Since Virgil Griffith launched Wikipedia Scanner, it’s been open season on organizations whose IP addresses are linked to changes made to entries on the popular DIY encyclopedia. For example…

• PRWeek’s UK edition notes that “PR agencies are flouting Wikipedia rules demanding they do not edit the site. At least six of the PRWeek top ten UK agencies have edited the site in the past year…FD is the biggest offender filing 25 edits, primarily concerning clients Russ DeLeon and Ruth Parasol—founders of the online gambling company PartyGaming.”
• Wired is taking and publishing submissions in a posting titled “Vote on the Most Shameful wikipedia Spin Jobs.”
• Over 750 blog posts have been contributed on the topic, and a Google search produces 342,000-plus results.

Cumulatively, the number of organizations being outed for making changes is staggering. Rarely does a single tool produce such an overwhelming indictment of institutions’ predlicition for spinning facts and history.

But a little perspective is in order.

I’ve been visiting the Scanner over the last few days, spending time in the “Editor’s Picks.” For example, I spent a fair amount of time clicking through edits made from Exxon Mobil’s IP addresses. As of right now, there is a whopping 1,205 edits made by Exxon Mobil. Outrageous, right? Well, no. In fact, only about 20 of the edits seem to have anything to do with entries related to Exxon Mobil. 

What, then, were all these other edits? Here’s a very small sampling:

• Correction of a typo (“choose” instead of “chose”) in an entry about Disneyland’s Autopia ride
• Addition of a paragraph about how refrigeration cycles work in an entry about air conditioners
• Removal of an offensive addition to the biography of country singer/actress Dolly Parton
• Removal of a gratuitous addition (I LOVE YOU JENNY) to a listing on the history of American Football
• A section on “Economy” was added to the listing for the city of Natchitoches, Louisiana, suggesting that the city’s tourism industry needs little promotion.

Are these blatant abuses by Exxon Mobil? Clearly not. These are employees who also happen to be Wikipedia fans; they don’t care whether they’re at home or at work when they make their corrections and additions. But because their entries from work computers get aggregated with all other revisions from the company’s range of IP addresses, they get added to the total. Without these entries, Exxon Mobil probably wouldn’t have attracted the editor’s attention and become an editor’s pick.

A look at Amgen’s results shows a similar pattern. Most changes were made to entires dealing with a band called Caustic, the late Monty Python alum Graham Chapman, the term “mnemonic,” the New Progressive Party of Puerto Rico, the Red Hot Chili Peppers, and the like. Reviews of other companies produced the same kinds of results.

And what about the edits to Exxon Mobil entries, the ones that clearly violate Wikipedia’s policy? While some are egregious (deletion of content about the impact of the Exxon Valdez oil spill, for instance), others simply correct facts. For example, the entry about Mobil 1motor oil originally stated that Mobil 1 was introduced two years after Amsoil “marketed the first API certified synthetic engine oil.” The change reflects the fact that Mobil 1 was, in fact, the first API (American Petroleum Institute) certified synthetic motor oil.

At this point, you also have to wonder how many of the edits to company-specific content were business-motivated—that is, deliberately executed by the PR staff—and how many were individual employees who visited the company’s listing and said, “Hey, that’s not right,” and made the change without understanding the consequences. Companies now need policies that limit employee edits to company listings in Wikipedia.

I’ve been on the record opposing the Wikipedia rule that bars companies and their agents from editing company content. Honest efforts to correct mistatements of fact are prohibited by the rule (such as changing the number of employees from 500,000 to 50,000 because the original author added an extra zero) while unethical companies will simply make their inappropriate changes from non-work computers or use proxy services that mask their identities. Meanwhile, thousands of people who don’t work for the company but do have a biased point of view merrily post entries that obviously were never crafted with objectivity in mind.

But my objection doesn’t matter. The rule is the rule and if companies can’t play by it, they deserve whatever heat they take as a result of being outed courtesy of the Wikipedia Scanner. And, to be fair, most of the companies that have been the subject of news reports since the Scanner opened for business have been caught trying to rewrite history.

Still, it’s important to separate the wheat from the chaff, and not much of an effort has been made so far to draw the distinction between truly unethical manipulation of content, minor factual revisions, changes made unwittingly by front-line employees and changes to non-company content by employees accessing Wikipedia from work computers.