One of your Twitter followers posts that the mayor will call it quits tomorrow.

Your social media gut says share the information with a dose of skepticism and see if anyone can corroborate it. Your journalistic head says don’t touch it — you don’t know where it’s been.

The issue of whether journalists should forward information they find in social networks without verifying it cuts to what has traditionally been a core principle of journalism: we publish true things only.

But does retweeting constitute “publishing” as we’ve traditionally defined it?  News organizations such as the LA Times and the Associated Press say it does — and journalists should apply the same standards as they would with any story. Others such as Reuters and the BBC suggest retweets constitute a process of journalism, not an end product.

I’ve been working with three colleagues on the Canadian Association of Journalists Ethics Committee over the past couple of months to prepare a draft of guidelines for retweeting or forwarding information found in social networks. We conclude that journalists can gain a lot by participating in real-time social networks. But if they retweet, they should apply a checklist of questions about sourcing and aim for transparency about what they know — and don’t know.

The draft will be discussed Friday, May 28, 2010 at the CAJ conference in Montreal. Take a look and tell us what you think.

[I'm at the Mesh conference in Toronto]

The Guardian was the first news organization to really harness the power of crowdsourced data with its MP expenses database.

Now it has the lofty goal of distributing its technology worldwide to expose wrongdoing.

The man who has the job of opening up The Guardian‘s data technology told an audience at the Mesh technology conference in Toronto this morning: “The plan is to make the API (Application Programming Interface) open source.”

“We’re also exploring the possibility of making it a data clearinghouse,” in essence, allowing other people to upload their own data.

Chris Thorpe, The Guardian’s Developer Advocate for its Open Platform, set out the grand goal of  helping the oppressed and the impoverished worldwide. “We want to heal the world,” he said. “Openness is the friend of good.”

In June 2009, The Guardian hired a programmer for a week and built a portal to distribute more than 400,000 government documents of MPs’ expenses. The project prompted thousands of people to join The Guardian’s reporting team in pouring through claims and highlighting ones for further investigation. It was a huge step forward in newsgathering.

The Guardian API aims “to shine light on corruption,” Thorpe said. ”Let’s be a world liberal voice.”

The audience pressed Thorpe to state The Guardian’s stance on the current trend toward news paywalls. To that, he said, ”The people putting up paywalls will really reduce their influence in the world.”

“We’re seeing good advertising revenue from being open,” he added, but refused to talk numbers.

He said The Guardian is starting to realize “We have to behave more like a newspaper” by increasing audience engagement with more in-depth content.

“What we’re finding is the more pages people view, the more people are likely to click on adverts,” he said.

The Halifax Chronicle Herald signalled today it’s going to beef up its commenting system to improve the level of discourse on its site. Good to hear. Much of what passes for comment on news websites continues to be the lowest possible level of name-calling and uninformed bravado.

The Herald’s director of news content Dan Leger states:

“So we’re working on ways to end abuse, partly through registration and software to encourage use of real names and identify abuse. It won’t be perfect, but perhaps we can temper the over-the-top attacks.”

The issue of weak commenting systems was highlighted nationally last Wednesday when Halifax weekly The Coast said it would surrender the IP addresses of commenters following a Nova Scotia Supreme Court order. It’s the first time a news site has done that in Canada. Two senior fire officials at the Halifax Regional Municipality had sought to unmask the identities of these commenters, alleging they defamed them by painting their behaviour as racist and incompetent.

The move resulted in many people online and on local radio imploring news organizations to adopt a policy requiring real names for commenters as a means of raising the quality of online conversation.

They wanted to hear from someone like Howard Owens, who argues a real names policy is practical and enforceable at his New York State news site The Batavian.

I check public databases for names that match in the zip code provided. If no match, the user is asked to provide either by fax, e-mail or in person a copy of a picture ID.

Owens, himself, calls his policy “a ‘best effort’ practice,” siding with the news executives Richard Pérez-Peña cited in his New York Times article last week who “say that merely making the demand for a name and an e-mail address would weed out much of the most offensive commentary.” He may be right.

But while this may work at smaller news organizations, it is likely to be unworkable at sites that get tens or hundreds of new registrations a day. Pérez-Peña says as much in the Times story, which details the growing unease of news organizations with anonymous comments.

Real names strikes me as an offline solution to an online problem.

I argued in an interview on CBC Radio’s Mainstreet Nova Scotia last week that the big problem with the comments areas on most news stories is they seem to be abandoned by the news site itself. The result is the so-called “broken windows” problem — it becomes a wasteland vacated by moderates who have been long-since shouted down by blow-hards.

Author/blogger Scott Rosenberg states: ”Show me a newspaper website without a comments host or moderation plan and I’ll show you a nasty flamepit that no unenforceable ‘use your real name’ policy can save.”

I’m much more intrigued by Gawker’s approach. It implemented a more robust software system last year that gave its staff and community a way to promote the “funniest, thoughtful, intelligent, well-argued” comments. It divided its community into two tiers:

• a small community of “starred” commenters “who have proven themselves to be engaged, intelligent, humorous, fair-minded, thoughtful, rational.” These people can promote “well-written, thought out, intelligent and/or otherwise notable comment” below stories.
• The rest, whose comments will be obscured behind a “Show all comments” link

A key aspect is that the discussion is guided by an engaged community with an interest in creating a place people want to be. In addition, the community is fluid, with new commenters rising to become stars, and people who abuse their star power falling to become relative nobodies.

Gawker Media CTO Tom Plunkett suggested in this graph last week, that the site recovered — and thrived —  from an initial drop in comments that resulted from stripping people’s status based on the number of followers they had — and making many comments more difficult to view. He concludes: “purging commenter accounts is not a solution for the out-of-control commenter community. Nor is a large moderation staff.”

Will this completely stop defamatory comments? Probably not. Editor moderation is likely the only way to do that. But it’s a bottom-up — not top-down— prescription that seems a better fit with the medium.

There was a torrent of announcements from Google this week — among them Living Stories, which has intriguing prospects for journalists.

As Dan Gillmor suggests, a better name would probably be Living Topics. The service, which is an amazing technological feat, groups news stories and deconstructs them for better online viewing. Google says the automated initiative, which it launched in co-operation with the New York Times and the Washington Post, does three things:

• Puts the entire coverage of a story under a single URL
• Chunks up the story elements by theme and form
• Customizes the reading experience so each person sees story developments new to them
  

The main benefit appears to be that it automatically generates context for stories by creating bite-sized content summaries and aggressively linking between them. For example, users can sort the story content by subtopics, major characters, quotes, external links, images, graphics, video, audio, etc. The size of the story summary also indicates its importance and newness, according to this Google video.

The Living Stories site isn’t optimized for my iPhone in any way. No doubt that’s coming; this sorting and grouping feature would seem tailor-made for mobile users.

Each topic starts with a dynamic topic summary and timeline. But the service doesn’t just reformat existing content. At the story level, there are some subtle differences from the versions that appear on the news outlet’s website. Take the War in Afghanistan Living Story and a story that’s part of it (Afghan Says Army Will Need Help Until 2024) , which is also on the Times’ website. In the Living Stories version …

• a link to Hamid Karzai, for example, goes not to a detailed bio like the one on the Times site, but to a pop-up snippet description
• there’s a link (“pledged to begin withdrawing American troops”) to a related story that isn’t made on the Times site
• an interactive map, which is not in the Times story, shows the location of the capital, Kabul.

The service has the potential to improve the online reading experience in a number of ways:

• by narrowing the inverted pyramid, so stories are less broad at the top — and more to the point — because the context surrounds it
• by including more summaries, which usability advocate Jakob Nielsen says are crucial for online reading
• by more aggressively linking content than human editors can — thereby improving the user experience on small screens

Still, there seem to be limitations. The importance of topic pages would seem to be diminishing as people increasingly consume content at the story level via Twitter and Facebook links. And I can think of many stories that won’t fit easily under topic designations.

But it’s a worthy innovation. I’m waiting for the mobile version.

The Halifax Chronicle Herald has joined a growing number of news organizations that are live blogging events to break news quickly and build engagement around them.

It live blogged 8 events this past summer during Nova Scotia’s provincial election. Seven were political events — party platform releases or debates — the other was a discussion of the Herald’s use of social media. The Herald used CoverItLive, the leading (free) live-blogging service based in Toronto.

Rick Conrad, editor of thechronicleherald.ca, calls it a “vital” means of engaging readers and its effect is evident in the quality of conversation: “You get a more defined audience – almost a more educated one – because it’s something they are interested in.” He acknowledges the events attracted the occasional “crank” but “99% of the people were there for the right reasons and wanted to have a real discussion.”

Some tips from the Herald’s experience:

• Promote, promote, promote: People need advance notice to schedule a live blog into their day. Conrad promoted them on the site, in print, and on Twitter. The result was some success in expanding the site’s audience: “For the social media blog, I got the distinct sense it attracted people who wouldn’t normally think of picking us up every day or coming to us online.”
• It takes two: If you’re covering an event, you generally need at least two people – one to report and the other to engage the audience. Says Conrad: “With the platform releases, [the audience] would come with a lot interesting — and tough — questions like “What does this mean?” When you’re listening to what the leader is saying and trying to go through the documents, it’s difficult to make a snap analysis.” Conrad suggests bringing a reporter along, if you can. He concedes it’s difficult because a reporter frequently has to file a story as well. But he adds, “Readers really like to talk to a person directly involved in the story.”
• Let the audience lead: When to drive the conversation and when not to? It depends on the topic, says Conrad. If the audience is getting information from other sources it often works best to let them lead the conversation. “With the election night blog, we essentially took the approach that ‘this is their night’ and we would pop in with the occasional comment on the results … Whereas with the platform releases, I found there was more interaction with me because I was their eyes and ears.”
• … but keep it civil: Conrad says he approved the vast majority of comments, but rejected a few: “For me, the reason for doing the live blogs for the election was to give as many people as possible the chance to comment.” However, a few were “nasty” and wildly speculative. “We didn’t think that was appropriate. You want [the conversation] to be substantial and you don’t want it to spiral down.”
• Use it to assess your own efforts: People are especially engaged when they feel they can make a difference. The Herald received almost two comments for every person who attended the session in which it asked how it could make better use of social media. While Conrad acknowledges the difficulty in drawing conclusions from raw numbers, he says the tone of the conversation was spirited. “For me, the people who were involved in that blog were really into it.”
• Keep it available for replay: The number of people who replayed the Herald’s live blog the day following election night was almost twice the number who participated live. Some people don’t have time to follow a live stream — or aren’t available when it’s scheduled. But many are interested in seeing which topics resonated with the audience.

Conrad says he would like to live blog a provincial budget release and he hopes to experiment with the tool at a concert or a live arts event.