How to build the anti-Facebook
1. Détournement
Social networks come and go, but only one of them has captured a billion humans. Given the events from 2016 through today, it’s pretty clear that so-called social media has in fact been destructive to the fabric of society.
But because of network effects, it’s very hard to believe that another general-purpose social network will be able to compete with Facebook unless Facebook makes systematic terrible mistakes over many years or is broken up by governments.
The way to defeat Facebook is not to build another social network — it’s to build something completely different, the anti-Facebook, while still something that offers comparable familial, social, professional and informational results to human beings as a group.
I have a fairly clear picture of how to do this. I call it Samizdat, abbreviated to S, in honor of tens of thousands of Soviet dissidents in the past.
I don’t think it would be a huge amount of work to implement — perhaps a few thousand hours of programming from a small number of talented programmers to get something useful and reliable.
Whether people would use it is of course at least partly subject to the vagaries of fate, but it might finally allow us to regain some sort of control over our data destinies.
In this series of Medium articles I’m going to explain how this might work, trying to avoid technical buzzwords as much as possible.
Here goes!
Keep the good parts
Let’s start with the five Facebook features that must be retained.
You need to be able to keep in touch with your friends, family and colleagues in both active and passive ways — active, where you reach out to others, and passive, where you just see what’s happening without saying anything. Call this feature people.
S has to be free of charge.
You need to be able to block individuals or groups in S.
S has to be easy to use. It can’t be much more difficult to teach to someone who isn’t very good at computers than Facebook was when it first came out.
And S has to have a network effect.
A social system is worthless without people. It’s intolerable without blocking intolerable people. Maybe paying a bit of cash wouldn’t be so bad, but few would do it, so it has to be free. If it isn’t easy, most people won’t use it. And the network effect is a bit creepy, but if the value of the system does not increase as more people join it, you’ll die like Friendster.
The anti-Facebook
As history shows, if you wish to conquer a huge empire, it’s best to start your own from as far away as possible. So I’m going to deliberately say no to some key Facebook features and see what pops out.
No crowd requirement
A classic social network only works at large scale. Facebook would be useless with a single person.
Invert it — so S has to be perfectly useful for one single person on their own!, then get more useful if a second person joins, because of the network effect, and so on.
No servers
Facebook relies on very roughly a million central servers pumping out content and ads.
Invert that— S has no servers of its own.
So how does information get around? Peer-to-peer is a thing but granny might not want to do that and it might be impossible on your mobile device. But what about piggybacking on existing services?
I’ll let that percolate...
No real name requirement
Facebook requires you to use your real name, and enforces this fairly aggressively. One human may only have a single account. Identity is your Facebook name is your real world name is the name on your passport.
Invert this. S must allow perfect anonymity with continuity of identity. One human might have multiple identities in S; conversely, a name might belong to multiple humans.
No proprietary software
Facebook is written in some bastardized version of the debased language PHP for the front end and probably a dozen other languages for data analysis and process control — all tightly secret to Facebook.
Turn this around. Everything is open source. (You saw this coming.)
No new protocols
Protocol means “how you exchange data between machines”.
Companies have their own. Google has their beloved but byzantine Protocol Buffers. Facebook no doubt has their own style of “Face Buffers”.
Invert it. S has to use existing protocols — open source of course.
This fits snugly with that earlier, percolating idea of piggybacking on existing servers.
No continuous delivery
Facebook continuously delivers information — it lets you browse on an infinite page forever. This is addictive and hooks into mindless, automatic behavior. We have all been there.
Flip it over. S should use separate, discrete updates. S should be useful if you get one update a year, or one a second. S should encourage a civilized workflow where you interact with it daily and not many times an hour.
No ads
Summary of part 1
Samizdat or S, the anti-Facebook, should allow communication with people, and blocking them; it should be free, and easy, and have a positive network effect.
S should require no crowd nor real name, no server or proprietary software or protocols. Updates should happen discretely, not continuously and no ads.
Part 2 to appear soon — quite soon if there is any interest.
