Can An Algorithm Discover The Key To Laughter?
STEVE INSKEEP, HOST:
Along with decent finances, it takes a lot of talent and practice to play in an orchestra. The same goes for being able to make people laugh. And even some of the most brilliant comedians can have a hard time of it. Let's listen to the usually great Johnny Carson in one of his not-so-great moments on the "Tonight Show."
(SOUNDBITE OF TV SHOW, "THE TONIGHT SHOW")
JOHNNY CARSON: You know, kids, I tell you, are very fickle. Today at Zody's, I saw a nine-year-old step on a Smurf's head to get to a Cabbage Patch Doll. I mean, how quickly it changes. How quickly you change. How quickly we all change.
INSKEEP: Carson's sidekick, Ed McMahon, had to laugh at that, but nobody else was required. Not knowing how a joke will go over is part of the exhilaration of being a comic. But imagine if comedians could get it right every time. Some Web-based startups are trying to figure out the algorithm for funny. Alex Schmidt reports.
ALEX SCHMIDT, BYLINE: Peter McGraw, director of the Humor Research Lab at the University of Colorado Boulder, is not fooling around.
PETER MCGRAW: This is not a place that has shelves and shelves of rubber chickens and whoopee cushions.
SCHMIDT: Actually, the lab is one of the first mainstream psychological attempts to study people's reactions to humor and how they can be used in the real world - in public service announcements, for example. McGraw says humor is notoriously tough to study.
MCGRAW: It's pretty easy to make people sad. But when it comes to humor, what one person finds funny, another person is offended, and yet another person is bored by it. And so to conduct this research really broadly ends up being difficult.
SCHMIDT: McGraw thinks figuring out the key to laughter could be used in the business world. He's advising a new startup called Laffster.
DANIEL ALTMANN: Log onto the app. It's showing you how to vote here.
SCHMIDT: I'm going to click through the directions. I'm testing the most recent Laffster mobile app with CEO Daniel Altmann. Kind of like Pandora recommends music or Netflix recommends movies, Laffster recommends comedy. The first video I click in the app is Joe Biden's appearance on the TV show "Parks and Rec."
(SOUNDBITE OF TV SHOW, "PARKS AND RECREATION")
VICE PRESIDENT JOE BIDEN: And you must be Leslie Knope. Welcome. Welcome.
(Soundbite of Laughter)
AMY POEHLER: (as Leslie Knope) My name just came out of your mouth.
SCHMIDT: I thought the clip was pretty funny, so I gave it a thumbs-up. Videos are defined by attributes, like sarcastic, demeaning, subtle or dirty. By noticing what I like, Laffster tries to recommend more movies, keep me laughing and keep me watching. Daniel Altmann wants to license out the technology, too.
ALTMANN: Can someone else plug in the Laffster technology and drive three times the amount of videos, three times the amount of ads watched and ultimately drive revenue? Because they're coming there, and they're interacting. They're engaging.
SCHMIDT: Other startups are trying to harness humor differently. Julia Kamin is founder of dating website Make Each Other Laugh, currently in beta. If two people on her site laugh at the same stuff, her software will send them on a date. There will be no joke categories or analysis at all. In fact, she doesn't want to understand the magic of why two people click through humor.
JULIA KAMIN: One of the things that gives us excitement in life is that there are things that are always going to be elusive, no matter how much we're able to crunch data.
SCHMIDT: Kamin quoted author E.B. White, who famously said that analyzing humor is like dissecting a frog: The frog dies, and who cares about its insides? Humor researcher Peter McGraw draws a different conclusion.
MCGRAW: I like to say that if frogs are like jokes, there's a lot of sick frogs out there.
SCHMIDT: And, McGraw says, if by dissecting frogs we can improve humor, then by all means, pass the scalpel. For NPR News, I'm Alex Schmidt.
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.