From Ben Smith:

Though I've spent the last several years at major newspapers - the New York Observer and the New York Daily News most recently - I've done much of my reporting on blogs, and have developed an instinct to let my readers know whatever I know, as soon as I know it. The medium typically allows you to refine and update a story as it changes -- including saying, "Well, my original source had it wrong."

But the scale of this story was simply too big to report that way, to share information with high but imperfect confidence -- and without making that level of confidence crystal clear. I should have waited for a second source, or hedged the item much more fully. Or simply waited for the news conference like everybody else.

As Politico editor John Harris emailed me this afternoon, "I believe a blog item is different than a story--not in standards of accuracy or fairness--but in the ability to report and reveal a breaking story in real time: You write what you know when you know it.

"BUT, and here's where you went wrong and we let you go wrong, you can not write more than you know.

"You had a source, who has been reliable in the past and who had reason to know what was happening in this case, telling you something relevant to a big story. That reporting was worth sharing with readers, but ONLY with the caution that our information was still fragmentary, not definitive. We should not have made a flat predictive assertion about what Edwards was going to do.

"The lesson, which we both know but re-learned, was the importance of precision," he concluded. "If we had put as much emphasis on what you still had not confirmed, and that there was still doubt about Edwards's intentions, we could have shared your reporting with readers and not misled them."

As is probably to be expected, some harsh words in the comments section of both the above piece and this one (1000 between the two).

Smith also links to some reaction.