Oops – it looks like Bing got busted using information from Google’s search results to power their own search engine. Of course, they had to deny this as well, because who wants to actually admit that they are ripping off a competitor’s product?
The evidence was pretty damning though, and if you want to know exactly how Google came to this conclusion be sure to check out the link above.
But what happened next was pretty low and scummy – even for Microsoft. While the Google blog was shocked enough about Microsoft’s flat denial in the face of evidence, things would get even crazier by Wednesday. In their latest response to the allegations, Bing spokespeople actually claimed that Google’s investigation amounts to “click fraud and a honeypot attack.” I’m not sure they fully understand what those words mean…
See, it turns out that Microsoft is grabbing a whole lot of information from anyone who has the Bing toolbar installed. While there’s no saying how wide the scope of behavior monitoring goes, it seems pretty clear that this toolbar is designed to look at what results people click on during their Google searches. Since Microsoft has been bribing people to install this piece of spyware with all sorts of incentives designed to reach broad demographics, they’ve been able to collect a whole lot of Google’s data and refine it with the actual choices users have made.
Does that make Bing better than Google? Absolutely not, because now we know how easily its results could be manipulated with a few computers signed up for the toolbar. This is what Bing condemns as click fraud, but frankly they’ve created a system with a series of huge vulnerabilities. Several tool-bar based web organization systems have come and gone, and the fact that a few Google Engineers with IE8 + Bing Bar were able to change the results so quickly shows exactly why none of the previous attempts have lasted the inevitable onslaught of money-hungry SEOs.
Does this make Bing evil, or just plain sloppy?
It should also be stated clearly that Bing isn’t relying on Google 100%. They’ve got their own bots out there following links, and they’ve got their own set of metrics for evaluating which ones are important. For a few search phrases that are important to me, I’ve seen fairly significant differences in how the two search engines sort their data.
This also wouldn’t be the first time a web-based information project used signals from user behavior to build data. Advertising companies like Google and Infolinks do it all of the time, and Google has hinted in the past that user behavior can be tied to search results as well – at the individual and aggregate level.
So I can’t go out and say that what Bing is doing is inherently evil. They’re working with the information they can get their hands on, and they’re taking a slightly more aggressive approach than some of those who innovated such behavioral measures.
Where Bing screwed up was by mixing up their response! If you’re actually doing something, it isn’t a very good idea to deny it as a knee-jerk reaction. When your hand is in the cookie jar, you don’t tell mom she’s imagining things unless you’re ready to get slapped.
And further: if you’re going to play hardball by measuring and borrowing from your competitor’s product, don’t cry about it when they use your products against you, or sabotage your data collection efforts, or simply call you out on it.