It dawned on me today that one of the most popular web sites on earth is improperly designed at the most fundamental level. Yup, Facebook is conceptually all wrong and in this post, I’ll explain why.
In order to justify the bold claim above, I need to review a quick bit of history. Let’s set the way back machine to the late 80s, in the days before the internet became the ubiquitous mother-of-all-networks it is today. Back in those days, computer networks were viewed as monolithic products: network software came with the network infrastructure (protocols, etc.) AND the applications, all bundled together. The most popular product of this generation was Novell Netware, which at one time claimed 90% of the market share for computer networking software. Anyone used Netware lately? What made a product that owned 90% of the market virtually disappear in the blink of an eye?
The reason Netware failed is because a much better infrastructure (the internet) emerged, which was:
- more open because it used standards based protocols, whereas Netware required the use of Novell’s proprietary protocols
- more ubiquitous because it could by used by anyone, anywhere, whereas Netware was targeted to business users
- more flexible because it worked with lots of applications, whereas Netware limited the set of available applications
- and most important of all: the internet was owned by no one, so no one party got to control it
The internet gradually chipped away at Netware’s dominance and in the mid-90s it really blew up with widespread use of the most successful app of all time: the World Wide Web. Some companies, like Cisco, Sun and Microsoft anticipated and exploited this sea change. Others, like Novell, desperately clung to their proprietary and monolithic vision of computer networking.
Why didn’t Novell adapt? Why didn’t they modify their software to embrace the internet protocols? My guess is that Novell rightly understood that the lion’s share of their investment and value was embedded in their network infrastructure and protocols. Unfortunately for them, the most strategic piece of their portfolio was precisely that which was replaced by the internet.
Ok, fast forward back to 2010. Facebook’s user base is growing exponentially. Even your grandmother is posting stuff on your wall. But take a closer look at what Facebook really is: it’s a monolith, just like Netware. It’s part infrastructure (the so-called “social graph”) and part application. And just like Novell in the 80s, Facebook wants to own both pieces.
In another parallel with Novell, Facebook’s real value is tied to their infrastructure, the social graph. Facebook, the application, is a nightmare. Yet despite the interface annoyances, dictatorial policy changes and security headaches, we all keep coming back like moths to a flame. The reason we keep returning is that we need access to that social graph. We need to communicate with our friends and loved ones. As long as Facebook owns that infrastructure, we’ll keep using their crummy app. If you want to access that social data, there are no alternatives; there is no competition.
Now let’s imagine a different world: one with a clean separation between social networking applications and infrastructure. Imagine a non-profit organization (sort of like the group that oversees internet engineering) that maintains the social graph. Let’s call this entity socialgraph.org (a made-up name, although I suspect that domain name is already taken). Imagine this group provides a repository of social information along with secure programmable interfaces (APIs) to access that data. Imagine complete and user-friendly control over who, when, where, how and with whom your social data is shared. Imagine policies that don’t change without your consent.
I contend that this organization would do a much better job of managing and protecting our data than Facebook does, for one simple reason:
- This organization would care only about the social graph. It would not care about applications, user interfaces, acquisitions, revenue, ads, subscriber base or any of the other business concerns that drive Facebook’s policy makers. The top (and really, the only) priority of this organization would be to maintain and protect your, my and our social data.
This new world creates an ecosystem where social applications can flourish. Don’t like that Facebook UI? Try this new one I just read about. Don’t like having to maintain separate data for Facebook and Twitter? No need to do that any more – those and all other social networking apps feed off of a common infrastructure. Got a great idea for a new social networking app? You can make your idea a reality by leveraging the common infrastructure.
Who wins in this world? We do – we get more services and more innovation and we get to choose the apps we want, with the services we like best, at the lowest price we can find, all tied into the social data we care about.
Who loses in this world? Facebook. Did you read about (or experience) the recent push to integrate facebook into every website in the free world? This is Facebook’s way of saying “we know we’re not going to build a lasting, winning business model with our app – our real value is tied to our social graph so we’re going to try to embed that infrastructure everywhere”. In essence, Facebooks wants to establish their social graph as the de facto standard.
In the late 80s, the roots of the internet tsunami were already well underway – Novell was doomed and didn’t know it yet. But at the current time there is no obvious contender to replace Facebook’s social graph. It may already be too late. But wouldn’t it be great if some enterprising person or team created socialgraph.org and everyone joined it? How awesome would it be to “take back our own data”? At the end of the day, this is our data, it represents who we are and who we care about, and it should be owned and managed by an entity whose top priority is keeping it safe, secure, reliable and easy to manage. I don’t know about you but my recent experiences with Facebook suggest that’s not their top priority.