Identifying hoaxes

I’ve always felt that it’s better in principle to help people to identify potential hoaxes for themselves and check them out accordingly, rather than take the pure dictionary approach (i.e simply list each hoax as it comes along). In practice, a well-maintained ‘dictionary’ site is a treasure. Even people who can detect a hoax (or semi-hoax) at 50 paces sometimes need information that goes beyond their own scepticism and intuition.

For instance, in order to convince others, who can be very reluctant to let go of a pet meme. After all, it’s not very nice to be told, however politely and considerately, that you’ve fallen for some mean-spirited hoaxer’s attempt to boost his own ego at your expense.

Adam Pash’s post at Lifehacker on How to Identify and Avoid Spreading Misinformation, Myths, and Urban Legends on the Internet does a good job of providing the aspiring hoaxbuster with some resources for checking a suspected hoax. He uses as an example a heavily retweeted quotation incorrectly attributed to Martin Luther King Jr. after a Facebook update regarding the death of Osama Bin Laden was misinterpreted or misquoted. The list of attributes to a message that ought to trip an anti-hoaxer’s sensor is longer and more varied than that single example suggests – maybe I should get back to that – but  the list of checking resources is worth summarizing here (but I also recommend that you check out the Lifehacker blog).

• Unsurprisingly, he recommends Snopes (not just for its encyclopaedic listings and well-researched commentary, but also for the up-to-date info and discussion on its forums).
• He also suggests BreakTheChain.org and TruthOrFiction.com and points to a longer list of resources by Tim Malone.
• He draws attention to the use of Google’s date range filter.
• He also draws attention to the difficulty of confirming a scam when the scammer floods the Internet with fake recommendations. (To all intents and purposes, a manifestation of Black Hat Search Engine Optimization.) However, similar issues arise with messages that are classified as (semi-)hoaxes rather than as scams. To take a simple example, you’re likelier to get a useful hit if you add the word “hoax” to your search terms straightaway: otherwise, you may get several pages of sites where the hoax is simply uncritically quoted or recycled.
• And finally, one I hadn’t come across before: Dustin Luck’s Debunkadunk custom search engine, which throws your search terms at a range of specialized web sites like Snopes.

David Harley CITP FBCS CISSP
Small Blue-Green World
ESET Senior Research Fellow