Here is a guest post from Amy Fry, a San Diego-based librarian with whom I’ve done some research on aggregated databases. She was struck by the way a sloppy mistake in handling information led to a plunge in a company’s stock prices – and what the implications might be for information literacy. If you’re low on energy and thinking a cup of strong coffee might wake you up – hang on; this post might just do the trick.
On September 8, a reporter for Income Securities Advisors, using Google, found a 2002 article from the South Florida Sun-Sentinel about United Airlinesâ€™ bankruptcy. The article was undated in the paperâ€™s archive, but used a site header displaying the current date, so the Google News crawler, indexing the site Saturday night, applied the date of September 6, 2008 to the story. Mistakenly identifying the article as current, the reporter summarized it and sent it to her editor, who posted it to the ISA wire service. Aggregated by Bloomberg (though independent from Bloomberg News), the headline was seen by Wall Street traders, and even though the company caught the mistake and removed the headline within 13 minutes (and Bloomberg itself posted a correction), a trading frenzy had already caught hold causing United to lost 75% of its stock value in under an hour.
This story contains a powerful lesson about information literacy.
One: Proper metadata is important.
Metadata experts have been trying for years to promote universal standards for describing and applying information about content objects, online and elsewhere, and this is why. Why was this article undated when other articles from the same news archive were dated, and how can a header date be mistaken for the date of unaffiliated content? The answer is: improper application and use of metadata. One reason we teach students to use library resources is that we believe that properly indexed information, with standard subject headings and descriptive metadata that is uniformly formatted and properly mapped, aids the user in finding and evaluating information. As this story shows, such indexing can also help information seekers avoid costly mistakes. The problem of universal metadata standards is complicated, but our hard work as information scientists is not wasted in solving it.
Two: There is no substitute for critical thinking about sources.
The reporter, and her editor, did not think critically about where her information was coming from and why it might require a second glance. Even if she didnâ€™t have the background to already know that United had declared and emerged from bankruptcy within the last 10 years, proper critical thinking about sources should have caused her to ask why this story was being fed to her first through Google News from a south Florida Tribune-affiliate instead of the Wall Street Journal or another primary information source of financial news. We teach students to examine a variety of points to determine the authority of an information source, like an identifiable author, author affiliation, publisher and publisher affiliation, traceable references, and external peer review. All of these can help them ascertain if sources they find are reliable, even if they do not have extensive prior exposure to the subject of their research question. This story proves that there are no shortcuts to determining the authority of sources, and no substitute for critical thinking.
Three: Sometimes aggregators are misleading.
Aggregators play a valuable, but complicated, role in information provision. Bloomberg not only provides information to its subscribers â€“ it also aggregates information from other services and packages this information with its own. Operating under the â€œmore is moreâ€ and â€œbigger is betterâ€ philosophy has become commonplace in the world of information aggregation, and librarians tend to agree (see Fister, Gilbert and Fry in the July 2008 issue of portal). But, as this story shows, it comes with certain pitfalls. Aggregators have neither the means nor the desire to vet every item of information they provide in their products, but the distinction between their role as aggregator and their role as authoritative information provider is blurred. Often their own status lends authority to the information they package â€“ touted as unintended when that information proves to be faulty. As this story demonstrates, more oversight of aggregators and by aggregators, and a demand of quality over quantity, should be a priority for librarians, especially in this age of information overload.
Four: Google is more powerful than we even realized.
If any one of you has been underestimating the role of Google in the information food chain, STOP. Google has enormous power to direct culture through the control of information. While the company sticks to its mantra of â€œDonâ€™t Be Evil,â€ this story proves that high-stakes real-world results can be achieved in moments through Google without Googleâ€™s knowledge or intervention and even without intentional sabotage. Google has changed the way we find, use and even produce information â€“ but with great power comes even greater responsibility. Librarians have raised important points about the ethical dimensions of private information ownership in conjunction with the Google Books digitization project. We have warned students to be careful when using Google as a research tool. A private company is not required to act in the public interest. Academic librarians, as educators, are. As more and more information is accessed through and archived by private companies (for example, despite its content, EEBO is still a proprietary resource), librarians must take on greater responsibilities as watchdogs for the public interest. Even if our roles are changing, our mission must not.