Poynter.org works hard to meet the highest standards of journalism excellence, and I learned late Wednesday that we have not consistently met those standards.

A centerpiece of our editorial work has been the Romenesko blog, which invented a form of aggregation that is widely and deservedly respected. It is also imperfect.

Thanks to the sharp eye of Erika Fry, an assistant editor at the Columbia Journalism Review, I now know that Jim Romenesko’s posts exhibit a pattern of incomplete attribution.

Though information sources have always been displayed prominently in Jim’s posts and are always linked at least once (often multiple times), too many of those posts also included the original author’s verbatim language without containing his or her words in quotation marks, as they should have.

Here’s a recent example, with the original author’s verbatim language in bold:

Chicago Tribune says Mayor Emanuel refuses its public records request (Our story | Original story)

The Tribune says Mayor Rahm Emanuel refused its requests for his emails, government cellphone bills and his interoffice communications with top aides, arguing it would be too much work to cross out information the government is allowed to keep private. After lengthy negotiations to narrow its request for two months of these records, the paper was told that almost all of the emails had been deleted. The Tribune notes that Richard M. Daley repeatedly denied similar requests when he was mayor, “but it’s not the practice in major cities across the nation.” The paper reports:

"The [Emanuel] administration provided cellphone records that did not include a single telephone number for either incoming or outgoing calls, making it impossible to discern how the phones might be used to conduct city business. The city said it would be 'extremely burdensome' to determine which numbers were public under the law and which were not.

"Emanuel doesn’t have a city-issued phone and uses an aide’s phone to make city-related calls, [spokeswoman Jenny] Hoyle said. The Tribune requested the records for that phone, among others."

The paper found that the kinds of records it wants from Emanuel are routinely available — in many cases with a phone call or an email request– in Atlanta, Boston, Hartford, Houston, Miami, Milwaukee, Phoenix and Seattle. | Chicago Reader (July 21, 2011): “In his first months in office, Mayor Rahm Emanuel has been sending and resending the message that he wants his administration to be a model of transparency and openness.”

One danger of this practice is that the words may appear to belong to Jim when they in fact belong to another.

This style represents Jim’s deliberate choice to be transparent about the information’s origins while using the source’s own words to represent his or her work. If only for quotation marks, it would be exactly right. Without those quotation marks, it is incomplete and inconsistent with our publishing practices and standards on Poynter.org.

Our guidelines -- which have been published on our site since 2004 -- state:

We credit the authors and creators of the various forms of journalism we publish. We apply appropriate scrutiny to work by staff and contributing writers to prevent plagiarism, intentional or otherwise. We do not intentionally mislead with words or images. We do not deliberately deceive as we gather information.

Our practice is to enclose verbatim language in quotation marks, and to set off longer excerpts in blockquotes. While I have no reason to believe this practice has spread beyond one writer, I will check the work of other contributors to determine for certain whether anyone else has been guilty of the same shortcut.

Jim’s situation is unique in that he is the only author in the site’s recent history to publish posts himself without being edited prior to publication. His editors read behind him after he publishes, and often read the original source material, but none of us have noticed the duplicative language.

Some of you may find this sourcing entirely acceptable and disagree that it is unclear or incomplete. Some of you may find it abhorrent and a journalistic sin.

In an attempt to understand the situation as completely as possible, I consulted colleagues whose judgment I value and learned that even within Poynter there is wide variation in how this practice is viewed.

As I listened to their perspectives, these questions formed in my mind, along with some preliminary answers:

  • How much does intent matter? The format and transparency of Jim’s posts credit the original source and direct people to it. There was no attempt to mask the material’s origins or deliberately pass it off as someone else’s.
  • How much weight does the attribution carry? If you begin a sentence with “Newspaper Y says...” do readers believe the words that follow are the newspaper’s or your own? As a reader, I assume text originated with the writer unless quotation marks or blockquotes indicate otherwise. It is incumbent upon the writing and publishing team to signal the reader as clearly as possible. We did not.
  • Is there a different standard for aggregation? This question may generate the most -- and most important -- discussion. I would argue that proper sourcing and attribution are required across all story forms, though they may be accomplished differently. I do not believe incomplete attribution is inherent to or an appropriate part of aggregation. I coined the term “overaggregation” to refer specifically to instances when too much original material was used in an aggregated post, not as a description of sloppy sourcing.
  • How long has this been going on? Jim says this is how he’s been writing for his 12 years at Poynter and a spot-checking of stories going back to 2005 finds multiple examples of the same practice. For years, those posts were quite short and this practice was less noticeable and perhaps less prevalent. It appears more frequently in recent longer posts; however it will take a thorough review to conclude definitively how extensive this practice has been.
  • How much of our work rests on unspoken understandings? This is a cultural question that concerns me deeply. We spent weeks in 2004 developing explicit publishing guidelines with the understanding and expectation that they would be adopted. How often, how consistently and universally did we articulate our values and standards and confirm that others share them? Not enough. Never enough.

I will continue to wrestle with these questions, as may some of you.

To our knowledge no writer or publication has ever told us their words were being co-opted. That raises some questions of its own. Surely many writers whose words appeared in Jim’s posts have read them there.

In fact, often those writers or their editors are the ones who send us links in hopes we will feature their stories. They are not seeking, nor do they deserve, to have their words used without proper credit. They hope to receive the attention of other journalists who rely on us to point them toward the most interesting journalism issues of the day.

We plan to continue doing that work.

Effective immediately, Jim’s work for Poynter will change in a few important respects. First, it will follow our standards of attribution. Second, it will be edited before it is published. I asked Jim Wednesday night to refrain from publishing while we sorted out this situation, and he has done so. Jim has offered to resign and I refused to accept his resignation. In August, Jim announced his plan to semi-retire at the end of this year.

We are in uncharted territory, marked by uncertainty, which suggests caution. We will continue to evaluate this situation and to be as transparent as possible about what we learn and decide.

You can and should draw your own conclusions.

If you’d like to reach me by phone my office number is 727-553-4336 and my email is jmoos at poynter.org.