Top Stories

Two clicks sparked a chain of events that led to a sell-off of United Airlines stock Monday amid a report that the company had declared bankruptcy.

Our original article described how the lack of a publication date on a 2002 story played a key role in Google's crawler picking up the story as new. The short version is that the old story appeared in a "most viewed" area of the South Florida Sun-Sentinel's Web site over the weekend, which caused a Google search agent to index the story in Google News. Google assigned the incorrect date of Sept. 6 to the undated story.

Late Wednesday afternoon, Tribune Co. revealed what led to the article appearing in the "most viewed" area: Shortly after 1 a.m. EDT Sunday, someone viewed the 2002 story. Because site traffic was so low at the time, the company said, that single visit probably caused the story to be included on the list of the five most viewed business stories on SunSentinel.com.

Then, 36 minutes later, according to Tribune Co., someone viewing a story about airline policies on canceled flights clicked on the "most viewed" link to the United Airlines article.

Less than a minute later, Google's search agent visited the site and crawled the United Airlines story. "This time," Tribune Co. said in the news release, "despite the fact that the URL to the story hadn't changed, despite the fact that Googlebot had seen this story previously, it was apparently treated as though it was breaking news."

Once the story was indexed in Google News, according to Tribune Co., visits to the story increased Sunday. Most of those visits were driven by Google, Tribune Co. said. A reporter for an investor's newsletter found the story through a Google search Monday, which triggered the market pandemonium.

		RELATED
		Google 101 article on how the search engine crawls and indexes the Web

Finger-pointing between Google and Tribune Co. continued Thursday. In its news release, Tribune Co. said that it has identified problems with Google's crawler, Googlebot, not being able to differentiate between breaking news and frequently viewed stories on its Web sites. Tribune Co. said it has asked Google to stop using Googlebot to crawl its sites for inclusion in Google News.

"We asked Google to stop using Googlebot and use site maps instead," said Gary Weitman, Tribune Co's senior vice president for corporate relations, in an e-mail. "We're happy to have Google acquire information from our sites, we just asked them to do it more accurately because we were aware of problems with Googlebot."