I love this.
[NelsonMuntz]Ha ha![/NelsonMuntz]
Apparently “search overload” has forced Microsoft to start sneaking peeks at their neighbor’s test sheet.
It’s great how microsoft’s marketing campaign was all about how the other search engines had too much noise (those ads with stupid people riffing on everything they associated with a word).
Microsoft has it’s own special definition of “innovate.”
You have to admit they’ve been consistent though.
Well, to be fair Google ripped off Bing’s image search a few months ago.
At the risk of sounding daft… I think this story is a bit overblown. At first, it certainly surprised me to hear that Microsoft was incorporating search results from other engines into its own results, but as I thought about it longer, it got less and less crazy.
For instance: aggregators are ubiquitous on the web; we have sites that collect and disperse all sorts of things ranging from news articles, to blogs, to music. How useful would a search engine aggregator be? Pretty useful - if you have a way to sift through all of the crap (i.e., if you just combine search results from multiple search engines, without using any kind of heuristics, you’re going to be left with a mess).
So, Microsoft includes search engine results in the data that it collects and feeds into its own search algorithm. It’s not like Bing was literally showing us Google results as its own (after all, it did take a couple of weeks for the Google plant to work its way to the surface).
I think it’s not a bad idea, actually.
Ha, I had a search bot program I liked back in the 90s … look! I found the page!
Mata Hari. Back then, on dial up it was slow, as in set up the search and go to bed and go through the results in the morning, but it was fun, and really would use all 140 search engines so you didnt just do a google search, and then research with yahoo and so on. I remember it had some specialized search engines it would use, and if you had a lex-nex log in, it would search that too.
I’ve read this and it looks like a very clever ploy by Google that 99% of people will never see through because they don’t really understand the issues involved.
Basically, Google has done this:
-Set up a website that will be hit by a Google search of a random string of letters ( I think they set up 20 like this), the Google engineers were searching for terms like “hnppallkuonomphnomnahkup”
-Insure that the random search term returns one and only one website
-Go home, use Internet Explorer’s browser-based search tool. Like every major web browser, IE has a built in “search” bar. The search bar allows you to search for any term you want using any search engine you want, without having to actually go to the page of the search engine itself. It’s a feature you’ll find in Chrome (where the search bar is shared with the address bar), Firefox, Opera, and Safari.
-Set IE’s search tool to use Google’s search engine.
-Put the input string in at bing.com, post pictures showing that bing is returning the exact same single site result as google from a long string of random letters.
What is this really showing? All it’s showing is that Microsoft tracks what users enter into the search box–by the way, Google does the same with Chrome.
What Microsoft specifically is doing is tracking the search term entered into the box, and then tracking what page the user loads immediately after that, this captures what a user was “looking for” when they searched. Microsoft does this in order to add a “popularity weight” to search results. Both Google and Microsoft’s search algorithms have probably over 1,000 different parameters. The ranking of results isn’t strictly based on any one thing, but by tracking what a user has searched for in the search box and then tracking which website they have selected from the list of results, they can gauge which result was the “most appealing” to a user.
In this case, there is a random string of letters being input into the search box, and it is returning one and only one page. When the Google engineers then navigated to the page, that page entered Microsoft’s databases. Since it had never been encountered before, and since that search term had never been encountered before, it basically created the same search result in Bing as you would get in Google.
However, it’s really not fair to cry foul over this. Microsoft has made it clear it tracks what users search for using that search box. By the way, Google tracks what search terms you enter in the Chrome Address/Search box. By the way, for years Google widely distributed this little browser plug in called the Google toolbar that had its own search box which also tracked what users searched for, for the same purpose.
This is standard practice.
Savvy observers have noticed that any more complex results and Bing and Google show both different results and different rankings for the same results. This is because both are using different algorithms. There’s as of yet no credible evidence that part of Microsoft’s algorithm is “copy the top x results from an equivalent Google search.”
This seems more like a marketing ploy by Google than a ripoff by Microsoft.
Except that the Google toolbar can’t track what page you clicked on after doing the search unless you used the Google search engine. IE has some way of feeding that information back to Microsoft regardless of the search engine used. I don’t know if the Chrome browser tracks the pages you click on after doing a web search; if so, Microsoft would have a legitimate tu quoque.
Microsoft has played dirty as well. Awhile back Microsoft was pointing out that Google sends information to its serves as you type queries in Chrome. Microsoft indicated this was a “very bad thing”, because Google had knowledge of “everything you type.” They contrasted that with their IE8 search box which only communicates with Microsoft after direct user interaction sends the query out to the web.
And of course in that scenario Microsoft didn’t have much of a gripe.
I believe there are opt-outs in both IE and Chrome to avoid sending stuff to the relevant development companies. Additionally, Chrome and IE are free products, and in the case of Chrome you can even get it sans Google if you feel like compiling your own version of the open source Chromium project. Collecting search statistics honestly makes the internet better, not worse. I’ve seen no evidence that search terms have been linked to “personally identifiable information” by either Microsoft or Chrome. So aside from any performance hit your system might take by performing data collection, I see little valid beef against either company for their actions in this regard.