We live in our own separate worlds: Your Google search results are based on your surfing history; Netflix recommends TV programs for you based on what you've already watched; the mobile app Zite, which began as a side project for grad students at the University of British Columbia, offers a magazine-like experience that's compiled based on your interests and reading history. We asked Mike Klaas, Zite's 31-year-old co-founder and chief technology officer, about the perils and promise of personalization.
Zite is one of a few popular apps that aggregate content– there's also Pulse and Flipboard. What's your big draw?
For Zite, I think the main value proposition is finding content for you that you wouldn't be able to find in any other way. It's not just a better way to read your RSS feeds or your Twitter streams–or, if you have a favourite site, reading it on your mobile tablet. It can provide that, but the real value is finding you that article on the blog you never heard of that is exactly what you wanted today.
Some people think that's becoming a problem: More and more, we're only getting content that is tailored to our stated interests and reading history, and we're trapped in so-called filter bubbles that prevent us from stumbling across articles or information that could contradict our world view. Does Zite reinforce those filter bubbles, or break them down?
I actually think recommendation technology is a way out of the filter bubble problem. Contrary to the premise, I don't think technology actually causes bubbles. I think humans are great at creating their own bubbles. They self-select who they're following on Twitter, who they follow on Facebook, they tune out neutral media and increasingly select for themselves tighter and tighter sets of sources, and only hear the same content from the same types of people. And a serendipity engine is exactly what you need to break out of that.
I know you were acquired by CNN last year, so maybe this isn't a concern, but–is there a business model here?
I think we have a bit of a unique opportunity. Recommending a good ad for someone isn't that different than recommending a good article. It's context-dependent, it needs to be relevant to their interests and what they're interested in at the moment. There's a reason we haven't done it yet. Not only is it difficult technologically, it also requires a large ecosystem: You need a large pool of ads to pull from if you're going to try to do some really deep personalization. In the meantime, we've focused on a sponsorship model.
Is the increasing number of newspaper and magazine paywalls a problem for Zite and its users?
I'm not concerned about that at all. One of the largest values of Zite is that we can provide content that isn't just The New York Times. We go to the independent blogs, we can go to the medium-tiered sites, where there's still a ton of really great content.
After CNN acquired Zite, the company moved to San Francisco. Your CEO has talked about how, after Microsoft acquired his previous company, most of the original staff left. There must have been some apprehension. How different are things because of the change?
Technology acquisitions happen all the time, and it's not really controversial to say they don't usually work out well. CNN did something really smart that I don't see happening in many other acquisitions. They realized they're not a super-technology-focused company. They wanted to bring some innovative technology, Silicon Valley atmosphere and culture into their everyday activities. But they knew if they tried to integrate us, mesh us, they'd completely destroy us, right? They understand that part of the value of a start-up is its agility, its ability to take risks, ability to move fast in a way that big companies can't. And so they kind of set up firewalls between CNN and us to make sure we're not unduly hampered by them, while still providing a way to slowly integrate more and more over time.
On Twitter, you describe yourself as a coureur de bois. Care to explain?
That's just a personal joke. Coureurs de bois were the early French-Canadians who canoed around and traded furs. I guess I feel a philosophical attachment to them.
They were travellers, like you–you've now relocated to San Francisco. But your Twitter bio says you're based in Vancouver.
Yeah, that's not accurate any more, I'm afraid. Thanks for reminding me to change that.