Many thanks to Ashley Huston in the Dow Jones PR department for chasing down answers to the questions I posed last week when I asked whether the WSJ was rewriting old articles.
In the case of the article about John Thain specifically, the news broke on the 22nd, and then the story evolved over the course of the day, eventually ending up on the front page of the physical Wall Street Journal on Friday the 23rd. As the story was reported and edited over the course of the 22nd, the story on the website got longer and more detailed; eventually, it ended up with exactly the same copy as appeared in the newspaper. (Yes, I’ve checked the story on the website against a PDF of the story in the paper: they are indeed the same.)
In general, Huston told me, "the version that runs in the paper is considered the final version. A new story may begin the next day as a story develops."
How come, then, the story on the website is dated January 26, rather than January 22 or 23? Says Huston: "there was a technical glitch as
we managed the layout of pages days after the story ran, and a new date
was accidentally added." Essentially, it was a simple error: maybe someone edited the interactive stock chart, or something like that, which automatically led to the date being changed.
The WSJ is, still, first and foremost, a daily paper; it runs on a 24-hour news cycle, although Huston did say that "it’s an evolving discussion as we publish more and more online
and operate as a 24/7 operation". As such, I can see why it might update a story over the course of one day and finally pin it down for perpetuity only when the newspaper version appears. Eventually, however, it would be great to be able to see some kind of Wikipedia-style history of changes.
Blogs are different: because they don’t conform to a 24-hour news cycle, I think they should always make it very clear when they’ve been changed or updated after their original posting time. And I would hope and expect that newspapers’ blog entries would follow the same rule, even when they end up appearing in the paper.
In a similar case which has been brought to my attention, an LA Times article headlined "Legislative Inquiry Finds Palin Abused Her Power" (you can find it linked to here, among other places) now redirects to a significantly different story, with the same date, headlined "Palin ethics lapse cited". But the LA Times does still host the original version of the story on its website, in the archives. I suspect, although I haven’t checked, that the same thing is going on here: the final version of the developing story is the version which appeared in the physical paper.
I’m a blogger who links to stories as they’re developing, and so it’s annoying to me when I think I’m linking to one thing and I end up linking to something else entirely — as happened with the John Thain story. And when the newspaper is the one breaking the news, I think there’s a strong case to be made that the first important news story — the one which broke the news and caught everybody’s attention — should remain on the web in perpetuity, rather than being massaged and lengthened and edited as people respond to the story.
I hope that the WSJ appreciates the importance of maintaining a historical record of what it said and when, and that it will treat its long front-page stories as separate beasts from its breaking news alerts, rather than considering them to be no more than evolved versions of exactly the same story. That way, people researching what happened when will be able to see not only the following day’s reportage, but also the stories which really moved the market.