BlockByBlock

Ideas and discussion from BxB2010 Summit

Credibility Accuracy Reliability and related subjects

leave a comment »

Being part of “the media” brings along some baggage. There’s the whole “you brought down Richard Nixon” thing right alongside “you cheerleaders said there were weapons of mass destruction in Iraq,” and it doesn’t matter if you’re the national daily or you’re the local guy with a website.

For the local guy with a website, you’ve the added problems of the fact that you’re right there, right in the middle of the community, and sometimes you’re up against established newspapers.

Following is loosely about a discussion between Tracy Record of the West Seattle Blog, Matt Hampel who works with open data, Scott Rosenberg of MediaBugs, Daniel O’Neil of Every Block, Bora Zivkovic, a science blogger with Science in the Triangle, Barbara Iverson of Chicago Talks. I’m an editor at Wicked Local, and I jump in at some parts.

Rosenberg noted “There’s this paradox for the hyperlocal sites.” There’s the community around the sites that put a lot of trust in what hey read, meanwhile there’s a lot of talk about journalism imploding, and a lack of information you can trust. O’Neil agreed there’s a bad reputation out there for media, but for the local sites, it sometimes gets down to “I like that guy” or “That guy helped my sister” as the reason to trust a site.

Zivkovic thought it was a matter of scale. You’re most likely not going with a weapons inspector in Iraq to check the veracity of a report, but you can walk down the street to the scene of a crime and check it out. O’Neil disagreed, saying the small town is different from NYC, where you can’t just walk down the block and see the shooting for yourself – oftentimes the “community” is thousands and thousands of people.

Rosenberg: Do people classify the reporting then as a local thing? Does that mean they trust locally-produced news differently?

Tracy Record noted she and her husband, the main producers on her site, were anonymous for years. There was no face, no this-guy-helped-my-sister personality that was behind the site, so they had to gain their credibility from the work itself. And while her site has become this nationally-linked to site for news in her area, she still gets questioned because she’s tagged as a blogger. People still give more credence to the small legacy-newspaper organization, even if the staffer who is reporting for the newspaper has decades-less experience than Record. The reason? “Well, they’re a newspaper.”

A discussion ensued about a story the West Seattle Blog reported upon the day before, where there was a dispute among media reports as to how many were detained by police. Record said she was being second-guessed by readers because she’s not with a newspaper.

Rosenberg: Is there a place you discuss other media reports?

Record said “If we get something wrong, we’ll get it in the story.” She noted that when the story changes, she needs to note what was previously reported, and how that was either wrong or has changed. (IE, now one suspect has been released so five, not six, are being held for a crime.)

Zivkovic said at Science in the Triangle, the site has been separated into news and blog. The “reporters” on the news side have their credentials stated up front, giving them authority, and the bloggers are sometimes trained in journalism, sometimes not. But those with credentials are lending some of that authority to the whole site.

Record noted there’s an about us page doing that, but folks are still taking “blog” out of West Seattle Blog and assume the folks are idiots.

Rosenberg noted that there’s a wisdom out there that once someone learns “fact 1” that’s all he knows about the subject, and his opinion doesn’t change. His experience was being at Salon.com when it was a site you had to pay for. While that is no longer the case, there are still people out there who don’t read Salon because they’re sure it is behind a paywall. Record nodded, saying she herself remembers “Salon Premium.”

O’Neil asked what the signaling is to show you have credibility. Is it design? Is it tone? His website is all about data – the web design is (his words) flat, the phrasing is flat. It feels authoritative. Rosenberg noted that there are a lot of cues in design and presentation, but in the long run, credibility is gained with the dynamic record you create over time.

O’Neil noted his site goes to great lengths to shows there’s some gray area when it comes to plain old facts. Take the time the plane hit a bird and landed in the Hudson River. Every reporter on the planet wanted data about bird strikes and planes out of the FAA. However, the FAA knew the data they collected was from self-reporting by air traffic controllers. Some like filling out forms. Some see it as their job to be compliant. Some folks have other priorities. So the data they have makes the folks who actually report every bird strike as being more dangerous, while the ones who don’t fill out the forms at all look like safer airports. But it is hard explaining that to a journalist on deadline, and we all need to give other humans a break.

Rosenberg said “the bird strike story is a great one, and it is the media outlet that is going to be perceived as screwing up”. The ones who do the best job at error reports get punished. The reward mechanism is screwed up.

Powers: But when you’re going on data and government reports and they’re wrong, the public is going to note how you messed up by going on that data – despite everyone agreeing that a police report is by public officials for the public record. If the cop messes up the charges in his report, and I write about it, it’s not like I’m writing from 30,000 feet – I’m in the community with the 25 family members of the arrested guy saying I screwed up. You can correctly report information that is wrong, but folks will question your credibility.

Iverson: My students scrape up the reports that are not the final thing. But my local paper treats the police blotter as something that came from on high.

O’Neil: To get all futuristic about it, that large family will have their own platform someday, be it facebook or something else, and they can report their side of the arrest.

Rosenberg: Interesting. Longterm, if the community knows that these police reports are going to be public, they’ll know it is important to be done right.

Iverson: But you Google yourself. ….

O’Neil: and you’ll pop up. It’s fine.

Record: Since the West Seattle Blog is updated frequently, Google indexes the site quickly. Forums aren’t “Pre-moderated,” and sometimes people will post inappropriate information. She takes it down quickly, and she does some “Truth-Squadding” but it is still indexed and searchable.

Rosenberg: There’s still an ever-shifting procedure for correcting on the web. Some do the strike-through, but news sites (not blogs where it was common) often see that as inelegant. What it does for journalists is it makes the process for getting something corrected more formal and slow. What should the norm be?

O’Neil noted that his website has a weird model, being giant lists of data. “We use the no-cache model.” They find inaccuracies, like restaurant inspections where the information has changed, and since they don’t want to mess with data, they just take it off the site.

Rosenberg: Which is different from journalism practices, which is, you correct the error but you do not deface the original. At Salon, we used to change it when we made a mistake like it didn’t happen. Sometimes it is libelous, and you don’t want to leave the original in place. But in databases you can’t do that.

O’Neil: And it’s an evolving practice. You can’t “nuke” an item that has been circulated in the public and has comments and has become part of a public discussion, but something that has zero pageviews, you can.

Rosenberg: We’ve had people come to MediaBugs to report an error because the story has been changed from the time it was posted on the web and then putting it in print. He suggests  a policy like Wikipedia, where you index the versions of the story.

Discussion followed over the coolest way to do that – do you mouse over a highlighted text and it notes where you can get past versions highlighted? How do you handle a story you’re updating with the real-time news? IE, First version is “shots fired” and then you update a few minutes later with “police responding.” Rosenberg noted in WordPress there’s a “Post Revisions Display.” Zivkovic said that in reality, most folks won’t check, but Record noted that it would be something easy to do but would bring credibility to the site. Zivkovic agreed it would be “huge credibility.”

Zivkovic noted there are six editions of “Origin of the Species,” and you need to note which one you’re quoting from if you’re going to be accurate and accountable. Granted, Darwin wasn’t transparent about that at the time, and it wasn’t clear between editions that there were changes, but modern comparisons have shown there are six individual versions of that one book.

So, you need to link out. Zivkovic, a science blogger, links to the original science pages.

O’Neil likes the links of twitter. It is “a clearing house of humans” and you can always find them and link to them.

The discussion broke up with a quick repeating of what was important:

Be open about corrections. Show when stories have been revised.

Note your credentials when relevant, know being accountable will, in the long run end up as part of your permanent record of stories.

Get active about truthsquadding

Link!

Give people a freaking break when they’re reporting birdstrike data

Advertisements

Written by katpowerswl

October 8, 2010 at 7:33 pm

Posted in Uncategorized

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: