How to break an airline - blog on out of date data.

If you haven't seen this, read it - its important. Here is Forbes' take:

At 10:53 a.m. Eastern Daylight Time Monday, a reporter at a Florida-based investor information service did a routine Google search and found a story saying United Airlines was declaring bankruptcy.

It seemed credible: It was from the Chicago Tribune and it was posted on the South Florida Sun-Sentinel, a sister paper also owned by Tribune. United Airlines parent company UAL (nasdaq: UAUA - news - people ), based in Chicago, has a long history of troubled finances.

He didn't call the company, but he did write up a headline, "United Airlines: Files for Ch. 11, to cut costs by 20%" and published it, not only to subscribers of the Income Securities Advisor, but also through Bloomberg terminals, to which ISA, in the parlance of the Internet news industry, aggregates its content.

Its impact on United's stock was swift and terrible. In the span of 10 minutes, 24 million shares changed hands. The stock, trading at $12.45, crashed to $3, according to Nasdaq. So severe was the market's response that Nasdaq halted trading from 11:06 a.m. to 12:30 p.m. The exchange says the rest of the day's trades will stand.

The problem, as most in the financial news industry and traders around the world now know, was that the story was six years old. The Chicago Tribune story ran Dec. 10, 2002, and appears to have been republished or resurfaced accidentally Sunday afternoon on the South Florida Sun-Sentinel's site.

This is not that far off the way the blogosphere works, especially the increasing evidence that the larger blogs are more interested in publishing early rather than accurately, and/or getting traffic by being as polarised as possible - again decreasing accuracy (this research shows that having political blogging being balanced and factual is not a stable state - and generally does not drive traffic - and we hypothesize from observation that most other fields are the same - the recent Google Chromegasm being an example in the tech field)

This may be accidental, but it doesn't take much analysis to realise that all those skills used to pimp and place "positive" PR can as easily be used to plant false and damaging data. Once aggregated on a large search engine, it becomes a truth all of its own. This is magnified when auto-trading algorithms started to trade on the news and then others started to sell based on the dropping price from the earlier trades.