Dispatches from the living amongst journalism's walking dead

Tag: crowdsourcing Page 1 of 2

In Defense of Crowdsourcing: You Get Out What You Put Into It

Yesterday, Dashiell Bennett via Atlantic Wire* brought some pointing-and-laughing to the plight of  Darren Rovell, a CNBC sports business reporter who was burned by a source he found via crowdsourcing and then pouted publicly online about it.

Bennett, however, takes a leap that defies logic: He blames the act of crowdsourcing for this error. Blaming crowdsourcing for failed reporting is akin to blaming phones and email – it’s merely a method to find sources, the end result still requires actual work.

Last fall, Rovell asked his Twitter followers how the NBA lockout was affecting their businesses. He received a response from a man who claimed to be the head of an escort service catering to NBA players and fans who noted that his business was down 30%. After a few questions over email, Rovell added the source’s comments to his story. As it turns out, said source was a high schooler punking this reporter. (Wait, high schoolers watch CNBC???)

Here was Bennett’s break-down of what happened:

Rovell was done in by two classic journalism mistakes. The first, less obvious one, is that crowdsourcing is a lousy way to gather news. As Rovell himself suggests in his CNBC mea culpa… people will say almost anything if they think it might end up in print, and people you don’t know and never meet can’t really be trusted. It happens to lots of people, because it’s very tempting to rely on these kinds of tips. The information comes so easily, but it needs to be taken with twice the amount of salt. The second is a more traditional maxim: If a story is too good to check, it probably isn’t true.

Well, he’s right on one count: You really can’t trust information that you get from any source anywhere – via crowdsourcing or otherwise – so you have to do a little reporting and fact-checking on things like this. He is wrong, however, to suggest that this somehow proves that crowdsourcing in and of itself is a lousy practice for journalists. In actuality, crowdsourcing can be a very effective way to find sources, but it’s what you do with those sources that determines the outcome of the story.

Crowdsourcing wasn’t Rovell’s problem, failing to take a couple of extra steps to find out if this too-good-too-be-true source was for real was his problem. Had he even called this guy on the phone and asked a few key questions, the kid’s story would have likely fallen apart in a matter of minutes.

Not that Rovell helps his case by including this in his “apology”:

He duped me. Shame on me. I apologize to my readers.

As a result I will do fewer stories on the real life impact of big events which I do think the public enjoys.

There will always be people out there who want their 15 minutes of fame and not really care how they get there.

The lesson was not “I shouldn’t accept anonymous sources I interview over email at face value”, it was “I’m not talking to you people ever again.” Sigh.

Reporters should take his story to heart as a cautionary tale, it shouldn’t scare people off from crowdsourcing altogether. Instead, note these basic steps of verification for all hot tips you receive as a reporter:

1. Evaluate the credibility of the source

2. Follow up on the information with reporting – including interviewing the source

3. Evaluate the credibility of the information 

4. Corroborate the info you receive against other sources

5. Evaluate your options based on the info you have

 

Here are some resources that might help you find your own verification process:

 

* Yes, I realize that in merely responding to this opinion post,  I’m falling directly into Atlantic Wire’s trap (also known as Forbes and Gawker Media’s trap), but it has to be said.

Growing Your Audience: Advice from Bloggers and Readers

I’m giving a workshop to the Digital Ninjas at the New Haven Register at 1:30 p.m. (EDT) today about engaging and growing blog audience. If you’d like to follow along when it starts, it’ll be online on Chris March’s blog.

I’ve been crowdsourcing some advice from the bloggers and blog readers in my social networks to include in today’s talk. Got something to add? Leave  comment or tweet me @mjenkins.

[HTML1]

UPDATE: Slides from today’s presentation are after the jump.

New Crowdsourcing, Curation and Liveblogging Training

Hi all! I’ve been traveling a lot for Digital First lately to spread the gospel of social media to my colleagues. So, if you’ve seen my presentations before, you’d know that I make very wordy Powerpoints so that people who weren’t there to see me prattle on about my favorite things can still follow what we went over (also, they keep me on task in-session).

So here are some recent training sessions that might be of use to you, your staff (or students, if you teach). Please let me know if there’s anything out of date or if you know of new tools I should be touting around DFM and on the interwebs at large.

Basic Overview of Twitter, Facebook and Google+

Social Media for Photographers

Social Media Roles for Editors

Curation for Journalists (including a step-by-step guide to Storify)

Crowdsourcing and Verifying Tips From Social Media

Liveblogging, Livetweeting and Chats

Advanced Social Media Techniques (gathers a lot of the above)

B.S. Detection for Journalists

Eds Note 10/5: It was brought to my attention that the links in this slideshow are not clickable in the embed here. I included them all below this post. 

Ever see a tip that’s too good to be true (it probably is) or a photo so amazing you just can’t believe it (don’t)? Sometimes you can’t just follow your nose to know what’s good and what’s bad on the social web – so you have to be extra careful in the verification and vetting process.

Following is the presentation I gave along with Craig Silverman of Regret the Error at the Online News Association Conference on Friday, September 23.

Our presentation went over how to verify tips, facts and images gathered via social media and the web. It also has a few case studies that demonstrate why this is so important.

[HTML1]

 

 

Links included:

 

Additional Reading:

 

Sarah Palin’s emails and a call for collaborative journalism

If you were committing an act of news on Friday, June 10, chances are every national news organization missed it.

Why? We all had boxes and boxes of printed emails of an ex-political official to go through. From the New York Times to Mother Jones/MSNBC/ProPublica, the Washington Post and my own employer – many national news sources spent enormous amounts of human capital to scan, upload, display, read, analyze and crowdsource Sarah Palin’s emails.

While I won’t delve into the newsworthiness of this effort (that’s a whole other Pandora’s box), I wonder why it had to take a village to carry it out. In the hours and hours everyone invested into this effort — what was missed?

Credit: RambergMediaImages

Credit: RambergMediaImages

Consider this… This information, in and of itself, was not exclusive. The FOIAed information was being released to all media that wanted it at the same time – so why did the media compete over the actual obtainment and presentation of the information?

Why couldn’t the nation’s largest news sources put aside their shared need to own information and just combine their efforts to quickly get these documents online in searchable form? Hey, stop laughing. I’m serious.

Think about it. When the documents dropped, one effort could have been taken to scan and text-translate these emails. There would have been one upload to a site like DocumentCloud, thus preventing the major backlog on that site Friday as we all tried uploading to the same place.

Instead of driving readers to several similar-but-different document displays, the heavy-hitters could have built a single site (off all our brand servers). This hypothetical .org could have an embeddable search functionality and open API that could be displayed on all news sites, from the Austin American Statesman to the Zanesville Times Recorder – and on several platforms.

It wouldn’t eliminate competition – it would just drive news organizations to compete smarter in a way that uses less resources. We could compete over crowdsourcing efforts, i.e. who built the best app on top of the API for readers to submit findings; who built the best contributor network; who had the best resources on the back end taking in the tips, etc.  We would also be competing on how quickly and intelligently we analyzed these found facts – and how we displayed them on our sites.

Think of the time that could have been saved had we split up the labor. Think of how much faster we coud have gotten to the actual journalism – and then moved on to the next story.

In an age where we’re all doing more with less, why can’t we come together over the simplest of agreements to aid the effort of newsgathering as a whole?  Why couldn’t it work?

Note: I’d be happy to start the beer summit for the top editors of news orgs to hash this out for future efforts. Seriously.

Interacting with the audience as a news brand

Last week I went over a few tips for setting a social media strategy and persona for your news org’s branded account(s) and tips for using those accounts as a brand. Today, let’s get into audience engagement on social media tools. These tips have served me well as both a brand and as an individual, helping me to establish great relationships on the old internets.

Audience Interaction in the Social Sphere

Responding: I’d suggest you try to always respond to those who reach out to the brand on Twitter or Facebook with questions, criticism and tips. You can respond via private direct message (if they follow you) or outward replies. If you’re squeamish about public replies, remember: Unless a Twitter user is following both parties, they will not see this interaction in their streams. If the reply is something you think other followers may be interested in, you might want to re-tweet the question/comment and answer it outright.

Note: After you DM a user, you will need to follow them from the brand’s account if you want a reply via direct message. People cannot direct message you unless you are following them.

News Tips: If someone sends a good news tip via Twitter, Facebook or by email, ask them follow-up questions (if necessary) and be sure to publicly acknowledge their contribution. You may want to re-tweet the tip once you have verified it.

Ask for help: If you want a photo or info from the scene of a story, ask for it from your followers.

Be thankful: Treat those supplying you with information as a respected friend – and they just might end up becoming regular tipsters and brand evangelists. If you get a news tip, photo or other info you’d like to use from your followers, be sure to thank and/or credit the user by name in social media messages and the story itself. Credit them on Twitter with (h/t @theirname) or similar when the link is shared. If you use a photo, make sure they are credited in the cutline.

Questions: If someone asks the branded account a question, answer it as soon as you have time (or ask someone else to do so). If you don’t know the answer, tell them you’ll find out and get back to them.

Criticism: If someone offers criticism, address it, even if only to say you’ll pass it to the right person internally. Try to avoid an extended back-and-forth with Twitter users and don’t get into embarrassing Twitter arguments. Once it seems to be approaching a point of no end, take the conversation offline.

Corrections: If you made a mistake, like a misspelled name, wrong link or factual error, it’s best to correct it in a follow-up tweet. Do not erase the first tweet unless you absolutely feel you must – and not without some acknowledgement of the mistake.

Start the conversation: Instead of always offering up a headline and link, add a question element when appropriate. Like, “What’s your favorite”, “do you agree” or whatever. If you want this to be an ongoing topic for the day, you may want to start a hashtag to accompany it.

Responses: If someone gives you a particularly good response, RT it with a link back or some notation as to what it’s about. Example: RT @someguy I think it’s a bad idea. // What do you think of Md’s new traffic law? http://bit.ly/ghgkg

Note: This is an especially good way to keep an ongoing topic going throughout the day. Use this to keep a hashtag going instead of tweeting out a boring old headline on the same thing again.

There’s a whole Internet outside of Twitter, so don’t forget it

Pew released a new study on Twitter demographics today that found only 8 percent of Americans on the web use Twitter. Of that 8 percent, only 2 percent use Twitter on a typical day. Keep in mind that about 74 percent of American adults are internet users, meaning that the Twitter users make up about 6 percent of the entire adult population.

This news shouldn’t be surprising, but maybe it is to those who live in the Twitter echo chamber.

When all of your friends, your coworkers, your spouse and the media you consume are on Twitter, it may seem logical to believe a great deal of America is as well. This is a dangerous assumption for journalists and media organizations to make – and I know I’ve been guilty of it from time to time.

While I still think it is very important for journalists to use Twitter, the following facts must be emblazoned on the brains of media Twitterati:

  • Twitter represents a very small group of people in your area.
  • Being popular on Twitter doesn’t necessarily make one popular or important in real life.
  • Re-tweets, replies and Twitter referrals do not adequately represent the larger interest in or importance of your work as a journalist.
  • Most people that use Twitter don’t use it to get news.

Now, the study. The Pew study did find some interesting demographic tidbits that should be making us rethink how we approach the tool.

  • There are more American women using Twitter as opposed to men (10% to 7%)
  • Internet users ages 18-29 are significantly more likely to use Twitter than older adults.
  • African-Americans and Latinos web users are more than twice as likely to use Twitter as are their white counterparts.
  • Urban residents are roughly twice as likely to use Twitter as rural dwellers.

So what does all of this mean?

Recommended reading: Investigative social media, new ideas and tools

Sorry it’s been so long, but it’s been crazy busy as TBD’s preparing for the holidays and other events. This’ll be a quick one, just a few links I’ve been reading of late. Have a happy Thanksgiving, folks.

Social media roundup

  • How Investigative Journalism Is Prospering in the Age of Social Media – Great ideas from several resources gathered by Vadim Lavrusik at Mashable on how to use social media in investigative reporting and newsroom projects. Includes tips on Crowdmap, Storify, Twitter crowdsourcing, data searches and more. A great post to pass on to the social media haters in your newsroom.
  • RockMelt: The User Manual– If you don’t know about Rockmelt or want to know more on how to use the new social browser, here’s a great guide from the NY Times.
  • 6 innovative uses of Tumblr by newsrooms – The big media companies are only now getting into Tumblr, but there’s a lot of possibilities out there for it.
  • Engaging Facebook fans with clever, conversational updates – Great ideas from Web Up the Newsroom for writing interesting status updates on a media outlet’s Facebook page to drive traffic to content and drive discussion online.
  • In this disturbing bit from FishbowlDC, a Washington Post editor says “crediting the original source of a scoop isn’t “a requirement or even important” because “all news originates from somewhere” and “unless one is taking someone else’s work without attribution (that is, plagiarizing it) any news story should stand on its own and speaks for itself as an original piece of work.” Hm.
  • How News Organizations Are Generating Revenue From Social Media – Another great Mashable rundown of the top ways online media is generating revenue using social media and more to hit new audiences.

On the TBD Front

How TBD’s election day experiments turned out

As promised, I’m reporting back on TBD’s election day experiments.

The Voting Issues Crowdmap seemed to be successful. We had a lot fewer reports than in the primaries, but considering D.C. turnout was lower and there were fewer issues reported overall, I’d say that’s accurate. We got 20 accepted reports in and some were very intriguing. Take this one:

My husband & I were told that we could not vote unless we left the children (11 month old & 3 year old) in the hallway. Then the security guard…told us we could take the 11 month old, but had to leave my 3 year old unattended in the lobby & if we didn’t want to do that, then we were unable to vote!

The Foursquare experiment was interesting, though it didn’t yield overwhelming participation. This was kind of expected, considering we didn’t have a lot of lead time and, it turns out, the venues were not easily found on the Foursquare mobile app. Next time, we’ll have to try enlisting Foursquare’s help instead of going out on our own. Here’s how it went down:

At the Virginia venue, 21 people checked in after voting and two left tips about their votes. D.C. had 66 people check in, Maryland had 51 check-ins and six tips about who they voted for in the governor’s race. Those who checked in at both the Maryland and D.C. election venues ended up earning a Swarm badge (if they didn’t have one already). All who checked in and said they were voting got Foursquare’s custom “I Voted” badge. If you missed it, Foursquare also set up a special site to track these voter check-ins as a measure of turnout.

Link roundup: Facebook deals, Times paywalls, ONA and news experiments

Geolocation meets deals

Last news first. Facebook announced today that it will be doing more with its location feature, including offering deals tied to location. This could spell trouble for other geolocation providers like Foursquare and Gowalla, group buying sites like Groupon and, sadly, news sites looking for revenue streams. Facebook is offering these deals for free right now – and who’ll buy the proverbial cow through the likes of us when they can get the milk for free from Facebook?

A consumer/business side take on Facebook Places from D.C. blogger Lisa Byrne at DCEventJunkie outlines the potential on the local level. Facebook seem to have a lot of options for businesses of many sizes and kinds (including charities) to take advantage of the deal service.

Paywalls busted

Also today, GigaOm declares It’s Official: News Corp.’s Paywalls Are a Bust. NewsCorp’s Times (in London) lost 90% of its online traffic after putting up a paywall earlier this year. Somehow, they paint this as success, as they see a smaller online audience that is paying for their service as better than a large one getting it for free. Advertisers, it seems, disagree.

Election Experiments

I blogged here about what TBD was doing for elections (will update today with how that all turned out).The Nieman Lab and Lost Remote documented what news organizations around the country were doing to cover the 2010 election using news media and social media tools. Some great ideas in these posts from the likes of the Huffington Post, NPR and Washington Post.

More Adventures in Storify

Speaking of newsroom experiments, we at TBD are still in love with the tool. Burt Herman, who created the tool, was in the office Monday to tell us a few tips and tricks as well as take suggestions for improvements. Burt is awesome. We’ve been trying it in lots of different instances and news situations. Here’s a few of them:

Online News Association Conference

This was my first ONA conference and I was lucky to have it be in D.C. I volunteered, so I didn’t see many panels, but I was on the Friday keynote panel about TBD’s launch. Since I don’t have great notes, here’s some posts that summed up a lot of the conference’s highlights.

On Jobs (also ONA)

I sat for an interview Friday at ONA with Kent State student Nicole Stempak about journalism jobs for college grads. She asked me to explain how I’ve been fortunate enough to create my own positions in social media and online news since I left college. A few people asked me to share it, so I’m posting it here.

Using Foursquare and Crowdmap to track local elections

With the midterm elections coming tomorrow, lots of news outlets will likely be launching their fanciest new toys and social media ideas to best pull in that coveted election night audience on the web. TBD’s no exception, though our election day experiments are based in off-site crowdsourcing to better inform our on-site coverage.

As we did with the Washington D.C. primary elections, we’re launching a Crowdmap to track voting problems across the District, Northern Virginia and parts of Maryland. We’re asking readers to report long lines, broken machines, ballot refusals, electioneering and such at the polls using email, Twitter hashtags or on-site reports. It worked pretty well in September, though this time I’m a bit worried about Crowdmap’s servers holding up. Right now as I post this, they’re struggling to load any of our maps.

I’m particularly excited to try out Foursquare in Tuesday’s election coverage. In our attempt to take a local approach to the National Post’s Foursquare “exit poll” experiment, I set up three new venues to collect voter check-ins.: Virginia Congressional Elections 2010, Maryland Governor Election 2010 and D.C. Election 2010. Once readers check in to the correct venue, we’re asking them to submit a tip to that venue telling us who they voted for and why.

While these check-ins and tips won’t be any kind of real measurement, it will give an interesting look at how many people on Foursquare are voting in local races and how those votes are leaning – if people use it. If enough people check in, they could earn a Swarm badge, at least.

Here’s hoping something takes off.

How to build, manage and customize a Crowdmap

So you’ve got a great idea for a user-contributed map you need to launch RIGHT NOW. Ushahidi’s Crowdmap makes it pretty easy, and hopefully this post makes it even easier. All examples shown are from TBD’s Crowdmap for D.C.’s election.

First of all, if you’re mapping a crisis, Crowdmap recommends checking our their Emergency Response Strategy first (pdf).

Also, check and see if anyone else has done your map idea with a Google Search. If someone else has already built a map of what you want to do in the same area, maybe you should just help them out instead of replicating the work.

The Quick Build

Sign up for a Crowdmap account at www.crowdmap.com and log in.

1. Click on Create New Deployment

2. On the deployment setup page, pick a url, name and tagline for your map. Keep SEO in mind here to make it easier to find. (You can edit this later, so don’t sweat it too much). Click Finish.

3. Click on admin dashboard for your map or go to http://yourmapname.crowdmap.com/admin

This is your map’s Dashboard. Bookmark it. Your map is now live and activated. If you need to launch it right now, you can – though there’s further additions and customizations you can do. Note: With the default settings, people will only be able to submit reports on the site.

More after the jump (had to do it for images…)

Page 1 of 2

Powered by WordPress & Theme by Anders Norén & Hosted by Pressable