From Dan Froomkin’s piece at Nieman:

Experts in journalistic ethics are encouraging reporters to take a more critical posture going forward.

“I think professional journalists have an absolute obligation to make lies transparent,” said Kelly McBride, who teaches ethics at the Poynter Institute.

The first step is “to do the reporting so that you can with authority point out that this is an act of deception,” McBride said. With the Romney ad, that was easy; the accompanying press release provided Obama’s full quote.

Step two, McBride said, is to assign blame.

“Professional journalists need to follow up and figure out who in the campaign is responsible for this,” McBride said — “and keep at it.”

“We have to call that sort of thing out,” Geneva Overholser, director of USC Annenberg’s School of Journalism and a former Washington Post ombudsman, wrote in an e-mail. “Journalists did that, last week, but it seems to me it needs to have been even stronger. This was such a blatant act of deception. Treating it just like any other fact-checked ad, so many of which contain something mildly misleading, is itself misleading.”

I encourage you to read the whole thing.

The second piece is a post by Brendan Nyhan at Columbia Journalism Review. (Disclosure: I’m a columnist for CJR.) Nyhan, an assistant professor at Dartmouth College, is doing some of the most interesting academic research into misinformation, and he’s now writing regularly for CJR. He looked at how some publications dealt with the Romney ad. Then he offered some advice on the best way to combat the misinformation:

… A better approach would be for reporters to characterize the accuracy of ads in their own voice and to invoke non-partisan experts like PolitiFact. In some cases, it may even be possible to find credible sources on the side of the candidate airing the misleading ad who are willing to state the truth. For instance, WMUR’s Sexton briefly paraphrases a GOP strategist conceding that Romney’s ad is misleading: “Republican political strategist Mike Dennehy says it definitely appears as though Obama’s words are taken out of context.” This approach might be especially persuasive to Republicans who are inclined to trust Romney and distrust the mainstream press.

Still, it’s important to be realistic about the effectiveness of this sort of fact-checking. My co-author Jason Reifler and I have shown in our research (PDF) that fact-checking frequently fails to reduce misperceptions among the ideological group that is most likely to hold the misperception and in some cases makes the problem worse (what we call a “backfire effect”). In this case, reporters should not expect to convince skeptical readers that ads they support are false.

Nonetheless, aggressive fact-checking can provide reputational incentives (PDF) for elites to make more careful claims.

Earlier this year I wrote a CJR column about the backfire effect, and how it presents a challenge for journalists:

We present facts and evidence, and it often does nothing to change people’s minds. In fact, it can make people dig in even more. Humans also engage in motivated reasoning, a tendency to let emotions “set us on a course of thinking that’s highly biased, especially on topics we care a great deal about”.

These two important cognitive effects can have a significant impact on society and debates in the public sphere. They also end up negating some of the debunking and reporting work done by the press.

*Correction December 2, 2011: The original headline on this piece misspelled debunking as “debunkng”. The first sentence also used the Canadian spelling of “across”. I use US spelling on this site, so I corrected that as well. I regret the errors, and thank rogerogreen for spotting both!

This offers an opportunity for me to explain how we handled the situation. I’m not saying we did it perfectly. I’d be eager to hear comments from readers about anything we may have overlooked or handled poorly.

As a starting point, here’s the text of the apology that was published:

Around 2:45 p.m. today, we published a story about Jericho Upshaw, a Halifax man charged with first-degree murder, and his association with the North End.

It relied on two sources who said they had encountered him in the neighbourhood on several occasions. Unfortunately, the story was wrong: we had the wrong Jericho. The details in the story were about another young man with the same first name. Our two sources had seen Upshaw’s photo in a local paper and identified him as the man they had encountered in the North End. They were mistaken, and we were wrong in publishing the story.

Within about five minutes of posting the story online and sending out a tweet about it, we realized we had the wrong Jericho. We removed the story from our site and deleted the related tweet.

Our practice is to not unpublish content. We don’t want to erase the record; we prefer to correct it.

But in this case, the story was based on a mistaken identification by two sources. It was built on the information from those two sources. As a result, we felt it was essential to remove the story and publish this explanation of what went wrong. We also redirected the URL from the mistaken story to this blog post in order to ensure that anyone who finds that piece of content will understand why it was removed.

On behalf of OpenFile Halifax, I apologize for this error and will do everything I can to make sure something like this doesn’t happen again.

Neal Ozano
Editor, OpenFile Halifax

This is the kind of situation we journalists dread.

“My heart sank,” said OpenFile editor in chief Kathy Vey in an email to me today. “Neal had asked me to look over the story for legal issues, because it was a murder case. There weren’t any but we had a much more basic problem. We named the wrong guy.”

The story was based on totally incorrect information. It shouldn’t have been published. So what do you do now that you know that? I hate unpublishing content; it should only be done in very specific circumstances. In my view, this was one of them.

The story itself was completely incorrect. Once we knew we had the wrong person, there was nothing left in the piece that offered value to readers. So it had to come down — and we had to replace it with something as soon as possible.

I worked with Halifax editor Neal Ozano and editor in chief Kathy Vey to draft the apology. We also contacted our tech team to alert them that we would need to redirect the URL of the unpublished story to the new blog post. That way, anyone who found their way to the offending story would be given an explanation of what what happened, of why the story was gone. Sending people to an error page is not acceptable. You have to make it easy for them to understand what happened.

The apology itself also needed to be clear about what the mistake was, how it occurred and what action we took once we realized our error. Again: help people understand what went wrong. Be clear. We also had to be honest about the fact that it was a failure on our part. You can’t just blame your sources. It was published on our site. We needed to accept responsibility and offer a clear apology.

A final consideration was the fact that our OpenFile Halifax Twitter account tweeted a link to the story. So as soon as the post was live and URL redirect was working, we tweeted the link to the apology from the same local Twitter account. That was then retweeted by me, Kathy Vey and by our national OpenFile Twitter account. Our Toronto news curator also retweeted it of his own accord. We wanted to make sure Twitter followers had a chance to see the apology.

As of today, the apology is still on the homepage of our Halifax site. (It will probably scroll off by the end of day.)

Now the job is to make sure this doesn’t happen again. We cover local news in our cities, but we don’t do a lot of crime reporting. So the story was a bit out of our comfort zone and area of expertise. We will be extra cautious with any stories of this nature in the future. We’ll also ask ourselves whether what we have is really news. We will add extra eyes to contentious stories.

What else should we do? What could we have done better? I’d love to hear your thoughts.

These are some of the common questions I’m asked about online corrections. I think a lot of news organizations continue to struggle with them, and to look for guidance. I’m pleased to say that thanks to the Canadian Association of Journalists, we have the best and most comprehensive guide to help answer these and other important questions about online corrections.

Yesterday the CAJ released “Best practices in digital accuracy and corrections”, a set of useful best practices to help newsrooms make sense of online corrections. It’s the result of hard work by a panel created by the Ethics Advisory Committee of the CAJ. I serve on the Committee and also served on the panel, though I’ll note I was added after the group had already done some great work.

I hope journalists the world over will read the document and use it as a basis for developing policies and practices. If you have feedback, I’d love to hear.

I wrote an accompanying article to go with the best practices document. Read it here. An excerpt:

For several decades, newspaper readers in the U.S. and Canada have found the day’s corrections tucked away on page two of their paper of choice. This placement became a de facto standard after the New York Times began the practice in 1972. Newsrooms followed the Times’ lead.

Today, in 2011, we have yet to see a similar moment that standardizes the placement, wording and handling of online corrections. There are a myriad of questions when it comes to correcting online content, and news organization continue to struggle with best practices. At the same time, online errors spread father and faster than ever before. They speed across social networks and are deposited into search engine caches and databases large and small.

It’s essential that news organizations establish best practices for online corrections and digital accuracy. That’s why the Canadian Association of Journalists asked its Ethics Advisory Committee to create a panel to do so. The panel was chaired by Toronto Star public editor Kathy English and included Bert Bruser, Tim Currie, Rod Link, Shauna Snow-Capparelli, Scott White and myself.

The result of our work, “Best practices in digital accuracy and correction,” was publicly released this week. I encourage all journalists to download and read the document.

In addition to a set of clear and actionable best practices, the report provides useful (and interesting!) background on the challenges of online accuracy and corrections, and a look at what leading news organizations are doing. It offers much needed clarity for newsrooms when it comes to online accuracy and corrections …

But what’s notable about the Times story is the information offered up in response by readers’ representative Deirdre Edgar:

When readers write in about errors, it shows they care, and that’s a good thing. Research done for the American Copy Editors Society this spring confirmed what the Readers’ Rep inbox tells us.

Fred Vultee, a journalism professor at Wayne State University, studied a group of readers over a three-month period. His findings:

• Readers who read more news tend to be more critical than people who read less.
• Dedicated readers expect a higher level of quality than casual readers, particularly in terms of grammar and professionalism.
• Readers notice grammar errors and find them troubling and distracting.
• Readers notice writing that is garbled and confusing, and when words are misspelled or misused.

Most readers are less concerned about errors of style and story structure than they are about professionalism and grammar. “They really don’t care if you abbreviate ‘road,’ Vultee said. “They don’t care if you start a paragraph with a number.”

I hadn’t heard about Vultee’s work, and there’s a summary of it here. There’s one other key finding that’s worth noting:

Readers can tell edited from unedited stories in significant ways.

That’s an interesting finding. I hope to learn more from Vultee about some of the differences noted by readers.

We’re a couple of weeks into it and, as expected, there have been some gems. It’s a positive move for the paper to standardize the way it presents printed corrections. Unfortunately, the paper has not been so diligent when it comes to adding corrections to its online content. The Mail is still a work in progress.

However, this week it received a favourable ruling from the Press Complaints Commission, a self-regulator for the British press. The Guardian’s Roy Greenslade laid out the details. The important thing to note here is that the PCC found that front page errors by the paper did not require front page corrections. From Greenslade:

In so doing [the PCC] took the opportunity to set out its thinking as to what constitutes due prominence, making clear that it “has strong regard for the location of the original article”. This, however, could not be the “only determining factor”.

Its ruling stated: “The commission will consider the full circumstances surrounding the complaint: the nature of the breach of the code; the scale of the error; the full context of the story; and the existence or otherwise of a designated corrections column”.

In other words, it is all about proportionality. There is a general rule, but each case must be treated on its merits (or demerits).

In these two cases, the commission clearly decided that although “the mistakes were sloppy, the issues were not personal to the complainant and had not caused personal harm.”

In addition, the PCC took the view that the errors did not render the coverage of either story to be wholly inaccurate. So page 2 corrections were sufficient in both cases.

Here’s what PCC director Stephen Abell had to say:

“The issue of ‘due prominence’ as set out in the Editors’ Code will never be an exact science and, as the adjudication makes clear, there will always be legitimate calls for newspapers and magazines to highlight corrections with greater clarity. It may be appropriate in some cases for a newspaper or magazine to publish a front page correction. However, whilst the errors in these two cases were unsatisfactory, the Commission judged that the publication of suitable corrections on page 2 was a sufficient offer which met the ‘due prominence’ requirement of the Editors’ Code”.

I met LaForge in person in the spring when I gave an error prevention workshop at the Times. (I offered to give the session; he accepted. I wasn’t paid.) I mention him now because the Times Topics blog reprinted a notable memo LaForge sent to editors at the paper. Among other information, it included a list of more than 10 proofreading tips “culled from years of journalism tip sheets.”

LaForge also addressed the issue of mistakes in the Times:

Mistakes in copy have been a problem since Gutenberg, but it is hard to shake the impression that we have been slipping more than usual, especially in articles that are rushed onto the Web site, bypassing some of our traditional steps. Yes, speed is important in the modern news competition. But readers still have high expectations, especially now that they are paying for online news.

And he explained that the paper has a policy of giving every article “at least two reads”:

This era of news publishing has put a greater emphasis on speed, across multiple formats and platforms. Thanks to blogging and continuous updates, more people in the newsroom find themselves in the role of publishing live material. The same forces have increased the workload and distractions faced by reporters, backfield editors, copy editors and producers. It can be tempting to cut corners. You might decide, unwisely, to save some time by bypassing the copy desk. There is rarely a justification for doing so. Our policy is for every article to get at least two reads, preferably one of them by an experienced copy editor, before publication.

And then you should check your work again, or have someone else check it.

But the gold of the post was that list of proofreading tips. Here we go:

• Break your mind-set: Read the copy out loud. Read it silently, one word at a time. Read it backward and focus on the spelling of words. Print a copy. Preview it in a different application. Change the format or the screen resolution. Justify or unjustify the type. Take a break and return to it with fresh eyes.
• Use spelling checkers but don’t trust them. In particular, be aware of homophone confusion: complement and compliment, accept and except, effect and affect, oversees and overseas.
• Memorize frequently misspelled and misused words. Here’s a list: http://www.yourdictionary.com/library/misspelled.html.
• Beware of contractions and apostrophes: their and they’re, its and it’s, your and you’re.
• After reading for content and spelling, proofread separately for punctuation.
• Beware of doubled words at the end and start of a line. A doubled “that” will often slip right by if you let it.
• Double-check proper names and claims of distinction (first, best, oldest, tallest, etc.).
• Double-check little words that are often interchanged: or, of; it, is.
• Check all the numbers, especially any reference to millions, billions or trillions. Do the math. Do the math again.
• Set aside a regular time to review stylebook and usage rules. This includes backfield editors and reporters. If you don’t want someone to change your story on style grounds (and perhaps introduce an error), learn the basics and follow them.
• Be aware of dates and days of the week, especially in advance copy or copy that has been held. Be aware of references to next month/last month around the time the month is changing.
• Make a personal checklist of the things you tend to miss. Use it on every story.
• Have someone else, preferably a copy editor, read behind you.

Good tips, and I of course love that he advocates the use of a checklist. Full post is here.

Thanks, @BriscoPhoto!

Here’s more about the chat:

A Pew study released last week found that only 25 percent of the news consumers surveyed said that news organizations get the facts right. Two-thirds say stories are often inaccurate, and nearly 75 percent believe journalists try to cover up their mistakes.

In a live chat today at 12:30 p.m. ET, Reuters’ Jack Shafer and Regret the Error’s Craig Silverman will talk about the connections between accuracy and trust. We welcome you to join the chat with your thoughts and questions. You can also tweet your questions using the hashtag #poynterchats.

It sometimes publishes a corrective article when a mistake is made, or it will occasionally go into online articles and scrub away mistakes. If it thinks it might get sued, the Star will offer an apology. But the paper’s editor believes corrections are a bad thing and she does her level best not to publish them.

Needless to say, I’m not a fan.

Officially, the Star ethics code states, “Whenever it is noticed that a misleading, inaccurate or distorted article has been published, it must be corrected at the earliest opportunity.” This is similar to the wording in the Media Council’s Code of Conduct. But what does that really mean in practice? To answer that question I sat down with Star Editor Catherine Gicheru.

Gicheru says that she has no problem admitting an out-and-out error — for example when the paper, earlier this year, mixed up an MP’s wife with another woman of the same name who had been charged in a fraud case, the paper carried a page two ‘Apology’ the next day.

But Gicheru says that she tries hard to avoid corrections whenever possible. The main reason, she says, is her concern about the paper’s credibility. “If you have too many corrections, then your credibility starts to plummet,” she says. And, she says, a lot depends on your competition: if they aren’t printing many corrections, you don’t want to print many either.

This a backward attitude, though it’s one that used to prevail at many news organizations.

A few quick points: If you have too many errors, your credibility can plummet. It’s not the same for corrections. And if your competition doesn’t publish corrections, you have an opportunity to demonstrate your commitment to accountability and professionalism.

The truth is one of the worst things a media outlet can do when it comes to errors is to refuse to acknowledge them, or to offer disingenuous corrections. There is evidence readers feel better about the quality of a newspaper when they see corrections. It demonstrates accountability.

There are some weeks when she fields as many as three serious demands for corrections, she says, and other weeks when there are none. Her first line of defence is to let the complainant simply blow off steam. Often, she says, that, plus a sympathetic manner and a believable explanation, is enough to cause the caller to relent. “Most often people cool down and say, ‘Okay, I can understand how it happened,’” she says.

Understanding how it happened is good, as is talking to people who request corrections. But it’s wrong to talk them out of a correction when there was a mistake.

Some of her other time-tested responses are to offer the person the opportunity to write a letter or a commentary. Or, she may offer the prospect of a friendly story at a later date to make up for the offending one.

Really? Gicheru would rather promise favorable future coverage than issue a basic correction? That’s unethical.

Gicheru also favours a technique that I’ve also seen used elsewhere: finding a reason to run a follow-up story that corrects the error without ever acknowledging that one has been made …

Anything to avoid a correction…

Gicheru says that if she decides that there is no other course but to print a correction, she will determine what to call it according to its severity. If it’s an honest mistake, it will be called a ‘clarification’. If it’s inexcusable, it will be called a ‘correction’. And if it has the potential to turn into a libel case, it will be billed as an ‘apology’. “You’re really pleading at that point,” she says.

I don’t think readers understand the distinction between a clarification and a correction, so I usually counsel news organizations to stick to one term. However, if you’re offering an apology then by all means call it that.

After hearing the many ways in which the paper’s editor goes out of her way to avoid corrections, Rothmyer offers this view:

Maybe it’s because I’m no longer in the hot seat myself, either as a reporter or an editor, but I personally would like to see more Star corrections (called, moreover, by their right name rather than the weasel word ‘clarification’).

Agreed, but what about the scrubbing and the unethical practice of promising favorable coverage after an error?

Hat tip to Stinky Journalism.

*Correction August 3, 2011: The original headline on this post left out the word “to”. Thanks to @scottleadingham for spotting the typo!

The episode led to a couple of interesting pieces of debate about how journalists should treat rumors on Twitter. From journalist Neal Mann:

Reuters blogger Felix Salmon also weighed in with a post on his Tumblr arguing that Twitter “behaves in many ways a lot more like a newsroom than a newspaper.” More:

Rumors happen there, and then they get shot down — no harm no foul.

I think that big flagship Twitter accounts like @Reuters or @WSJ should be held to a higher standard. But for the rest of us, we’re conversing on Twitter just like we converse in real life. In the newsroom, we say things like “did you hear that Piers Morgan just got suspended?” and that’s fine. Is it really that bad to say that kind of thing in the new newsroom called Twitter? I don’t think so. People are human, they believe rumors, make mistakes, jump to conclusions. Twitter is just a healthy reminder of that fact.

True, it’s a reminder. But this rumor originated with a fake account. It was pretty easy for Jon Snow to take an extra few minutes to check the source of the tweet and see that it wasn’t credible. You can debate and celebrate the self-correcting nature of Twitter, but it bears repeating that this wasn’t a case of confusion or conflicting information. A fake Twitter account tweeted a hoax and a bunch of journalists fell for it. That seems pretty straightforward. I thought Dan Gillmor put it well:

I don’t mean to sound dismissive of the points made by Salmon, a writer whose work I enjoy; but this incident doesn’t strike me as a good example of how Twitter can be such a good tool for correcting misinformation and moving reporting and discussion forward.

Finally, here’s a great Storfiy from Ross Neumann that tracks the spread and correction of the rumor.

It is hard to describe the interview that took place on KQED’s Forum show on May 25, 2011, as anything other than a train wreck.

Osama bin Laden was dead, and Frank Lindh — father of John Walker Lindh, the "American Taliban" — had been invited on to discuss a New York Times op-ed piece he’d just published about his son’s 20-year prison sentence. The moment host Dave Iverson completed his introduction about the politically and emotionally charged case, Lindh cut in: "Can I add a really important correction to what you just said?"

Iverson had just described John Walker Lindh’s 2002 guilty plea as "one count of providing services to a terrorist organization." That, Frank Lindh said, was simply wrong.

Yes, his son had pled guilty to providing services to the Taliban, in whose army he had enlisted. Doing so was a crime because the Taliban government was under U.S. economic sanctions for harboring Al Qaeda. But the Taliban was not (and has never been) classified by the U.S. government as a terrorist organization itself.

This distinction might seem picayune. But it cut to the heart of the disagreement between Americans who have viewed John Walker Lindh as a traitor and a terrorist and those, like his father, who believe he was a fervent Muslim who never intended to take up arms against his own country.

That morning, the clash over this one fact set host and guest on a collision course for the remainder of the 30-minute interview. The next day, KQED ran a half-hour Forum segment apologizing for the mess and picking over its own mistakes.

KQED’s on-air fiasco didn’t happen randomly or spontaneously. The collision was set in motion nine years before by 14 erroneous words in the New York Times.

This is the story of how that error was made, why it mattered, why it hasn’t been properly corrected to this day — and what lessons it offers about how newsroom traditions of verification and correction must evolve in the digital age…

– via “Corrections in the Web Age: The Case of the New York Times’ Terror Error,” a great piece by Scott Rosenberg and Mark Follman of MediaBugs. A must read.

It also has a list of recently corrected articles:

Both these things are improvements. The links to recent days of corrections makes additional corrections easily accessible from the page. The list of recently corrected articles also provides some nice context, as some readers may recognize a familiar headline and click through to read a correction they may have missed.

Three additions I’d like to see are an RSS feed for corrections, a link to a page where readers can learn more about the paper’s corrections policy, and a form people can use to submit a correction. (Have a look at the form on the Chicago Tribune’s corrections page.)

Apart from that, it would be great to see the Times embrace a report an error button for its content.

… In the field of corrections as anywhere else, “openness” isn’t binary — it has gradations and nuances. I like to imagine these as a sort of ladder of transparency that news organizations need to climb.

On the first rung of this ladder, journalists readily fix mistakes they learn about and conscientiously disclose and record the details of each fix. (Most newsrooms declare allegiance to this ideal but, sadly, our MediaBugs research shows, the majority still fail to live up to it.)

One rung up, news outlets effectively solicit error reports from their audiences, making it clear that they welcome the feedback and will respond. The Report an Error Alliance is trying to push more news organizations to climb up here.

On the next rung up, newsrooms also willingly expose their own internal deliberations over particular controversies, explaining why they did or didn’t correct some issue readers raised and leaving some sort of public trail of the decision. At some publications, the ombudsman or public editor takes care of some of this.

On the final, topmost rung, the news organization will assure accountability by turning to a neutral third party to maintain a fair record of issues raised by the public. This shows external critics that the newsroom isn’t hiding anything or trying to shove problems under the rug…

The starting point for the Commission will be that, if an article appears in print and online, the proposed remedy will often appear in both media. This note is not designed to be prescriptive, and will take into account the existence of differing practice. The test, in the end, will be whether the requirement of "due prominence" is met. The following points are relevant:

• Negotiation is a key part of the PCC process, and discussion between complainant, editor and PCC will be necessary in the placement of online – as offline – corrections and apologies. Clause 1 (Accuracy) of the Code states: "in cases involving the Commission, prominence should be agreed with the PCC in advance".
• Readers will access information on newspaper and magazine websites via different means (such as searches or links), so there is not automatically a correlation between the original location of an article and the placement of a correction or apology. The existence of a paywall may impact on how a site is initially accessed, and this should be taken into account. However, for stand-alone corrections and apologies, editors should give consideration to appropriate placement on the relevant section where the original article appeared (such as the "news" or "showbusiness" section, for example).
• If the resolution to a complaint is a stand-alone text (an apology, correction or letter), it will generally be appropriate to link to the original article under complaint (should it still be published online) and for the original article to link back to it. If the original article has been removed, then how long the apology, correction or letter should remain online should be the subject of negotiation with the PCC.
• Corrections or apologies that appear on the original article should be clearly marked.
• If the outcome of a complaint is that the text of the article is significantly amended, then consideration should be given to the publication making explicit reference to the existence of the alteration. How quickly the text has been amended will be a factor in this consideration.
• Care must be taken that the URL of an article does not contain information that has been the subject of successful complaint. If an article is amended, then steps should be taken to amend the URL, as necessary.
• Online corrections and apologies should be tagged when published to ensure that they are searchable.

I was hoping the guidance would endorse a dedicated online corrections page (here's the Guardian's), and come out against scrubbing away errors. PCC Watch also raised another relevant point:

In the longer term, the PCC will need to consider how it treats tweets and Facebook status updates from journalists where measuring online prominence is far harder. Equally, timely corrections of content that has appeared on apps will need to be addressed sooner rather than later.

… has become riddled with typos, grammatical mistakes and intolerable "small" factual errors that erode credibility. Local news coverage, once robust, has withered. The Post often trails the competition on stories. The excessive use of anonymous sources has expanded into blogs. The once-broken system for publishing corrections has been repaired, but corrections often still take too long to appear. The list goes on.

Much of this is a result of upheaval, disruption and necessary cost-cutting. Over the past few years, once-separate

print and online staffs have been combined. The traditional newsroom structure was blown up and reconfigured. New editors are in charge. Scores of staffers have been reassigned.

And in a Q&A with readers, Post executive editor Marcus Brauchli seems to agree:

The ombudsman is right that there have been too many careless errors lately–typos, grammatical errors, silly factual mistakes. I don't want to make excuses, because we shouldn't tolerate these sorts of errors. But by way of explanation I will say that we have made a number of changes in our processes in the last couple of years and are putting in a new editing system that will further change workflows for editors. We try to be diligent about publishing corrections, and the data are interesting: we published 1,054 corrections in 2010, up from 1,040 a year earlier and 961 in 2008, but down from a recent high of 1,319 in 2005. We have room to improve.

It's good news that the Post is bringing in a new system. The old one has been cut and altered to the point that it's a shell of its former self. Let's hope the new system improves workflow and quality. A challenge, to be sure. But a worthy one.

I’ve since learned what real, excruciating pain NPR triggered with its false news report that Rep. Gabrielle Giffords had died, which was repeated on npr.org, e-mail alerts, Twitter and picked up by other news organizations.

NPR correspondent Ted Robbins is based in Tucson. He was at the scene Jan. 8 when his cell phone rang shortly after NPR aired at 2:01 p.m. EST that Giffords died. The call was a friend, who is also a friend of Giffords.

The friend was sitting outside the hospital operating room with Giffords’ mother Gloria, holding her hand.

“Please tell them to stop reporting she is dead,” he begged Robbins. “She is in surgery.”

Robbins immediately called NPR but was told NPR was sticking to the story since it had two sources.

Scott Simon, host of NPR's Weekend Edition Saturday, got a similar call. Simon and his family are close friends of Giffords and her husband, astronaut Mark Kelly.

At 2:08, a distraught family member called: "Scott, where the hell is NPR getting that information?"

"The close family member is a huge NPR fan or was,” said Simon. “Until that moment, he found NPR more reliable than other news outlets. He had been told that Gabby was in surgery. But he was anxious, isolated, and wondered why NPR would report such a thing if it were not true.”

Simon phoned the NPR news desk and was told the information was based on "confirmation" from the Pima County Sheriff’s department and a congressional source.

He didn’t think that was good enough …

Simon sent these suggested operating principles to top management:

“There should be no room for doubt when a news organization declares someone dead. They should wait until the medical authorities directly involved declare death, or close family members announce it. There is simply no way that anyone else—not local police, not witnesses, not 'two governmental sources'—would be in a position to know for certain especially when there are now, between respiration and brain activity, at least a couple of medical gauges of death.”

– via How NPR Giffords' Mistake Hurt The Families, a blog post from NPR ombudsman Alicia Shepard.