Blog Post

Future of Media: Curation, Verification and News as a Process

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

As part of a “social media summit” this week, the BBC posted an overview of how its user-generated content desk handles reports from the field — verifying and curating them in much the same way that Andy Carvin of NPR has been doing for the past few months during the upheaval in the Middle East. As I’ve written before, there is a growing need for this kind of curation, but there is also the need to start looking at news as a process and not as a pristine, finished product.

While many media outlets have web editors who track reports on Twitter and other social media, the BBC is unique (as far as I know) in having a special desk that sits in the middle of the newsroom and pulls in reports from Twitter, Facebook, Flickr, YouTube and anywhere else it can find information. The desk staffers then try to verify these reports, and some of the ways they do that are fascinating. In a sense, they are trying to do CSI-style forensic analysis in real time, by checking things such as:

  • looking at weather and shadows to confirm that the conditions shown fit with the reported date and time of a photo or video.
  • checking the accents and language in a video or other report to make sure that they are consistent with the location, using BBC staff who speak those languages.
  • verifying locations against maps of a region, and cross-checking images against known images from that location.

Andy Carvin has also talked about how he sees verification as being a big part of his job in curating Twitter feeds from people in the trouble spots he is reporting on — feeds that are made up of people he has hand-picked because they are reliable in some way. Whenever a video is posted that purports to be of an attack, or reports of government leaders or high-profile individuals killed, Carvin spends a lot of time trying to get confirmation from others in the region or other reliable sources.

Storyful and curation

Mark Little, the founder of a curation tool called Storyful — which allows journalists and others to collect videos, photos and other content easily about a specific topic (in much the same way that a similar tool called Storify does) — says that his staff of curators take much the same approach that the BBC’s user-generated content desk does, in trying to confirm reports and videos that claim to be from a particular spot. They also:

  • review an uploader’s history and location to see whether he or she has shared useful and credible content in the past, or if they are a “scraper,” who passes off other people’s content as their own.
  • use Google Street View images to help verify the locations in a video or photo.
  • monitor social-media traffic to see who is sharing the content and what questions are being asked about it.

But what’s interesting about the Storyful founder’s post about this process is that he admits that sometimes he has posted things when he wasn’t sure whether they were correct — including a photo that he posted during the floods in Queensland, Australia that showed a crocodile in a parking lot. As he says in his post: “It was too good to be true, so we decided to drop the image into the social media conversation around the floods. Suddenly, we were part of a furious debate about crocodiles in Australia,” which ultimately proved that the photo was a fake. Carvin says that he also uses his network to check the veracity of reports, because he knows they will know more than he does.

Crowdsourced fact-checking

There are no doubt many mainstream journalists and media observers — including some at the BBC, who were critical of Carvin’s approach to Twitter reporting during the social-media summit — who would see this kind of approach as reprehensible, and argue that reports of any kind shouldn’t be posted on Twitter or anywhere else without verifying them. But others (including me) would argue that what Little did was a sensible way of approaching the new real-time nature of news reporting: in other words, post a report and say that it is unverified, and see if anyone can help you verify it.

Little calls this the “human algorithm,” and I think it’s a necessary step in the future of journalism and media. If anyone can report, and anyone can function as a journalist, then we need everyone to be able to help confirm and verify reports like the ones Andy and the BBC are getting. That’s not something a single person or even a group of editors at a specific outlet can do in real-time. Media analyst and journalism professor Jeff Jarvis has talked about the need to look at the news as a process instead of a finished product that is turned out at a specific time and in a specific format, and that is something more media outlets need to come to grips with.

In a sense, this is no different from the approach that news wires have taken in the past: a report comes in, it is fact-checked as much as possible and then it goes on the wire, and when it needs to be corrected, an update goes out. Twitter is becoming the real-time news wire for the world, and we need people who can make use of it as such — more people like Andy Carvin, as entrepreneur Bernard Lunn argues in a recent blog post. And we need new attitudes about how we look at journalism as well, now that everyone is doing it.

Post and thumbnail photos courtesy of Flickr user Luc Legay

12 Responses to “Future of Media: Curation, Verification and News as a Process”

  1. The “human algorithm.” What an interesting and telling term.

    As creators or curators of content, each individual or brand has a responsibility to understand and provide insight into the content we share. Truthfully, this is precisely why content curation tools are so valuable. By effectively eliminating the time it takes to locate interesting and relevant content, curators can focus their efforts on evaluating the value and validity of the information.

    Pawan Deshpande
    CEO, HiveFire (creators of content curation solution Curata)

  2. As a communications professional, I’ve been using StoryCrawler for content curation. Love working with this company. Great content and easy to then redistribute the material.

  3. Jonathan Marks

    Great overview and reasoned thinking. Thanks Matthew. It is obviously going to be easier for organisations like Huffington Post than the BBC to use the human algorithms. I think it also depends on whether what you’re trying to verify might save a life or prevent an accident, rather than checking if an incident is real or fake. But it’s very important that this process of verification and curation takes place. It’s what journalists do best. And then there are a few important stories where it is just as important to produce with the audience rather than just for them. Flooding in Australia, earthquake in NZ. That must be the biggest challenge in Northern Japan right now. NHK is a broadcaster, but lousy at understanding how to build a public conversation.

  4. Liz Hannaford

    As an institutional, mainstream radio journo for 15 years, I’m really fascinated by the idea of news stories as a constant process rather than something that stops once the final despatch has been written, filed and broadcast. But when should the casual audience member dip in? Won’t they always be left feeling they’re only halfway through the story, here? This kind of curator journalism demands a lot of the audience and a real sophistication. Listening to a crafted 6 o’clock news bulletin is much easier. The news stands still for a time allowing us to catch up.

    • Thanks for the comment, Liz — I agree that consuming news as a process is harder than just reading or viewing a static report, but I think there is a need for both. Even with something like Andy Carvin does, there is a place for someone to curate what he is doing and create a summary report that gives people a sense of what is going on. It’s an evolving market, that’s for sure.

    • To piggyback on what Mathew said, I think there is an increasing need to serve this more sophisticated news consumer. I think a big part of where many mainstream news orgs have gone wrong is treating all news consumer the same, assuming they all have the same informational needs, are all interested in the same topics, consume news the same way, etc. I think the future will partly be built on serving different demographics differently and the highly engaged, sophisticated, discerning news consumer appears to be a growing group.

  5. Zane Aveton

    Mathew, Great Post. I love the concept of giving informational “Power to the People” and smartly processing the results – including the point where news should always be viewed as evolving information.

  6. Another attribute to consider is the source’s social capital. Services such as PeerIndex and Klout numerically rank an individual’s social capital across various subjects (eg: technology, politics, arts etc) using proprietary algorithms that produce a normalized score between 0 and 100. By no means perfect, but definitely a tool in a curator’s belt for determining the credibility of a source from a social network.

  7. Matthew…really good post.

    The concept of the ‘human algorithm’ is interesting and while this terminology is new to me, the idea of mixed (human and machine) curation is already a component in some of the socially-oriented curation platform.

    I don’t track this from the news biz side but curation (with human controls) as a context filter for individuals and businesses to cut through the noise of the web and find relevant information is the key behind a group of start-ups trying to crack the information relevancy question.

    You point to curation as as a human aided ‘fact-checker’ for new organizations. Sites like Eqentia focus on the human-managed curation process to control automatic scraping for info and to ameliorate some the irrelevance that crops up from popularity driven filters.

    Good stuff, thanks.

    My take on filters and curation from the individual and business side here:

    “Context not content is king” @