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 …

AT THE Birmingham Mail, we pride ourselves in journalism that is honest, accurate and fair.
We do our best to correct errors or publish clarifications as soon as possible.
Our journalists adhere to the Editors Code of Practice, which sets the benchmark for high professional standards and is enforced by the Press Complaints Commission*. From today, any corrections or clarifications will be published on Page 2 so we can offer a more prominent platform for setting the record straight.
If we do slip up, we promise to put things right on this page in a clear, no-nonsense manner.
To ask for an inaccuracy to be corrected, all you have to do is: Email: stacey.barnfield@birminghammail.net, Phone: 0121 234 5491 Write to: The Editor, Birmingham Mail, Floor 6, Fort Dunlop, Fort Parkway, Birmingham B24 9FF.
*The Press Complaints Commission (PCC) is the independent body which governs the newspaper industry. The Code of Practice and details of how to lodge a complaint are available from Halton House, 20/23 High Holborn, EC1N 2JD. Website www.pcc.org.uk or email complaints@pcc.org.uk
Telephone 020 7831 0022. Helpline: 0845 600 2757.

The same note was published by the Huddersfield Examiner and others. The Daily Mirror made one slight alteration to the text to note that it was moving its existing corrections section to page two:

For years, the Mirror’s For the Record section has provided an opportunity to clear up mistakes.
Today, that column moves to Page 2 so that we can offer a more prominent platform for clarifications and corrections. If we do slip up, we will set the record straight on this page in a clear no-nonsense manner.

All Trinity Mirror papers will have page two corrections, according to Roy Greenslade of the Guardian.

First, Plommer shared that the paper’s corrections work is more focused on online than ever before:

Increasingly, in recent years, corrections’ priority has gone to online content, and I see no sensible alternative to this, because the digital database is the Guardian’s most accessible permanent archive, and one with instant global reach. This is where most people will see the original mistake, and where a corrected version stands the greatest chance of being spotted. Having the right version on the web also seems the best way to prevent us and others repeating an error in future stories.

The web is live, and impatient; corrections need to keep up. On our website’s corrections and clarifications page, a selection of online repairs is pulled together and steadily updated so that a rolling log is created throughout the day.

She explained that the paper’s print corrections column has primarily focused on correcting errors made in the print edition. However, that’s about to change (emphasis mine):

Enclosed in its finite space, the newspaper column is a window on to some of our correcting and clarifying. To date, it has mostly confined itself to making amends for errors published in the newspaper, on grounds that it would seem odd to explain web-only mistakes that lovers of the fish-wrap Guardian might never have clapped eyes on. This probably still makes sense, but from tomorrow, the exclusion zone will be breached so the paper column can point – in headline form – to some subjects corrected on guardian.co.uk. Word-for-word headlines should allow an easy web search, for anyone wanting to know more about what was corrected. The aim is to give a better taste of the work being done by many editorial departments to ensure an accurate record – and to make the column better reflect the Guardian’s range from newspaper to multimedia site.

Nice work by the paper to think about how it can spread its corrections between print and online, and help make readers aware of mistakes.

But that wasn’t the only news Plommer had to share:

Meanwhile, a new web page – entitled accuracy and standards – now brings together, under one digital roof, information on the guidelines, system of scrutiny, and other tools through which the Guardian and Observer aim to reinforce the business of getting things right, and being answerable when this objective is not met. It is found at www.guardian.co.uk/info/page/accuracy-and-standards.

The above link is a landing page that aggregates corrections from the Guardian and Observer, the readers editors columns, and lots of information about standards and policies at the papers.

The tweak to the printed corrections column and this new landing page demonstrate that the Guardian and Observer continue to think about how they can evolve and publicize their corrections. That’s one reason why they continue to be leaders in this field.

Oh, and I course need to mention that a correction was later added to Plommer’s column:

This article was amended on 28 October 2011. The original said that “the Guardian’s website was born in 1999″. This has been corrected because several Guardian sites were launched between 1995-1998. It was in 1999 that a unified and augmented site, Guardian Unlimited, was launched. The column above has been corrected, and an item will appear in the Corrections and clarifications column.

Note that the online correction specifies when it was added to the content. I see too few news organizations adding this to their online corrections.

The Mail on Sunday strives for complete accuracy in every article, but it is inevitable when working against deadlines that mistakes are sometimes made and misunderstandings arise.
We have always been ready to publish corrections, and from today we will ensure this happens quickly and prominently with the launch of this new column.

As the Guardian noted in its report, “No other tabloid runs a regular column of that kind, although many upmarket papers do so.” (The Guardian has one of the word’s most famous corrections columns.)

The announcement of the new corrections column came from editor in chief Paul Dacre last week. It’s a positive move, especially from a paper that often has to issue major corrections and apologies, and has in the past made them difficult to locate.

The next step for the Mail and its sibling papers is to create dedicated online corrections pages that are accessible via a link from the homepage. As of today, it’s still difficult to locate the Mail’s online corrections, particularly thanks to the paper’s extremely long and content-packed webpages.

An even more pressing issue, however, is the fact that the Mail isn’t actually correcting its online articles. As noted by FullFact.org and Tablid Tabloid Watch, all of the articles corrected in the Sunday paper remain uncorrected online. Online corrections are pretty useless if you don’t actually add them to the offending copy/link…

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!

Myers reports that the news organization will soon implement a new way to handle retractions, an update that better adapts to the digital world.

.. A notice will be posted atop the offending article at the same time that the advisory goes out, and an editor will strike through the incorrect portion of the article. If another article is pending, as in this case, the notice will say so.

As I pointed out in my report about the Johnston column, Reuters simply killed the column and in doing so it turned the original URL into an error page.

In essence, Reuters will move from an approach that made sense when it was a wire service with no online publishing home to one commonly used by bloggers to update and correct their posts.

“The correction/kill policy that is followed at Reuters is long-established by the wire service,” Ledbetter said. “There isn’t a procedure for taking down something that is wrong because for the vast majority of Reuters’ existence, there was nothing to take down.”

Under the new policy, the erroneous post would remain online even after Reuters published a follow-up.

“I think it stands as a transparent record of what occurred,” he said. “I think to take it down – while I can see some argument for that – it’s not being fully transparent with our readers about the process, and it could be subject to abuse.”

One advantage of this approach is that it retains reader comments, which disappear when a post is deleted.

This new approach makes a lot of sense. Now if Reuters would also make the online corrections page work on its sparkly new website, it would be making some good strides.

… It has been my experience, after reporting from many locations around the world and working with others who do, that errors are more likely when covering foreign events and topics, as misspellings, misunderstandings, inaccurate references to professional titles, and poor or incorrect translations from other languages can lead to mistakes.

– via Apparently, Global News Orgs Don’t Commit Online Errors, a great column by my CJR colleague Justin Marrtin.

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.

We’re entering one of those fertile, exciting periods when the fundamentals of publishing are, yet again, undergoing massive revisions thanks to new technology.

– via Will Murdoch’s The Daily Evolve Corrections for a New Platform?, a recent post from MediaBugs’ Scott Rosenberg on MediaShift Idea Lab, I site I help edit.

The committee that oversees the Editors' Code of Practice to which the British press subscribes voluntarily has changed the rules governing the publication of corrections.

From next year the prominence of corrections stemming from a complaint to the Press Complaints Commission, which oversees press self-regulation in the UK, will have to be agreed with the press watchdog prior to publication.

The Editors' Code Committee has changed Clause 1ii of the code, which covers accuracy, to now say: “A significant inaccuracy, misleading statement or distortion once recognised must be corrected, promptly and with due prominence, and – where appropriate – an apology published.

“In cases involving the commission, prominence should be agreed with the PCC in advance.”

Ian Beales, Code Committee secretary, said: "This amendment is designed to help kill the myth that newspapers and magazines routinely bury corrections.

“Research conducted by the PCC has shown this to be untrue – nearly 85 per cent of PCC-negotiated corrections and apologies appear no further back than the original transgression, or in a designated corrections column.

"It is also the case that most editors already consult with the PCC informally on the position of corrections.

“This change formalises that position in the hope that it will remove one, at least, of the misconceptions about the PCC and the self-regulatory system."

It will be interesting to see how the PCC will link placement to the nature of the error. Will front page errors require front page corrections? I also wonder why the change is necessary if, as Beals said, printed corrections are already receiving fair placement.

The new policy takes effect January 1.

… Most newspapers have longstanding policies on how errors are corrected in print, but if you ask editors and reporters about online corrections in their own newsrooms, you likely will get as many answers as people you ask.

… Mobile sites and apps for smartphones and devices like the iPad bring yet another dimension to the issue of corrections. Are developers, online content people and editors asking "What about corrections" when developing the specs and feature sets for those? I think not.

– Via Fixing errors online needs some correcting at news organizations. This post is in response to my recent CJR column about the lack of thought and effort news organizations are putting into online corrections. Jack’s last point is particularly important: news orgs haven’t even figured out what to do with website corrections; there’s no reason to believe they’ll put any thought into handling corrections via apps. Not good…

Content Analysis

Author – Is someone identified as the author of the site or article? Google them, look for a personal website. If their byline links to an archive of previous work, read through it to see if they cover the topic regularly. If they’re an academic, Howard Rheingold has a tip to check their credibility: “use the scholarly productivity index that derives a score from the scholar’s publications, citations by other scholars, grants, honors, and awards. If you want to get even more serious, download a free copy of Publish or Perish software, which analyzes scientific citations from Google Scholar according to multiple criteria. Again, don’t trust just one source. Triangulate.”

Content – Is the article citing facts and are they accurate? Or do you read it and realize there isn’t fact, statistic, quote or citation mentioned. Does it rely on generalized personal narratives that lack specificity?

Copy – Rosenberg advocates checking to see if the content is original. “Grab a chunk of text (a sentence or so), put it in quotes, and plug it into Google to see whether there are multiple versions of the text you’re reading.”

Links – Does it link out to reputable sources? Is it littered with keyword ads, or have no links at all? Also, critically, see who is linking to the site or page in question. Here’s what Rosenberg wrote about this: “If your hunt for links in turns up a ton of references from dubious sites, your article may be part of a Google-gaming effort. If you see lots of inbound links from sites that seem reputable to you, that’s a better sign. “

Comments/Tweets/Likes – Are people interacting with the content? Be sure to check if all of the comments are spam, and also to see if tweets come from active users rather than bots. Finally is anyone Like-ing the content or the site in general? One easy way to look for social media chatter for a given link is to install the ConvoTrack bookmarklet. Run it while on the site in question and it will show if people have shared the link on Twitter, FriendFeed, Digg, Reddit, HackerNews and some of the major blogging platforms.

Today's column looks at the aftermath of the Slate/Politico showdown from earlier in the summer, and how that caused Slate to change its correction policy. An excerpt:

When we spoke this week, [Slate editor David] Plotz called the dustup with Politico “embarrassing” and said it happened “because of a mistake by me, which was not to really realize and then not to notice or call attention to the fact that we were tweaking them for things we were doing ourselves.”

After the incident, Plotz decided the Slate organization needed to reexamine the “twenty-four hour window” rule that saw it scrubbing errors out of articles. He convened an e-mail discussion between the site’s top editors and initially suggested they shorten the window.

“We went back and forth on shortening the window and actually it was Jack [Shafer, the site’s editor at large and Press Box columnist] who weighed in and said, ‘Look this just doesn’t make any sense. The window is the problem and we have to be really transparent about this …’,” Plotz said.

The group quickly realized Shafer was right. At this point, I’ll disclose that Shafer and I spoke earlier this month in order to discuss Slate’s corrections policy and the issue of the window. I have no idea if this was before or after he made his suggestion to the group, but I can say that he and I are in agreement: the window needed to be closed.

So were these changes updates or corrections? And, in either case, why didn’t the site acknowledge the changing nature of its articles? The New York Times, for example, requires its continuous news desk to note when articles have been updated. And adding corrections for factual errors is a broad standard in the industry. So is Politico inserting new information without noting it, or is it also deleting errors without adding in corrections? [Be sure to scroll down to read my latest updates to this piece.]

Slate certainly made it seem as though scrubbing is an issue at the website. And then, close to three hours after the Slate piece went live, a correction was added at the top of the piece:

Correction, 6:45 p.m.: Shortly after we published this piece about Politico’s unacknowledged corrections, Politico contacted us to tell us that it had acknowledged almost all these corrections. (Politico appended the corrections after we asked about them but before we published our story.) To see our corrections of our mistakes about their corrections—still with us?—click here.

That’s a notable development considering the response that Politico’s top editor had when first contacted by Slate:

Editor-in-Chief John Harris said: “I’m not sure there needs to be a black-and-white policy.” But Politico “ought to think about” keeping a running list of corrections, and its unwritten policy. Politico’s unwritten policy is to append notices to articles “where something substantive was not correct,” Harris said. In many of the stories cited in the sidebar, Harris acknowledged, “there was no thought or judgment applied.

The fact that most of the problematic Politico articles noted by Slate were corrected after Politico was notified of the changes suggests that the site could indeed benefit from a “black-and-white-policy.” Honestly, though, shouldn’t a good policy be black-and-white, meaning clear? Isn’t that kind of the idea behind creating a policy?

It seems Politico has not been operating with a proper corrections policy, let alone decent procedures. (An unwritten policy is usually no policy at all.) As a result, articles were being changed, facts were being corrected — and nobody was telling readers. Bottom line: scrubbing is unethical and unprofessional. It breaks the contract of correction that has existed between journalists and the public for hundreds of years. If that’s too old school for you, Politico also violated Jeff Jarvis’ newer maxim of “publish and correct.”

Politico needs to get its corrections act together. Scott Rosenberg has also weighed in with a call for “versioning” — a similar approach to how software is updated — as a way to deal with the evolution of news stories on the web. It’s a good suggestion, and I know of one person who is already using this as a way to communicate updates and corrections. As a starting point, Politico needs to draft a clear policy, create the necessary related procedures, educate staff, and implement any necessary technical measures to enable corrections.

Let’s hope this Slate study provides Politico with the motivation to get it done with the kind of speed that the site is so famous for.

UPDATE July 21, 12:00 pm: It’s important to note that the Slate story has resulted in a rather suprising admission by the publication. This was added as an update to its story:

As Politico’s editors also pointed out to us, many of the changes Politico made to its stories would have been permitted under Slate’s own corrections policy. Under Slate’s policy, we do not notify readers about minor corrections that we ourselves catch within 24 hours of publication.

Well, that’s interesting. Hypocritical, too… This is the first I’ve heard of the Slate policy and they should change it. Twenty-four hours is a long time online, and the proper policy is to fix errors and correct them, regardless of time frame. I’m curious as to what is and isn’t a “minor correction” in Slate’s definition, but it seems to me that a factual error or typo that changes the meaning of a sentence should be noted whether you catch it 10 minutes or 10 hours after publication. The idea is to be transparent with your readers. This policy may help editors save time, and writers save face, but it does nothing for readers.

Slate does a lot of things well with their corrections, and I often cite them as an example of a publication that has a good online corrections format. But this is a bad policy and they should change it. There’s a certain sense of karmic justice, though, that Slate’s takedown of Politico has had the unintended effect of shining a light on a misguided Slate policy.

UPDATE July 21, 12:25 pm: Here’s a statement from Tim Grieve, deputy managing editor of Politico:

Slate has now posted no fewer than six separate corrections to its story and the accompanying sidebar, and we’re still pushing them to correct at least one more significant factual error.

They’ve admitted they were wrong about the number of stories on which we posted corrections; they’ve admitted they were wrong about the substance of one of those corrections; they’ve admitted they were wrong on a number of specific stories where they said we hadn’t posted corrections when we had; and they’ve admitted that they misquoted a WRITTEN statement I provided to them. They also mischaracterized comments made by POLITICO’s editor-in-chief; we’re still trying to get them to own up to that one.

Their story now begins with a 50-word correction and ends with a 125-word correction, and the sidebar is studded with doozies like this: “Correction, July 20, 2010: A previous version of this article incorrectly reported that Politico hadn’t yet acknowledged its correction.”

We all make mistakes, of course, and Slate caught us making some: We’ve managed to correct some stories on our site without posting the formal correction notices that our policy requires. We’ve gone back and added correction notices to all of the factual corrections Slate identified, and we’re taking steps to prevent this problem from recurring.

In the meantime, Slate now admits that it doesn’t even try to live up to the standard it would hold us to. I’m not sure if that’s irony or hypocrisy, but either way it probably should have given Slate pause before going on the attack.

Here’s an excerpt of a recent blog post from Rosenberg:

One of the early field results of the MediaBugs experiment is a simple one. It turns out that, in the case of many news organizations, including some pretty prominent ones, just figuring out how to tell the newsroom that there’s a problem requires persistence and stamina.

Consider this anonymous error report we received at MediaBugs a few days ago. It said that the Wall Street Journal, in a recent book review, had misspelled the name of the author being reviewed. The book is Mac McClelland’s For Us Surrender Is Out of the Question. The Journal spelled her name “McLelland.” (The publisher’s page listing the book, which I’ll take as an authoritative source, spells it with the extra “c.”) …

Rosenberg then spends several paragraphs detailing the process he undertook to pursue this correction. Imagine how many people would have abandoned their quest for correction after the first few minutes of trying. (Most of them, if you ask me.) Writes Rosenberg:

I went to these lengths because, right now, this is my work. But we shouldn’t have any illusions about normal members of the public. They won’t jump through these hoops. They will conclude — rightly or wrongly but very understandably, either way — that the newsroom doesn’t actually care about hearing about its mistakes.

Now here’s the beginning of a new post from Follman:

Recently a MediaBugs user reported that an Associated Press story had misidentified the “Seinfeld” character George Costanza as Jerry’s “neighbor” on the show. Eventually the AP’s west coast entertainment editor, Steve Loeper, responded to an inquiry about the matter, and the AP subsequently decided to publish a correction.

It was a positive outcome, but here’s the rub: Getting to it involved no less than contacting five different people, sending eight emails and making three phone calls — and it took more than three weeks to get a result.

I suspect the MediaBugs team will continue to encounter this kind of disorganization and lack of accountability. As Rosenberg wrote, “If we want to understand why people don’t trust the media, this might be a very good place to start.”

It also helps explain why the number of published corrections is tiny when compared with the actual number of errors.

… The Post has reversed its embarrassing inattention to correction requests. Since my initial column, section editors have received regular reminders about correction requests that have been pending for more than 14 days. Where a backlog of several hundred neglected requests once existed, the number now is only four. The database shows a handful of other pending requests that are being addressed and corrections likely will appear soon.

Following publication of my first column on March 22 of last year, Assistant Managing Editor Peter Perl successfully pushed the newsroom to whittle down the huge backlog of requests. The task of riding herd on corrections was subsequently passed to Senior Editor Milton Coleman, who started another push several months ago …

Coleman credited two veteran Post copy editors, Bill Walsh and Martha Murdock, with improving corrections to make them “more fulsome, clearer and more transparent.” That’s helpful to readers, who in the past often complained about Post corrections that made it impossible to know the original error.

Where the paper still needs to improve:

Readers periodically complain that their requests for corrections, typically e-mailed to corrections@washpost.com, are never acknowledged. And sometimes it takes too long to run a correction that should have appeared within days.