Stop a passerby and ask them “what is the internet?” Their answer is guaranteed to focus on technology. We find it hard to separate this global machine, which uses around 5% of the planet’s electricity, from its underlying purpose.
Just look at the most used sites on the internet — search engines at like Google and Bing. They use (literally) millions of powerful computers powered by arcane algorithms to constantly scan, filter, sort and rank everything published online.
From this perspective the net seems to be a technological arms race — whoever has the most servers, the best software, the smartest engineers — they will be the winners.
But a new breed of services is emerging that challenges that paradigm. Instead of relying on brute computational force, these systems concentrate on finding the least costly path to an answer. They do this not by crunching data but by connecting human beings.
A great example is Aardvark, a “social search engine” founded by some ex-Google eggheads.
Aardvark doesn’t index the internet, it indexes its users. You ask Aardvark a question and it finds the best person to answer that question.
Google has spent untold fortunes on its “natural language engine” — a system that understands and interprets our questions regardless of how we phrase them. This is how it can return such similar results for questions like “how to fry an egg properly” and “best way of frying an egg”.
But the founders of Aardvark have access to much better natural language engines — human beings. So instead of spending energy on teaching their computers the difference between, say, “therapists” and “the rapists” they can spend it on matching the right people with the right questions.
Humans instinctively trust other human beings over any kind of mass medium. You’d rather ask your friends for a good plumber than pick one at random from the Yellow Pages. The main reason we use services like the Yellow Pages is because our own physical network of friends has information gaps.
What a system like Aardvark does is exponentially widen our network. They’re not the first service to do this — social networks like Facebook and Twitter have already accomplished the feat on a global scale. But the genius of Aardvark is it’s knack of efficiently matching strangers to strangers based only on a shared subject of interest.
Aardvark is the evolution of an older idea called crowdsourcing which involves asking a large group of people to help solve a problem. One of the criticisms of the idea is that the majority of the work is not done by the crowd but by a few suitably skilled members. Aardvark simply cuts out all the noise and goes straight for the informational jugular.
One of the service’s apparent flaws is its reliance on altruism. Why on earth would you spend time answering some random stranger’s question for free? Then again, that’s exactly what many of us would have said about Wikipedia ten years ago. Intellectual satisfaction has proved to be enough of an incentive — Superfreakonomics be damned.
So should Google be worried? Well, apparently they already are. They acquired Aardvark in February for a cool $50-million. Not a bad payday for two years of work.
But there are plenty of other new services that capitalise on this same idea of organising humans around data (rather than the other way around). Hunch, for instance, is a social “recommendation engine” which constantly learns and adapts based on what its users recommend and what recommendations they follow. And there are dozens more projects incubating around the globe.
What’s funny is that all these services are finally getting back to what the internet promised us in the first place — to connect the whole of humanity with our shared knowledge. After all, it isn’t a network of machines; it’s a network of people.