Back in early 2006 I passably noticed the news about Congressional staffers white-washing Wikipedia biographies.
This was wonderfully karmic and I thought to myself, "This would be hilarious to do on a massive scale." But schoolwork comes first, and I sat on the idea until I had been accepted to grad school in 2007. Then in summer 2007 I set to work on the WikiScanner.
Little known at the time, I knew you could download the full Wikipedia database. Additionally, Dan Kaminsky had told me about the ip2location database which had a thorough mapping of IP#'s to owning organizations (also obscure at the time). I dusted-off my PHP skills and wah-lah---the WikiScanner, generator of mild yet delightfully karmic PR disasters. It felt delightful to bring sleazy wiki-white-washing into the court of public opinion.
I expected WikiScanner to make the standard Wikipedia newsreel as well as a mention in Wired, but I did not expect the massive interest from the print media. To my surprise, even normal people cared about the integrity of Wikipedia.
Some of my favorite WikiScanner fruit was:
- Imperial Household Agency no longer allowed to edit Wikipedia
- Minnesota Republican National Committee spoils Harry Potter
- Diebold repeatedly white-washes their own page on voting machine security
- A New York Times frontpage story discussing their own ejaculation on George Bush's page and Condoleezza Rice being a concert penis I gained immense respect for the NYT for this. They were the only large media company who discussed their own Wikipedia transgressions. The BBC couldn't bring themselves to do it.
- The ACLU elucidates the Pope's pederasty responsibilities
- WikiScanner was repurposed for social science research!
WikiScanner was expensive to host (several thousand USD per month) and I eventually took it down. I'd like to bring it back some day. Maybe I will.
Link: Old WikiScanner FAQ
In summer of 2008, now a graduate student at Caltech, I supervised undergraduates Rishi Chandy and Daniel Erenrich on creating a modern, new and improved WikiScanner. We developed a suite of Wikipedia-related tools known collectively as "WikiWatcher" which included: WikiScanner2 (Daniel), Wikiganda (Rishi), Poor Man's Checkuser, and BeaverScope. I launched these at the Hackers on Planet Earth (HOPE) conference.
Link: Presentation at HOPE