The binary nature of the web means your accidental, happy discoveries are anything but.
Search and software giant Google has, almost imperceptibly, integrated itself into the fabric of our daily lives. We use it to find and track news and information, to connect with friends, to increase our productivity, to locate ourselves in both the physical and the virtual worlds. And with the data it collects on us through each of its services, it has an extraordinary ambition: the tracker of our activities wants to become the catalyst for innovation and director of our dreams.
With all the ephemeral and seemingly disconnected data that it holds on us, the company hopes to “one day tell people things they may want to know as they are walking down the street, without having to type in any search queries”, reports Scott Morrison in the Wall Street Journal. “Think of it as a serendipity engine,” said Google’s Eric Schmidt at the TechCrunch Disrupt conference last September.
“Serendipity” is the latest holy grail in the Silicon Valley software zeitgeist: an ill-defined buzzword that developers use to describe services that will connect people with online ephemera they would not normally find on their own. Yet a website’s success relies on delivering successes, and something that tries to predict serendipity will fail almost every time. “If you can plan it, how is it serendipitous?” asks reader ShockJockey on the Guardian‘s Technology blog. Indeed.
Serendipity is a description applied to an innovation post-hoc, after the process of bumping into something unexpectedly turns into a new idea, a medical discovery, a new recipe, a new relationship or a new artistic movement.
In other contexts, with different frames of reference, these innovations might be attributed to divine intervention. Here, in our pragmatic, delivery-oriented social world, Google’s theoretical “serendipity engine” – and the technology behind websites such as Amazon, StumbleUpon, or Last.fm – aims to serve up a result that will appear novel and unexpected, but is in fact tightly algorithmically defined, based on a user’s profile and derived using the data collected on him or her.
Computer-mediated “serendipity” is therefore constructed by carefully controlling the connections people make between pieces of information and one another in order to maximise delight. This keeps the consumer brand-loyal and the company’s shares high. It does not take people in totally new directions; rather, it filters out the stuff that Computer Says is Irrelevant.
In real life, serendipity is a multifaceted, shape-shifting social and human process that is impossible to reduce into immovable computer code. Sure, a delightful and inspiring discovery may result from sticking a string of words into a search box, but what makes this outcome serendipitous depends on what you do with the information spat out by the engine. You have to recognise it, know what to do with it, and make the essential lateral connections necessary to transform it into something meaningful. You have to analyse and synthesise it, consider your own experience and the knowledge of your personal and extended networks. All Google – and any other site that purports to offer “serendipity” – really wants is to build a predictive algorithm that filters out the happy accidents that they don’t know about, and which might turn into something more.
In his book, Where Good Ideas Come From, Steven Johnson proposes several ingredients for serendipity. They are almost all impossible to replicate on a machine: sleep, coffee, a creative walk, a hot bath, a broken treadmill, chaos and an open system are among them. Yet Johnson still describes the web as “the greatest serendipity engine in the history of culture”, because it allows us to traverse Wikipedia – from an article about fountain pens to one about monarch butterflies via links to chocolate-chip cookies and the chancellor’s real name. That is not, I would argue, serendipitously discovering something about insects; it’s discovering something about insects via a path laid out by the people who edit the articles. It’s not making conclusions, drawing parallels or creating something new.
Bookmark this page for “Digital serendipity” and check back regularly as these articles update on a very frequent basis. The view is set to “news”. Try clicking on “video” and “2” for more articles.