This article is from the source 'guardian' and was first published or seen on . It last changed over 40 days ago and won't be checked again for changes.
You can find the current article at its original source at http://www.theguardian.com/technology/2014/mar/10/upworthy-viral-news-sxsw-eli-pariser
The article has changed 3 times. There is an RSS feed of changes available.
| Version 0 | Version 1 |
|---|---|
| Upworthy’s Founder Talked At SXSW… And You’ll Never Guess What He Said | Upworthy’s Founder Talked At SXSW… And You’ll Never Guess What He Said |
| (about 3 hours later) | |
| Upworthy is one of the fastest growing news websites in the world right now, fuelled by its stream of stories designed to be shared on social networks. | Upworthy is one of the fastest growing news websites in the world right now, fuelled by its stream of stories designed to be shared on social networks. |
| It has plenty of critics, but the rapid rise of the site – famous for its capitalised headlines – cannot be denied. So what does Upworthy’s success mean for the wider news industry, and what can traditional media companies learn from it? | |
| CEO Eli Pariser appeared at the SXSW conference today to give his perspective, interviewed by New York Times journalist David Carr. Pariser started Upworthy two years ago, and it now attracts 50m-60m visitors a month. | CEO Eli Pariser appeared at the SXSW conference today to give his perspective, interviewed by New York Times journalist David Carr. Pariser started Upworthy two years ago, and it now attracts 50m-60m visitors a month. |
| Carr asked him to define an algorithm. “Basically just a piece of code that sorts through information and outputs a response. What we’re talking about is content filtering algorithms,” he said. | Carr asked him to define an algorithm. “Basically just a piece of code that sorts through information and outputs a response. What we’re talking about is content filtering algorithms,” he said. |
| “It’s an editorial opinion encoded into an equation that powers things like the Facebook news feed or the Twitter discovery window... trying to figure out what’s good: what people want to consume. These little pieces of code are more powerful now than a lot of the most powerful editors in media.” | “It’s an editorial opinion encoded into an equation that powers things like the Facebook news feed or the Twitter discovery window... trying to figure out what’s good: what people want to consume. These little pieces of code are more powerful now than a lot of the most powerful editors in media.” |
| Pariser talked about his research into stories about Afghanistan: they don’t get clicked or read much, yet they’re on the front page of sites like the New York Times every day. | Pariser talked about his research into stories about Afghanistan: they don’t get clicked or read much, yet they’re on the front page of sites like the New York Times every day. |
| “They say this is the important stuff we all need to know as citizens, as we’re at war there,” he said, suggesting that Upworthy was set up after he started worrying about how these kinds of stories would find an audience in the social networking universe. | “They say this is the important stuff we all need to know as citizens, as we’re at war there,” he said, suggesting that Upworthy was set up after he started worrying about how these kinds of stories would find an audience in the social networking universe. |
| “We’re moving from this world in which everyone types in newspaper homepages. Increasingly people are logging onto Twitter and Facebook, and they’re expecting an algorithm to surface the most important stuff to them,” he said. | “We’re moving from this world in which everyone types in newspaper homepages. Increasingly people are logging onto Twitter and Facebook, and they’re expecting an algorithm to surface the most important stuff to them,” he said. |
| Upworthy’s key metrics have shifted away from measuring unique visitors and page views to something called “attention minutes” according to Pariser. | Upworthy’s key metrics have shifted away from measuring unique visitors and page views to something called “attention minutes” according to Pariser. |
| “It’s trying to look at our content the way an algorithm would see it. Are people sticking with it? Do we know that people are still there, or are people walking out of the room?” he said. “For a long time that’s been something that’s bad to do... an affront to your editorial authority.” | “It’s trying to look at our content the way an algorithm would see it. Are people sticking with it? Do we know that people are still there, or are people walking out of the room?” he said. “For a long time that’s been something that’s bad to do... an affront to your editorial authority.” |
| Pariser suggested that if a journalist spends weeks writing an investigative piece, that gets buried in the newspaper or website and that nobody reads, that doesn’t accomplish the “civic mission” that it was intended to fulfil. | Pariser suggested that if a journalist spends weeks writing an investigative piece, that gets buried in the newspaper or website and that nobody reads, that doesn’t accomplish the “civic mission” that it was intended to fulfil. |
| “We think a lot about three metrics: importance, satisfaction and quality. Quality I think of as ‘the subjective traits of something that’s really a fine piece of journalistic craft’. A high-quality New Yorker piece versus a high-quality Time piece... this ineffable little piece of the soul of the publication that comes out when you have quality,” he said. | “We think a lot about three metrics: importance, satisfaction and quality. Quality I think of as ‘the subjective traits of something that’s really a fine piece of journalistic craft’. A high-quality New Yorker piece versus a high-quality Time piece... this ineffable little piece of the soul of the publication that comes out when you have quality,” he said. |
| “Quality is the subjective thing, and then the satisfaction, the engagement, that’s the objective signals that people are actually enjoying this. Are they sharing it, are they staying with it, are they clicking it? And importance is the civic importance or the social importance of the piece of content.” | “Quality is the subjective thing, and then the satisfaction, the engagement, that’s the objective signals that people are actually enjoying this. Are they sharing it, are they staying with it, are they clicking it? And importance is the civic importance or the social importance of the piece of content.” |
| He suggested that because important, high-quality but not necessarily engaging stufg has “had a leg up for so long” it hasn’t had to compete on being compelling. | He suggested that because important, high-quality but not necessarily engaging stufg has “had a leg up for so long” it hasn’t had to compete on being compelling. |
| “It gets on the front page anyway. Now we have this real challenge where the story about Afghanistan or new policing methods or whatever has to compete in the same pool with Kim Kardashian or Candy Crush or everything else that’s begging for attention from the algorithms,” he said. | “It gets on the front page anyway. Now we have this real challenge where the story about Afghanistan or new policing methods or whatever has to compete in the same pool with Kim Kardashian or Candy Crush or everything else that’s begging for attention from the algorithms,” he said. |
| “Unless we figure out how to make the important stuff really engaging, I don’t know that it reaches a broad audience.” | “Unless we figure out how to make the important stuff really engaging, I don’t know that it reaches a broad audience.” |
| Pariser said we’re moving into a period with a power curve where “news junkies have never had it better... but for most people who don’t seek out content about important stuff, and expect to just have it surfaced in their media environment may be having that happen less. And to me that’s worrisome.” | Pariser said we’re moving into a period with a power curve where “news junkies have never had it better... but for most people who don’t seek out content about important stuff, and expect to just have it surfaced in their media environment may be having that happen less. And to me that’s worrisome.” |
| Carr suggested that Upworthy is “mission-driven” – fundamentally progressive in its politics, even if it doesn’t always like to admit it, or put a specific label on it. | Carr suggested that Upworthy is “mission-driven” – fundamentally progressive in its politics, even if it doesn’t always like to admit it, or put a specific label on it. |
| “We don’t pretend that we don’t have a point of view: being ambivalent about gay marriage or climate change. We don’t put a label on it politically, because a lot of stuff we do isn’t just ‘progressive’ stuff,” said Pariser. “It’s not something that fits into a political box. We don’t want to be a site just for people who self-identify as having a specific ideology.” | “We don’t pretend that we don’t have a point of view: being ambivalent about gay marriage or climate change. We don’t put a label on it politically, because a lot of stuff we do isn’t just ‘progressive’ stuff,” said Pariser. “It’s not something that fits into a political box. We don’t want to be a site just for people who self-identify as having a specific ideology.” |
| Is there a danger that as Upworthy grew so fast, did the traffic become more important than the original mission, wondered Carr, while also suggesting that Upworthy may be creating its own filter bubble for people with a certain view of the world. | Is there a danger that as Upworthy grew so fast, did the traffic become more important than the original mission, wondered Carr, while also suggesting that Upworthy may be creating its own filter bubble for people with a certain view of the world. |
| “If you’re reaching 50, 60, 80 million people, you’re outside any one niche. It does actually get outside of those enclaves, and you start to have a conversation across a pretty big chunk of the culture about this stuff,” he said. | “If you’re reaching 50, 60, 80 million people, you’re outside any one niche. It does actually get outside of those enclaves, and you start to have a conversation across a pretty big chunk of the culture about this stuff,” he said. |
| Pariser talked about one of his first big viral stories at the MoveOn website: a video of a young man talking about his two gay mothers, explaining passionately why they shouldn’t be discriminated against. | Pariser talked about one of his first big viral stories at the MoveOn website: a video of a young man talking about his two gay mothers, explaining passionately why they shouldn’t be discriminated against. |
| It had been watched 200,000 times or so, but once its headline was changed to “Two Lesbians Raised A Baby And This Is What They Got” from its original “Zach Wahls Speaks About Family” the video took off – it’s since been watched more than 18.4m times. | It had been watched 200,000 times or so, but once its headline was changed to “Two Lesbians Raised A Baby And This Is What They Got” from its original “Zach Wahls Speaks About Family” the video took off – it’s since been watched more than 18.4m times. |
| “With Google SafeSearch off, we own the number one search for ‘two lesbians’,” said Pariser. “I hope you screensaved it, because it’s your life’s work!” joked Carr, drawing laughter from the audience. But Pariser said that the rebadged headline helped the video spark a popular debate about gay marriage. | “With Google SafeSearch off, we own the number one search for ‘two lesbians’,” said Pariser. “I hope you screensaved it, because it’s your life’s work!” joked Carr, drawing laughter from the audience. But Pariser said that the rebadged headline helped the video spark a popular debate about gay marriage. |
| “The important thing is to get people to really great content that they love. The headlines are only as good as they accomplish that goal,” he said. “We don’t do well unless people love it so much that they share it.” | “The important thing is to get people to really great content that they love. The headlines are only as good as they accomplish that goal,” he said. “We don’t do well unless people love it so much that they share it.” |
| Pariser played a video about a young African-American man talking about his experience of getting harassed and frisked by police, seemingly for the crime of having an empty bookbag – complete with an audio recording of the incident. He suggested the video played an important role in helping people understand that experience. | Pariser played a video about a young African-American man talking about his experience of getting harassed and frisked by police, seemingly for the crime of having an empty bookbag – complete with an audio recording of the incident. He suggested the video played an important role in helping people understand that experience. |
| “Seeing that experience is enraging and helps me care about something I only knew about in the abstract,” he said. “This has been seen by literally millions of people. That is what we’re here to do: find pieces of content that have something to say, like this one does, and get it in front of people.” | “Seeing that experience is enraging and helps me care about something I only knew about in the abstract,” he said. “This has been seen by literally millions of people. That is what we’re here to do: find pieces of content that have something to say, like this one does, and get it in front of people.” |
| Is there a danger that Upworthy has ended up over-exercising its headline muscle and delivering clickbait, where the videos don’t meet up to the expectations engendered by the headline? Pariser disagreed, suggesting that the metric of social sharing is a judgement on whether content is clickbait or not. | Is there a danger that Upworthy has ended up over-exercising its headline muscle and delivering clickbait, where the videos don’t meet up to the expectations engendered by the headline? Pariser disagreed, suggesting that the metric of social sharing is a judgement on whether content is clickbait or not. |
| “Facebook is looking at clicks, but it’s also looking at whether people share something. And sharing a piece of content is a significant validation of that thing: you’re putting it out to all of your friends, and saying this is something you should spend your time on,” he said. | “Facebook is looking at clicks, but it’s also looking at whether people share something. And sharing a piece of content is a significant validation of that thing: you’re putting it out to all of your friends, and saying this is something you should spend your time on,” he said. |
| “They’re looking at a whole bunch of signals. Like is a softer signal probably ,share is a pretty strong signal. And if I were them I’d also be looking at what happens in between: are people clicking into something and then coming right back out. Or are they paying attention and then sharing it? That looks from an algorithmic point of view pretty good.” | “They’re looking at a whole bunch of signals. Like is a softer signal probably ,share is a pretty strong signal. And if I were them I’d also be looking at what happens in between: are people clicking into something and then coming right back out. Or are they paying attention and then sharing it? That looks from an algorithmic point of view pretty good.” |
| Pariser said his mission is to prove that the news industry shouldn’t give up on trying to reach the majority of the population with important stories, just because they would traditionally have been presented in a dry, unpalatable way. | Pariser said his mission is to prove that the news industry shouldn’t give up on trying to reach the majority of the population with important stories, just because they would traditionally have been presented in a dry, unpalatable way. |
| “We’re seeing there is a big appetite for this stuff, if people are told it will be fun and interesting and compelling. It doesn’t have to be choking down yesterday’s Brussels Sprouts,” he said, citing a video about the US healthcare system as a prime example. | “We’re seeing there is a big appetite for this stuff, if people are told it will be fun and interesting and compelling. It doesn’t have to be choking down yesterday’s Brussels Sprouts,” he said, citing a video about the US healthcare system as a prime example. |
| “We had to pick out a little piece of it and say ‘wow, this is really amazing’, and we found that when people came in, they stuck through that video. It got 5m people watching a video about the US healthcare system,” he said. “To me, there’s no purity in doing the beautiful thing that doesn’t reach anyone.” | “We had to pick out a little piece of it and say ‘wow, this is really amazing’, and we found that when people came in, they stuck through that video. It got 5m people watching a video about the US healthcare system,” he said. “To me, there’s no purity in doing the beautiful thing that doesn’t reach anyone.” |
| Is Upworthy simply gamifying the news, asked Carr. “The news needs to be a little more interesting, a little more engaging. It does!” said Pariser, who argued that often it’s a case of initial expectations: if people expect to find a video or story more interesting, that’s exactly what will happen. | Is Upworthy simply gamifying the news, asked Carr. “The news needs to be a little more interesting, a little more engaging. It does!” said Pariser, who argued that often it’s a case of initial expectations: if people expect to find a video or story more interesting, that’s exactly what will happen. |
| Pariser compared Facebook and Twitter’s importance for news. “The average American spends about two minutes pr month on Twitter, and about seven hours per month on Facebook. It’s not even the same order of magnitude. One in every six internet minutes is spent on Facebook,” he said. | Pariser compared Facebook and Twitter’s importance for news. “The average American spends about two minutes pr month on Twitter, and about seven hours per month on Facebook. It’s not even the same order of magnitude. One in every six internet minutes is spent on Facebook,” he said. |
| “There’s this funny thing where people talk about social media – ‘Oh, I do social media’ – and there are these little ecosystems like Instagram or Twitter. But there’s really one that looms over all the rest, and that’s Facebook right now.” | “There’s this funny thing where people talk about social media – ‘Oh, I do social media’ – and there are these little ecosystems like Instagram or Twitter. But there’s really one that looms over all the rest, and that’s Facebook right now.” |
| Carr asked about Upworthy suffering when Facebook changes its news feed algorithm: is it too dependent on that big social network for its traffic, which can get squeezed at a moment’s notice? For example its slump between November and December 2013. | Carr asked about Upworthy suffering when Facebook changes its news feed algorithm: is it too dependent on that big social network for its traffic, which can get squeezed at a moment’s notice? For example its slump between November and December 2013. |
| “We had a really great month in November: 80 million. Other than that month traffic has been more or less hanging out at 50 or 60 million,” said Pariser. “There’s a tendency because it’s a black box and because the Facebook gods are very powerful gods indeed, to reverse-engineer a very powerful algorithm in very simple ways.” | “We had a really great month in November: 80 million. Other than that month traffic has been more or less hanging out at 50 or 60 million,” said Pariser. “There’s a tendency because it’s a black box and because the Facebook gods are very powerful gods indeed, to reverse-engineer a very powerful algorithm in very simple ways.” |
| His point: Facebook didn’t dial up Upworthy’s position in its algorithm in November then dial it down in December. There’s no knob to, in Carr’s words, “fuck Upworthy: I think it’s called the Upworthy fuckolator!” | His point: Facebook didn’t dial up Upworthy’s position in its algorithm in November then dial it down in December. There’s no knob to, in Carr’s words, “fuck Upworthy: I think it’s called the Upworthy fuckolator!” |
| “If I were Facebook what I would be doing was trying to set a playing field for what constitutes quality content. By that token we’re doing very well: we’re one of the top websites on Facebook right now,” said Pariser, before dealing with a question about the growing number of critics of Upworthy. | “If I were Facebook what I would be doing was trying to set a playing field for what constitutes quality content. By that token we’re doing very well: we’re one of the top websites on Facebook right now,” said Pariser, before dealing with a question about the growing number of critics of Upworthy. |
| “Did we prefer the 18 month honeymoon or the three-month vicious backlash? The honeymoon was better! It’s not fun, because we are doing something we believe in, and when folks we respect want to take it apart, it doesn’t feel good,” he said. | “Did we prefer the 18 month honeymoon or the three-month vicious backlash? The honeymoon was better! It’s not fun, because we are doing something we believe in, and when folks we respect want to take it apart, it doesn’t feel good,” he said. |
| “The challenge is always to listen thoughtfully to the critique that’s smart. And on the other hand with the stuff that’s dumb, just be happy that we’re reaching enough people that it matters.” | “The challenge is always to listen thoughtfully to the critique that’s smart. And on the other hand with the stuff that’s dumb, just be happy that we’re reaching enough people that it matters.” |
| Upworthy has been researching its community: the subjects they’re most interested in and enjoying on the site. The three top topics are climate change and energy, income inequality and poverty, and human rights, which Pariser said is a big source of pride for his team. | Upworthy has been researching its community: the subjects they’re most interested in and enjoying on the site. The three top topics are climate change and energy, income inequality and poverty, and human rights, which Pariser said is a big source of pride for his team. |
| Now the company is partnering with organisations in those categories – Climate Nexus, ProPublica and Human Rights Watch respectively – to double down on these topics. A marriage of organisations who know what’s going on in these areas, and a company like Upworthy that can get their work to a large audience. | Now the company is partnering with organisations in those categories – Climate Nexus, ProPublica and Human Rights Watch respectively – to double down on these topics. A marriage of organisations who know what’s going on in these areas, and a company like Upworthy that can get their work to a large audience. |
| Pariser was asked about trust, and said that his team fact-check every piece of content independently, to ensure it doesn’t – for example – publish a video that’s fake. | Pariser was asked about trust, and said that his team fact-check every piece of content independently, to ensure it doesn’t – for example – publish a video that’s fake. |
| It has also tried to make corrections “as viral as the original: we feel that is good behaviour online, to own your mistakes as quickly as possible, as big as possible”. In contrast to the traditional newspaper tactic of burying corrections as small items further back in the paper. | It has also tried to make corrections “as viral as the original: we feel that is good behaviour online, to own your mistakes as quickly as possible, as big as possible”. In contrast to the traditional newspaper tactic of burying corrections as small items further back in the paper. |
| Pariser was asked about sites like Viralnova, which emerged after Upworthy and adopted its template. Are they a problem for the company, given that there’s not a huge barrier to entry in this space? “If you consider the business to be ‘viral anything’ then yeah, a lot of people are going to be good at that in the future,’ he said. | Pariser was asked about sites like Viralnova, which emerged after Upworthy and adopted its template. Are they a problem for the company, given that there’s not a huge barrier to entry in this space? “If you consider the business to be ‘viral anything’ then yeah, a lot of people are going to be good at that in the future,’ he said. |
| “If we look 10 years in the future, there’ll be dozens of organisations that wil lbe good at that. or us, this is why we drew a line in the beginning: we’re never going to do stuff that is fun and entertaining, but not substantive in some way.” | “If we look 10 years in the future, there’ll be dozens of organisations that wil lbe good at that. or us, this is why we drew a line in the beginning: we’re never going to do stuff that is fun and entertaining, but not substantive in some way.” |
| “The reason we did that is there’s a longstanding model of drawing people in on celebrity news or cat pictures or whatever it is, and you hope they make their way across to the political stuff. But what that often leads to is 90% cat pictures and 10% really substantive civically useful stuff. We want to build the muscle of making that stuff.” | “The reason we did that is there’s a longstanding model of drawing people in on celebrity news or cat pictures or whatever it is, and you hope they make their way across to the political stuff. But what that often leads to is 90% cat pictures and 10% really substantive civically useful stuff. We want to build the muscle of making that stuff.” |