When you have a son who is smart enough to get a PhD on his own and who then goes out and receives acclaim in national magazines you gotta be proud of him. That's the deal with this post...my son, John, and his wiley experiments with eBayers and their buying habits....Read on, he's the John Morgan mentioned about 2/3ds of the way down the story. Read the link to see the whole story.
ArjayHerding the Mob
John and Nina Swanson run a business selling vintage postcards on eBay. To keep customers happy, the Swansons reply to buyers promptly and ship on time. This policy is reflected in their eBay feedback score — a rating based on responses to prior transactions. Positive comments are scored as one point. Neutral and negative remarks are recorded as zero and negative one, respectively. The Swansons have a score of over 2,000.
Six years ago, University of Michigan information studies professor Paul Resnick asked the couple to participate in an experiment. Resnick wanted the Swansons to continue selling postcards through their established profile, but also to offer the same goods and services through seven fake identities. Initially these bogus profiles would have no reputation; later they would be given negative scores. The Swansons agreed.
After 470 auctions, Resnick found that the Swansons’ main account, with its high customer rating, earned an average of 8.1 percent more per transaction than the fakes. It was the first hard proof that a feedback score — a number generated by a collection of unrelated people — carries quantifiable real-world value. “What we’re seeing here is a new kind of trust,” Resnick says. “It’s a kind of impersonal trust geared to situations with lots of interactions among strangers.”
In other words, the crowd matters. Today we harness the masses for everything from choosing the next pop star on American Idol to perfecting open source software and assembling Wikipedia articles. But perhaps the most widespread and vital uses for group input online are in scoring systems. In addition to eBay feedback, these are the customer ratings that Amazon.com and Yahoo Shopping post with product reviews. They’re the feedback scores that Netflix tallies to help subscribers decide which movies to order. And they’re the up-or-down votes that sites like Digg and Reddit (part of the Wired Media Group, which also includes WIRED magazine) rely on to determine which stories to feed Web surfers.
But as rating systems have become more popular — and, as Resnick shows, valuable — there has been what some would say is a predictable response: the emergence of scammers, spammers, and thieves bent on manipulating the mob. Call it crowdhacking.
In some cases, crowdhackers are looking to boost sales or increase traffic to their Web sites. In other instances, they’re simply ripping off unsuspecting consumers. Either way, the more we base decisions on the wisdom of crowds, the greater the incentive to cheat.
The feedback system on eBay was the first widely used community-scoring program. Launched in 1996, it was a way for people to feel comfortable buying things from strangers. Under eBay’s rules, only people involved in a given transaction can rate it, and eBay won’t remove remarks once they’re posted.
This scheme quickly became the gold standard, serving as a model for user-rating systems everywhere from Amazon to Yelp. But people soon realized that the setup was easy to manipulate.
Cheats on eBay typically work like this: A scammer builds up a positive profile by selling hundreds of low-end items, then uses that high score to burn customers on big-ticket sales. That’s what police say an Arizona woman named Nancy Dreksler did in 2003. According to police reports, once Dreksler had acquired positive feedback by peddling inexpensive CDs and DVDs, she sold over $100,000 worth of nonexistent items and fled with the money, leaving more than 500 buyers empty-handed. Arizona authorities say they may yet file charges; meantime, Dreksler has pled guilty to theft and securities fraud charges in a separate Nevada case.
John Morgan, a professor at UC Berkeley’s Haas School of Business, says reputation gaming is surprisingly common on eBay. Morgan recently published a study in which he found more than 6,000 examples of buyers and sellers engaging in transactions solely to boost one another’s scores. These auctions frequently had titles like “100+ Feedback” and a price of 1 cent. Often, the item for trade was a booklet explaining how to increase feedback by reselling that same booklet.
“We saw a number of sellers who used sham transactions to build reputation, laid low for a period of time, and then reentered high-value markets as apparently ‘reputable’ sellers,” Morgan says.
eBay says it constantly hunts for cheaters. According to spokesperson Catherine England, the company uses sophisticated fraud-detection tools to spot suspicious activities and “individuals who may be attempting to inflate their feedback.” She declines to identify these tools but concedes that they are “not 100 percent perfect.”
Other commerce sites have even fewer controls. Yahoo Shopping, for instance, lets anyone post a review, making it easy for merchants to boost their ratings by submitting multiple reviews under false names.
A glaring example of this was disclosed last year, when then VP of Yahoo Shopping, Rob Solomon, admitted to Forbes that the company’s merchant-rating system had been “rigged” by a Brooklyn-based company called PriceRitePhoto. Somehow this shop had managed to get stellar Yahoo ratings, despite many negative reports from disgruntled customers. Blogger Thomas Hawk wrote a post detailing how he’d been threatened by PriceRitePhoto’s owner after writing negative comments about his experiences ordering a camera. Yahoo finally banned the store, but it returned within months under the name Barclays-Photo, according to the Better Business Bureau. After several more complaints, Yahoo removed BarclaysPhoto from its listings. As of late January, it continues to operate (with suspiciously excellent ratings) on eBay.
Over the past few years, crowd scoring systems have made their way to news and article aggregators. Instead of recommending products or services, these sites solicit community rankings to help steer readers to interesting online stories and postings.
The biggest and best known of these is Digg. Members submit articles, along with a short description and link, to the Digg system. Other members can then look through these articles and choose either to “digg” or “bury” the stories. Articles with the most diggs make it onto the site’s widely read front page.
Page 1 of 2 next »
No comments:
Post a Comment