Posts about credibility

Credibility is not binary

I’ve seen a couple of efforts lately to help determine who’s credible online and though I understand the need and the motive, these attempts are fundamentally flawed and perhaps even more damaging than they are helpful.

The latest is Newscred and Techcrunch describes it today. I used a Techcrunch beta invite to poke around.

It’s very simple — though that’s the problem; credibility isn’t so simple. They list articles and you get to “credit” or “discredit” them. These scores are, in turn, compiled for writers and publications.

The first and most obvious problem, which TechCrunch points out, is that this is bait for grudges. Fox from one side, the Times from the other will get discredited by their detractors all day long. One man’s bias is often the other man’s truth.

The second and more fundamental problem is that there’s no basis to decide credibility. Does one error ruin an article’s credibility? How many discredits does it take to ruin a reporter’s or a publication’s? And then what does that mean? That they lied? That you don’t believe them? That you don’t like them? That they make mistakes? That they don’t report enough? That they use anonymous sources? That they relied on bad sources? That they wrote it badly? That they weren’t transparent?

And who’s doing the judging? Are they credible? Who’s judging the judges, then?

Over the years, I’ve heard of various attempts to determine credibility or bias algorithmically, in an effort to take out this human bias in the process of finding bias, but that’s just an engineer’s wet dream. Again, the problem is definition (not to mention technical limitations of analyzing text and ideas).

Newstrust has tried to do this in a subtler way, with star ratings and comment, but it faces the same issues: Who’s doing the rating? On what basis?

I think these folks are attacking the problem from the wrong perspective. They’re trying to play whack-a-mole with credibility and identify all the bad stuff — just as news people, long accustomed to packaging the world in a pretty box with a bow on top, keep wanting to kill every bad comment on their sites. They’ll fail. Life insists on being messy. The task of identifying the bad stuff is so large — there is, indeed, too much junk — that these folks try to scale their effort with simplicity or technology. Won’t work. They’ll never find all the bad stuff. Ultimately, this can be dangerous because good people who do good work can easily be besmirched by bad judges with grudges.

Instead, I think it would be far more useful to concentrate on finding the good stuff. That is the real challenge in the new architecture of news and media, in the ecosystem of distribution and aggregation. When all the articles on a given topic are brought together by Daylife (where, disclosure, I am a partner) or Google News the need and the true service is to find the best articles because that’s what we want to spend our time on. (A restaurant guide with only bad reviews doesn’t help me eat.)

We also need to find ways to surface original reporting so we can support that reporting with our attention (and with traffic and ads). This is why I believe that there should be an ethic in professional journalism, as there is in blogs, to link to prior work and sources. All roads should link back to the original reporting.

There is still clearly bias in this approach of finding the best. Many will recommend Paul Krugman, many won’t trust that recommendation. Who’s doing the recommending still matters (and so it would be very helpful to have transparency among them). But by highlighting the good rather than trying to expunge the bad, we would try to support good journalism wherever it is done — msm or blog. And that’s really the point, isn’t it?

On top of that, every news site should have a means for people to help correct errors — that’s as simple as adding comments (though doing so does add a cost to police them). Correcting errors makes one more credible; that, too, is an ethic of blogs. And that, too, will improve the journalism, just as you improve mine in comments here. At the end of the day, there’ll always be disagreements, though. Look at the post below about airlines; there’s plenty of argument there. Is that really about credibility? No. It’s about conversation.

Reuters and trust

Tom Glocer, head of Reuters, posts a speech he gave on news and trust in Israel on his blog. It starts with a good summary of what an old-media exec should be saying about the new, two-way world and then he tackles the Reuters photo affair in Beirut, which Reuters did handle well, reacting immediately. Then he gives technical news:

I am pleased to announce today that we are working with Adobe and Canon to create a solution that enables photo editors to view an audit trail of changes to a digital image, which is permanently embedded in the photograph, ensuring the accuracy of the image. . . .

It is important to say that we sought this technical solution, not because we don’t trust our photographers – far from it. I am incredibly proud of the amazing and dangerous work our photographers and journalists do. They all too often risk their lives to get the photograph that tells the true story of a conflict or captures the horror of war. The threat of injury or death is a daily hazard for many.

No, we sought a technical solution so that we had total and full transparency of our work. It’s what we stand for. It’s what we’ve always stood for. And we hope that it will provide reassurance to editors and consumers of our services.

When I discussed this innovation with one of our best photographers, Ahmed Jadallah, in Dubai last week, he welcomed the transparency. He almost died three years ago in his native Gaza photographing the Intifada, but he wants the world to know his photos are 100% truthful.

Glocer also talks about the lessons of the incident:

So what does the Hajj incident tell us? There are three key lessons:

The first is accountability. The upside of the flourishing blogosphere is that beyond our own strict editorial standards, there is a new check and balance. I take my hat off to Charles Johnson, the editor of Little Green Footballs. Without his website, the Hajj photo may well have gone unnoticed.

The blogosphere provides accountability. They’re not always going to be right. Indeed, many of the accusations levelled at traditional media are partisan in nature – but some are not. We have to listen to the bloggers – we shouldn’t ignore them.

The second lesson is about the trust of our audience. We learned at Reuters that the action of one man – a man who wasn’t even a full-time staff member – could seriously hurt the trust in our news, built assiduously over 155 years. His stupid decision to clone smoke cost us.

We learned that your reputation is only as good as the last photograph you transmit, or the last story you file.

The final lesson we learned was this – more than ever the world needs a media company free from bias, independent, telling it as it really is, without the filter of national or political interest.

If you searched across the Web during the Lebanon conflict you saw many entrenched and extreme views – on either side. There were thousands of voices opining on the war from their own particular standpoint. This cacophony of voices is exciting and it does for the first time give a true flavor of all views. It is also provides a marketplace for ideas.

But I strongly believe that in the mixing of different voices we will always need a place for the news organization whose watchword is trust. Trust will be the differentiator in the new media dynamic. Your independence and impartiality will mark you out.

Telling the story truthfully is more important than ever. Reporting it without spin and without editorializing is critical if history is to accurately record events.

This was a theme I heard at the American Academy of Arts & Sciences event I spoke at the other night (and didn’t bother blogging because not much happened, by the way): People from big, institutional journalism said there will always be a role and a necessity for such institutions. I won’t argue, operationally, that there will be a role for institutions, although they won’t be as big as they are now.

But I don’t fully buy the organizational analysis of trust that they and Glocer make. Does the institution — by its actions, standards, or reputation — enhance credibility and trust? In some ways, yes; I trust, for example, that a New Yorker story will be fact-checked better than my blog. But still, we will all judge the the trust of any story or writer ourselves and that does not hinge on the building from which the come.

In his list of lessons, I think Glocer misses the big lesson he himself just taught only a moment before: the value of transparency and its ability to bolster trust. That will matter for all — individuals or institutions — who publish with the intent to be trusted.