Copyright ©2010 NPR. For personal, noncommercial use only. See Terms of Use. For other uses, prior permission required.

MARY LOUISE KELLY, host:

It's ALL THINGS CONSIDERED from NPR News. I'm Mary Louise Kelly.

ROBERT SIEGEL, host:

I'm Robert Siegel and it's time now for All Tech Considered.

(Soundbite of music)

SIEGEL: We're checking in with technology entrepreneurs and thinkers on ALL THINGS CONSIDERED. Today, Hunch.com's Caterina Fake. The phrase user-generated content gets thrown around a lot and we can thank Caterina Fake, among others, for that. Fake co-founded Flickr, the photo-sharing site that Yahoo bought for $35 million five years ago.

The idea that any amateur or professional photographer can upload photos for little to no charge and post comments about other user's photos may seem old hat now, but Flickr was part of the social media revolution. And Caterina Fake joins us now from KQED in San Francisco to tell us what she thinks is next for the Web. Welcome to the program.

Ms. CATERINA FAKE (Chief Product Officer, Hunch.com): Thanks, Robert.

SIEGEL: It seems like everyone's doing it now, sharing photos, sharing articles, posting videos. Did you know when you started Flickr that this was the future of the Web?

Ms. FAKE: It was in the air. You could sense in 2003, 2004 when we had started this that there was a sea change happening. Blogging, as well as social networking, were these things that were just sort of nascent. You could see it coming on the horizon and you could see the original impetus of the Web being contribution and participation, that kind of activity.

SIEGEL: Well now you're the chief product officer of Hunch.com. Wired magazine quoted you as saying, and I quote, "the ultimate goal of Hunch is to map every person on the internet to every object on the Net." What does that mean?

Ms. FAKE: The thing that happens in what is now known widely as Web 2.0 was people were participating and people were putting information online and there was a great deal of data. I think the next phase of the Web that comes from Web 2.0 is how does a person take all of this knowledge about themselves and their likes and their dislikes and where they are and all that kind of thing and make the internet more useful to them?

How can you - say you want to book a hotel in Dallas and you've never been to Dallas and nobody that you know has been to Dallas recently. The taste profile that Hunch has made about you would know whether or not you're interested in going to a boutique hotel or you're the kind of person who wants to stay at a chain hotel like a Marriott or a Hilton or something like that.

SIEGEL: We should explain that you have on Hunch.com a taste profile in which people answer various questions in a questionnaire. Questions such as, are you a Mac person or a PC person? Or, Which of these four kinds of lettuce would you be more likely to put on a salad? And the idea is you're trying to get some kind of profile of the individual that would lead you to make assumptions about what other purchases they might make.

Ms. FAKE: Exactly. People answer a great number of these questions; 120 questions is that average number of questions that a Hunch user will answer. And it turns out that a lot of these questions are predictive in ways that you would not expect. So whether or not you wear plain socks or striped socks might be predictive of whether or not you would like the movie "Napoleon Dynamite," for example.

SIEGEL: I must say, it remains a big question for me as to whether one can effectively quantify taste. I'll give you an example. I, and it turned out, a contemporary of mine, we're both baby boomers and both of us quit Netflix at precisely the same time for precisely the same reason, completely independent of each other, which was that after watching and liking "Garden State" a great deal, we got the message, if you like that movie, you'll like "Harold & Kumar Go to White Castle."

And somewhere in the first half hour of this movie I realized that someone had assumed I liked movies about doper teenagers in New Jersey and that was what made it. And you could write the algorithm easily. It just didnt get to the quality or the kind of movie at all. I was watching teenage entertainment.

Ms. FAKE: Netflix only knows what movies you've seen before. By radically expanding the dataset, we are able to make much better predictions for you and wouldnt fall into those traps that you see on Amazon and Netflix where you are recommended a John Grisham novel if you've just bought a John Grisham novel.

SIEGEL: But there's another difference that I want to ask you about. On social media sites like Facebook, people have pages in order to just, you know, to have interactions with other people. And they make of that space what they will. That's different from saying here's a place where we want to help you buy things. I wonder if sites of that sort are nearly so successful as Facebook. I suppose Amazon has been successful.

Ms. FAKE: In many ways I would say they are more successful in building a taste profile because social networking sites are much more performative. You are presenting yourself in a particular way. Your portrayal of yourself is often aspirational.

In Hunch's case, you may be the kind of person who is going to be recommended an Acura. But on a social networking site, you might portray yourself as the kind of person who drives a Lamborghini. And so, in many ways, the sense of who you are and how you portray yourself are often at odds.

SIEGEL: Flickr was a case of understanding the internet as being a space where people would volunteer information. In this case, photographs about themselves and photographs they had taken, and share it with others. And it was more important than finding a way to order a product, whether it was a print or a t-shirt with your picture printed on it. At Hunch.com, you're walking in the opposite direction, it seems. Going from how do I move from the social space idea back into marketing things.

Ms. FAKE: I dont know if that is completely true. There's a large swatch of Hunch that has nothing to do with purchasing products or services.

SIEGEL: So you could fix the Hunch client up with a good writer, with a poet.

Ms. FAKE: In fact, I was responsible myself for building the poetry topic on Hunch. Which poet should I read? And Hunch is itself a user-generated site. All of the poets that have been added to that topic have been contributed by users.

So I started out and I put up Wallace Stevens and Elizabeth Bishop and William Butler Yates and then somebody else put up William Carlos Williams and Marianne Moore and so on.

People are adding content to it and recommendations to it as in the examples that I just gave. But it's also contributory in that people are participating in it by building their own taste profiles for themselves individually.

SIEGEL: Well, Caterina Fake, chief product officer for Hunch.com, thank you very much.

Ms. FAKE: Thank you.

(Soundbite of music)

SIEGEL: And now some questions from the Hunch.com taste profile.

Ms. FAKE: Sure.

SIEGEL: Are you a Mac person or a PC person?

Ms. FAKE: I'm a Mac person.

SIEGEL: Do you generally prefer your sandwiches to be cut vertically or diagonally?

Ms. FAKE: Diagonally. Although, I have to say that what I do do when I cut my sandwiches diagonally is tear off those little corners of the bread and get to the meaty bits in the middle.

SIEGEL: And do you prefer arugula, iceberg, red leaf or romaine lettuce on your salad?

Ms. FAKE: Arugula.

SIEGEL: Arugula.

Ms. FAKE: Sometimes red leaf.

SIEGEL: And if it picked a magazine that you would subscribe to, what magazine would it pick?

Ms. FAKE: The Economist and the New Yorker, both of which I subscribe to.

SIEGEL: So it got you right.

Ms. FAKE: Yes.

SIEGEL: Oh, all right. Well, Caterina Fake, thank you very much for talking with us.

Ms. FAKE: Thank you.

Copyright © 2010 NPR. All rights reserved. No quotes from the materials contained herein may be used in any media without attribution to NPR. This transcript is provided for personal, noncommercial use only, pursuant to our Terms of Use. Any other use requires NPR's prior permission. Visit our permissions page for further information.

NPR transcripts are created on a rush deadline by a contractor for NPR, and accuracy and availability may vary. This text may not be in its final form and may be updated or revised in the future. Please be aware that the authoritative record of NPR's programming is the audio.

Comments

 

Please keep your community civil. All comments must follow the NPR.org Community rules and terms of use, and will be moderated prior to posting. NPR reserves the right to use the comments we receive, in whole or in part, and to use the commenter's name and location, in any medium. See also the Terms of Use, Privacy Policy and Community FAQ.