the full NYT piece.  Is it feasible that the data could be transferred in a ready-to-use form?  And can the contacts object that they did not themselves consent to a transfer of their associated information say to “Alt-RightBook”?


I think it would be reasonable to handle objections from contacts in the same way one would handle it with phone numbers. If you don't like the fact that someone transferred their phone number, you don't answer or block their number. In the same way if you don't like one of your FB friends contacting you from Alt-RightBook, you just unfriend them on FB.

This is technically possible, but infeasible for both political and techno-political reasons (nailing down a specification). I think we do need better social media for a better society, better politics, and better government. But it is hard.

I which I agree with Donald Trump Jr.

Facebook and Twitter will make incremental changes. Perhaps they will make some useful discoveries. Or we might have to wait for AIs that are good enough to grade on both truth and sanity.

wtf is this

Antitrust for the 21st century. The proposal is for a "a Social Graph Portability Act would reduce the network externality dimension " to reduce monopoly in social media services. We asked our bureaucrat boffins to legislate flying cars and we get this...

BTW is this choice architecture and behavioral economics? Is an econ student going to prototype a better system? Or is a VC going to think that "better" (in this sense) than Facebook can beat Facebook?

Maybe the economics is all about engagement, and not necessarily truth, justice, and the American way.

Cute how TC believes in such a thing as Alt-Right.

Equally cute is how Richard Spencer does too.

Cute that Rich thinks he is custodian of the language.

Hey, still better than the 'Dark Enlightenment' that similar types (think the ever so modestly named persona of Vox Day) love to bandy about.

Nonetheless, Richard Spencer most definitely believes in the alt right, though his “Hail Trump, hail our people, hail victory!” is anything but cute.

'Is it feasible that the data could be transferred in a ready-to-use form?'

Of course it is.

'And can the contacts object that they did not themselves consent to a transfer of their associated information say to “Alt-RightBook”?'

Not in the U.S., at least legally. Whether something along the lines of norobots.txt or do not track could be implemented is another question.

There are already many services that work like this. There is Gnu Social (, Diaspora ( and probably some others. Also the "social media" platforms that people were using before Facebook and other giants entered the field used to be more community oriented (i.e. not in the hands of one big centralized actor) like IRC (

There are two worlds. One is where you normal people live. In this world everyone uses Facebook, Google, or whatever centralized proprietary platform is most marketed or has the flashiest user interface. The other world is for us, the free software and DIY tech enthusiasts. In our world we have communal, open source solutions to all things technical and we prefer to use those whenever we can. I'm always very surprised and frustrated when people speak about Facebook, Microsoft, etc. "big evil" tech monopolists in a way that assumes there are no alternatives to them. There are, but you just don't care about them.

You have to admit, those are not the most attractive landing pages.


Must second this, it does make sense that we all have websites, like our hosts do.

Mastodon seems to be a solution for decentralized servers, however, right now my impression is that there are turf wars, and the parties are the free speech gang, also known as nazis and/or fascists, and the SJWs, also known as "where is my safe space?".

I've recently joined Mastodon, which is in principle like Twitter, but less centrally controlled.

If any other MR readers use Mastodon, I'd be very grateful for your suggestions of accounts to follow.

People should own their own social network graphs. Facebook does operate on non-portable ownership of people's graphs. That is obvious.

I don't favor a political government solution, I favor technical solutions. People like better access and control. Someone can build a technical system that enables more user portability and control and choice and that will win.

Is it feasible that the data could be transferred in a ready-to-use form?

Yes it is feasible. It's called federation, and the protocols for doing so have existed for years. The reason it's not done is that dominant social networking platforms have a vested interest in not making federation available so that networking effects provide a barrier to the success of potential competitors.

It's the same reason none of the various instant messaging clients work with each other. Not because they couldn't, but their developers don't want them to.

I'm not so sure this is worth worrying about. Facebook has only been around for 13 years, and it has only been a public company for 5 years. In cyberspace time, that's an eternity, but because cyberspace time moves so quickly; it's dangerous to start regulating it after just a few years of monopoly power. Facebook hasn't completely locked in its position. There was a time when Internet Explorer, Windows, and iPods completely dominated user experiences too. Somebody will eventually come out with a distributed social networking service that's sufficiently well developed that people will use it. There are plenty of apps which don't have the dominance of Facebook, but still have a big enough user base that they could threaten the company if they ever chose to. Yes, losing favor with Facebook would be risky, but also a potentially winning strategy in the long run.

@David Condon - Well that's very sensible, unfortunately the EU does not agree with you; during the late 1990s they wanted to tax data sent on the internet, and their fetish with privacy (CCTV cameras in Europe, even filming a public street, can only store 2 weeks of data before it must be erased, I kid you not) has already Balkanized the internet.

Soon it will be like antitrust with physical products: endless government regulations designed to ensure a 'level playing field' not unlike the browser & bundling wars of the late 1990s in the USA, remember those? "Sensible" does not equal 'politically feasible' in the real world.

Funny that you mention Internet Explorer and Windows, the combination which precipitated the antitrust case against Microsoft.

Monopoly is the point. Competition is good for consumers, but bad for businesses.

Only in textbooks. In the real world, you need a producer surplus and monopolies to have money for research and development. Yes, you read that right: real innovation is often done by Fortune 100 companies, not mom and pop producers in a perfect competition (read: near zero profits, MR = Price = straight line) cut-throat world. Exceptions: when there's so much money to be made from disruptive technologies that Fortune 100 employees can form their own monopoly, as in the stylized history of the transistor, the exception that proves the rule.

Bring back Standard Oil!

Of course the data could be transferred in a portable, ready to use form. Digitally signed email contains all the necessary functionality for "publish authenticated message," so pick any of the multiple standards for maintaining a contact list with associated public keys, and run with it.

The missing piece isn't the social graph per se, it's willingness of the monopoly network to transmit messages to and deliver messages from external networks.

There are tricky issues with governance that the NYT article doesn't touch on at all.

In the old days (physical letters and email), when you send someone a letter, it belongs to the recipient. There's no way to take it back. Similarly, in each of our address books we have (possibly outdated) contact information about the people we correspond with. So, we own information that was created by other people. If the author asked you to destroy a letter they sent you, you could refuse.

Social networks work differently. When you publish something, the recipients don't get any control, unless they save a copy. You can edit or delete something you published. This has advantages: our contact lists are more up to date and we can fix embarrassing mistakes. But it also means we don't normally save copies of any correspondence we receive and thus have no control over it - any of it could be modified or disappear. The popularity of Snapchat partially due to its auto-erasing photos feature suggests that control by author (rather than by recipient) is what many users want.

The older model made distribution and federation easy, which is why there are multiple email systems in practice. It's a rather libertarian way of doing things; every user can keep anything they read. (Though there are customs about not sharing private correspondence, and making copies might not be legal according to copyright law.) Messages cross borders when you publish them. On the downside, federation makes spam and abuse much easier, because it happens on the borders between systems, which are everywhere.

Modern social networks are more like separate nations. Each social network sets its own rules and they apply to all users. A federated system consisting of multiple social networks would need to explain to users that there are borders and that some users are "foreign" and follow different rules. Users would need to learn what to expect when their messages cross borders, and the networks would need to figure out jurisdiction when resolving conflicts between users. It wouldn't be as easy to understand or use, and it's not clear that many users would want it.

A common and more natural way to handle this is syndication. People can be members of multiple social networks, and you can post the same content to multiple places if you like. Presumably when you became a member of each network, you learned its rules and can adjust accordingly.

This was good.

This is exactly the issue Urbit is trying to solve. Mencius Moldbug aka Curtis Yarvin made some money with a mobile patent when he was young, then read a lot of old books and wrote a lot of good political theory, and is now trying to engineer a technical solution to what are essentially political problems. The folks at Urbit think it is complicated enough that you have to rewrite a lot of basic computing from scratch.

Comments for this post are closed