To, in essence, spy on Joe User's interactions with another company, with neither knowledge nor informed consent from either party?
Incorrect. In this case Joe User himself granted permission to Microsoft to collect the data. Microsoft asked Joe User to do this when he installed Bing Toolbar, and they explicitly told him they would be collecting information on the searches he did and the websites he visited.
You are saying the the Bing toolbar sucessfully recognizes this government search page as a search and realizes which field is the search term? And recognized which click the user did subsequently as the actual link to the searched item, rather than ads, next page, new search, etc? And their AI was designed to figure this out without somebody at Bing ever looking at this page? And those writers did not design this algorithm to specifically recognize Google's search pages?
This is not a difficult problem. Watch for when a user types something into a search box, watch what they do after, collect the data. There's going to be a lot of noise, but there are heuristics to extract the signal you want, even if you don't get them all.
Also I find it pretty amazing that they manage to extract exactly the same sample text from the page as Google did. This is particularly amazing when the destination page no longer contains that text.
It's in a meta-tag, specifically designed to be put in a search engine summary.
If you'd read the original article, this is exactly what twigged Google to the fact that this was happening.
Yet Google never proved that was the case, nor offered any other examples. Bing could have just as easily gone to Wikipedia and followed their redirect.
which is why the Google engineers used nonsense terms--it was easiest to see the results where few or no results were returned.
All it proves is that Microsoft is doing exactly what they said they would do with data you send them about searches you do.
Common terms? The only evidence Google offers is a gross misspelling of a rare eye surgery procedure. The only thing this proves is that Google is better at correcting spelling, and that given sparse information Bing will rank user data disproportionately high.
Ever hear of ?
Search engines commonly use it to provide text to go with the link. Guess what the meta description is for rim.com:
Looking for contact information, locations, executive biographies or what the latest RIM press release is saying? Click and discover more about Research In Motion.
At its core, a search engine is a mapping from a search term S to a set of ordered results R = {l1,l2,l2...}. The problem all search engines are faced with is, what links are elements of R, and in which order are the presented.
Microsoft has their own algorithms for answering this question (this is a fact, even though many people on/. want to believe they are copying Google's results verbatim).
The only thing Google exposed is that as part of Bing takes usage statistics provided by users voluntarily to augment their algorithms. Presumably, the user input is coming in the form (Search String, Selected Link) to see which links users click on, given a search term.
The only thing Google actually proved is that in situations where Bing is unaware of the link in the user-supplied data, they will index it. This is a completely reasonable, and logical thing to do, and any programmer tasked with the job would do the same.
No, Google's "experiment" revealed a flaw in Bing's algorithm.
Bullshit. They clearly take results for pages that are completely unrelated to their search terms and just spray them up, automatically.
For contrived corner cases, where isn't much other data, yes, user feedback will recieve a disproportionately high ranking it seems. Regardless, If Bing finds that 100% of the the people who search for "wrijgdssd" click on a link for "how to grow tomatoes" in another search engine, why should they not rank it higher? Even though there are no keywords matching, user behavior indicates they find the link interesting.
Even with this kind of help, the best Bing will ever be is just behind the curve, a high-order fit of the solution-space of an algorithm that they failed to better.
Correct, assuming this is the only thing Bing is doing. The link indicates this is one of many (1000s) of metrics used to rank and index searches. It is entirely possible to return different and better results than google while still learning from their results. In fact if they do the same for every search engine, and google does this for none, then their results will end up strictly better in the long run.
They're not just doing it when Google does searches
And they're not just doing it to Google searches. They're doing it to any search engine or even any website. The fact that they're indiscriminate makes it hard to say they're simply piggybacking or reproducing google's results.
Now, the users may have agreed to this, but Google certainly didn't.
And? The information they're collecting is usage statistics of their customers, who have installed the Bing toolbar or IE and have opted to share data with Microsoft. These statistics allegedly show what the user searched for, and what links they found useful from that search. This is novel data Google does not own.
Let me say this again because people keep missing this point: for the most part Microsoft isn't indexing sites using this method, but augmenting their page ranking algorithm. This data does not belong to Google, is not generated by google, and therefore Goolge has no claim to it.
In windows 7, yes, you can uninstall IE. The rendering engine stays behind for certain help systems that use it, but the application itself can be permanently removed.
The problem is that Microsoft are again leveraging their windows monopoly to gain an unfair competitive advantage over a competitor, they are effectively using the massive installed base of IE
Sorry, this argument does not apply in 2011. IE currently has ~55% market share and it's declining every day. People can uninstall IE from their computers, and install any of half a dozen browsers which are quickly gaining steam in the marketplace. Microsoft may be the dominant player, but they are not certainly do not hold a monopoly in the browser space.
I can't think of a single thing Microsoft has done that was an original idea.
That's because there's no such thing as an original idea. There are ways of doing things better, cheaper, faster, or differently, but there are very truly original ideas.
Just look at iPod, iPhone, iPad. All derivative, but completely successful anyway. Even multitouch in the original iPhone is not original, straight down to the gestures. Google wasn't the first to do search and they won't be the last.
I mean, even if I were to release a hover car or anti gravity device, or even a freaking time machine it would not be an original idea.
Probably Google let Bing watch by intentionally performing these searches with the toolbar installed
Thanks for following along, that's exactly what Google did.
When the experiment was ready, about 20 Google engineers were told to run the test queries from laptops at home, using Internet Explorer, with Suggested Sites and the Bing Toolbar both enabled.
When the experiment was ready, about 20 Google engineers were told to run the test queries from laptops at home, using Internet Explorer, with Suggested Sites and the Bing Toolbar both enabled.
So the Google engineers installed the bing toolbar, and when they did, Microsoft asked them if it was okay to look at the sites they visited and the links they clicked on, to which they replied "Okay." The Google engineers then typed in the phony terms, visited the phony links, and thereby sent the phony data to
Microsoft.
Microsoft sees it and says, "it looks like people are searching for this term and like this result. We've never seen this before, we should probably index this."
The key here is that Microsoft asked the user (Google in this case) if they could look at the data to improve Bing, the user sends the data to Microsoft (which they agreed to do), and now they're raising a storm because (surprise!) Microsoft did with the data exactly what they said they would do.
They do spy on (sorry, gather 'click stream' data from) IE users (through IE itself, or one of its add-ons). Read those EULAs veeery carefully, folks!
It's not in the EULA. It's a check box the user sees on install they can uncheck. MS is spying on users the same way I would be spying on you if I gave you a camera and said "I'm going to use this to watch what you do, is that okay?" and you said "sure."
When you install Bing Bar there is a checkbox that says:
Help Microsoft improve your online experience with personalized content by allowing us to collect additional information about your system configuration, the searches you do, websites you visit, and how you use our software. We will use this information to help improve our products and services.
You can either A) not install the toolbar at all or B) Opt not to give Microsoft this information.
IE didn't spy on them, the user (a google engineer in this case) opted in to send clickstream data to Bing.
Yes, in this case someone (again the google engineer) actually searched for and clicked on those obscure terms with the intent of sending the bogus data to Bing (which he had previously opted to do).
Since the data was so obscure and unique, the single data point from the google engineer received a disproportionately high weight in the search results.
I still don't understand how this is in any way dodgey or underhanded.
Step 1: User opts in to report anonymous clickthrough data to Bing
Step 2: User searches for a term, chooses a search result
Step 3: Microsoft gets the data and compares it against relevent information for that search term.
Since google chose a random, unique for their search term, there is nothing to compare the user behavior with so it receives a disproportionately high amount of weight. With actual search terms, what a user searches for on google will have significantly less weight in the rankings, and depending on their algorithm could be next to inconsequential.
But the fact remains that looking at how customers use the competition (especially the frontrunner) is prevalent in all industries, and is a really smart idea from a business standpoint, and only serves to benefit your customers. By the looks of Google's optional photo homepage, they are guilty of it too (and if they weren't doing more I would be shocked). It seems the only people who are upset about this are Google, and people loyal to Google (most/. users).
Kinect games brought absolutely no innovation in gameplay.
Because optical full body skeleton tracking has been done before in a $150 consumer device. I'm not sure you really know what the Kinect is. Eyetoy provides a 320×240 RGB image. Kinect provides 640x480 RGBD image, so it has an extra depth dimension which is the key innovation of the kinect.
Eyetoy enables rudimentary color detection, edge detection, and gesture tracking. The depth information provided by the kinect goes far beyond this, as the devices provides full body motion tracking, advanced gesture recognition, facial recognition, person tracking (with integrated motors), and voice recognition (with microphone array).
So far the coolest things I've seen with the kinect are automatic login via face recognition, and simultaneous full body tracking for two player games. This is something eyetoy was never capable of, or any console for that matter. Thus, innovative.
Do you have any idea what the kinect is? Eyetoy is an overpriced glorified webcam with some object segmentation and gesture recognition behind it.
Kinect is a stereoscopic camera combined with an infrared depth sensor. This provides the xbox with a 640x480 RGBD image it uses for full-body skeleton tracking, facial recognition, gesture recognition.
If that wasn't enough, it contains motors, an accelerometer, and a microphone array for body tracking and voice recognition.
Slashdot Mirror
Google proved that Bing is doing this for all searches that are captured using the Toolbar.
Microsoft asks for your permission to do this when you install the toolbar. This is not news.
To, in essence, spy on Joe User's interactions with another company, with neither knowledge nor informed consent from either party?
Incorrect. In this case Joe User himself granted permission to Microsoft to collect the data. Microsoft asked Joe User to do this when he installed Bing Toolbar, and they explicitly told him they would be collecting information on the searches he did and the websites he visited.
You are saying the the Bing toolbar sucessfully recognizes this government search page as a search and realizes which field is the search term? And recognized which click the user did subsequently as the actual link to the searched item, rather than ads, next page, new search, etc? And their AI was designed to figure this out without somebody at Bing ever looking at this page? And those writers did not design this algorithm to specifically recognize Google's search pages?
This is not a difficult problem. Watch for when a user types something into a search box, watch what they do after, collect the data. There's going to be a lot of noise, but there are heuristics to extract the signal you want, even if you don't get them all.
Also I find it pretty amazing that they manage to extract exactly the same sample text from the page as Google did. This is particularly amazing when the destination page no longer contains that text.
It's in a meta-tag, specifically designed to be put in a search engine summary.
If you'd read the original article, this is exactly what twigged Google to the fact that this was happening.
Yet Google never proved that was the case, nor offered any other examples. Bing could have just as easily gone to Wikipedia and followed their redirect.
which is why the Google engineers used nonsense terms--it was easiest to see the results where few or no results were returned.
All it proves is that Microsoft is doing exactly what they said they would do with data you send them about searches you do.
Common terms? The only evidence Google offers is a gross misspelling of a rare eye surgery procedure. The only thing this proves is that Google is better at correcting spelling, and that given sparse information Bing will rank user data disproportionately high.
That is not surprising, as when the user installs Bing Bar, it tells them it will do just this, and the user can opt out of it.
Looking for contact information, locations, executive biographies or what the latest RIM press release is saying? Click and discover more about Research In Motion.
That's what Google's experiment demonstrated.
At its core, a search engine is a mapping from a search term S to a set of ordered results R = {l1,l2,l2...}. The problem all search engines are faced with is, what links are elements of R, and in which order are the presented.
Microsoft has their own algorithms for answering this question (this is a fact, even though many people on /. want to believe they are copying Google's results verbatim).
The only thing Google exposed is that as part of Bing takes usage statistics provided by users voluntarily to augment their algorithms. Presumably, the user input is coming in the form (Search String, Selected Link) to see which links users click on, given a search term.
The only thing Google actually proved is that in situations where Bing is unaware of the link in the user-supplied data, they will index it. This is a completely reasonable, and logical thing to do, and any programmer tasked with the job would do the same.
No, Google's "experiment" revealed a flaw in Bing's algorithm.
Bullshit. They clearly take results for pages that are completely unrelated to their search terms and just spray them up, automatically.
For contrived corner cases, where isn't much other data, yes, user feedback will recieve a disproportionately high ranking it seems. Regardless, If Bing finds that 100% of the the people who search for "wrijgdssd" click on a link for "how to grow tomatoes" in another search engine, why should they not rank it higher? Even though there are no keywords matching, user behavior indicates they find the link interesting.
Even with this kind of help, the best Bing will ever be is just behind the curve, a high-order fit of the solution-space of an algorithm that they failed to better.
Correct, assuming this is the only thing Bing is doing. The link indicates this is one of many (1000s) of metrics used to rank and index searches. It is entirely possible to return different and better results than google while still learning from their results. In fact if they do the same for every search engine, and google does this for none, then their results will end up strictly better in the long run.
But you certainly have no leg to stand on when you complain about someone hearing your "private" conversation.
They're not just doing it when Google does searches
And they're not just doing it to Google searches. They're doing it to any search engine or even any website. The fact that they're indiscriminate makes it hard to say they're simply piggybacking or reproducing google's results.
Now, the users may have agreed to this, but Google certainly didn't.
And? The information they're collecting is usage statistics of their customers, who have installed the Bing toolbar or IE and have opted to share data with Microsoft. These statistics allegedly show what the user searched for, and what links they found useful from that search. This is novel data Google does not own.
Let me say this again because people keep missing this point: for the most part Microsoft isn't indexing sites using this method, but augmenting their page ranking algorithm. This data does not belong to Google, is not generated by google, and therefore Goolge has no claim to it.
In windows 7, yes, you can uninstall IE. The rendering engine stays behind for certain help systems that use it, but the application itself can be permanently removed.
I haven't read the privacy policy w.r.t. the bing toolbar, but my guess is this is all outlined very clearly there. I'll have to look at it later.
The problem is that Microsoft are again leveraging their windows monopoly to gain an unfair competitive advantage over a competitor, they are effectively using the massive installed base of IE
Sorry, this argument does not apply in 2011. IE currently has ~55% market share and it's declining every day. People can uninstall IE from their computers, and install any of half a dozen browsers which are quickly gaining steam in the marketplace. Microsoft may be the dominant player, but they are not certainly do not hold a monopoly in the browser space.
I can't think of a single thing Microsoft has done that was an original idea.
That's because there's no such thing as an original idea. There are ways of doing things better, cheaper, faster, or differently, but there are very truly original ideas.
Just look at iPod, iPhone, iPad. All derivative, but completely successful anyway. Even multitouch in the original iPhone is not original, straight down to the gestures. Google wasn't the first to do search and they won't be the last.
I mean, even if I were to release a hover car or anti gravity device, or even a freaking time machine it would not be an original idea.
Probably Google let Bing watch by intentionally performing these searches with the toolbar installed
Thanks for following along, that's exactly what Google did.
When the experiment was ready, about 20 Google engineers were told to run the test queries from laptops at home, using Internet Explorer, with Suggested Sites and the Bing Toolbar both enabled.
http://searchengineland.com/google-bing-is-cheating-copying-our-search-results-62914
I'd like to head out there and just use some brush burners to get rid of the snow on my driveway.
Wrong
From http://searchengineland.com/google-bing-is-cheating-copying-our-search-results-62914:
When the experiment was ready, about 20 Google engineers were told to run the test queries from laptops at home, using Internet Explorer, with Suggested Sites and the Bing Toolbar both enabled.
So the Google engineers installed the bing toolbar, and when they did, Microsoft asked them if it was okay to look at the sites they visited and the links they clicked on, to which they replied "Okay." The Google engineers then typed in the phony terms, visited the phony links, and thereby sent the phony data to Microsoft.
Microsoft sees it and says, "it looks like people are searching for this term and like this result. We've never seen this before, we should probably index this."
The key here is that Microsoft asked the user (Google in this case) if they could look at the data to improve Bing, the user sends the data to Microsoft (which they agreed to do), and now they're raising a storm because (surprise!) Microsoft did with the data exactly what they said they would do.
They do spy on (sorry, gather 'click stream' data from) IE users (through IE itself, or one of its add-ons). Read those EULAs veeery carefully, folks!
It's not in the EULA. It's a check box the user sees on install they can uncheck. MS is spying on users the same way I would be spying on you if I gave you a camera and said "I'm going to use this to watch what you do, is that okay?" and you said "sure."
When you install Bing Bar there is a checkbox that says:
Help Microsoft improve your online experience with personalized content by allowing us to collect additional information about your system configuration, the searches you do, websites you visit, and how you use our software. We will use this information to help improve our products and services.
You can either A) not install the toolbar at all or B) Opt not to give Microsoft this information.
IE didn't spy on them, the user (a google engineer in this case) opted in to send clickstream data to Bing.
Yes, in this case someone (again the google engineer) actually searched for and clicked on those obscure terms with the intent of sending the bogus data to Bing (which he had previously opted to do).
Since the data was so obscure and unique, the single data point from the google engineer received a disproportionately high weight in the search results.
I still don't understand how this is in any way dodgey or underhanded.
Since google chose a random, unique for their search term, there is nothing to compare the user behavior with so it receives a disproportionately high amount of weight. With actual search terms, what a user searches for on google will have significantly less weight in the rankings, and depending on their algorithm could be next to inconsequential.
But the fact remains that looking at how customers use the competition (especially the frontrunner) is prevalent in all industries, and is a really smart idea from a business standpoint, and only serves to benefit your customers. By the looks of Google's optional photo homepage, they are guilty of it too (and if they weren't doing more I would be shocked). It seems the only people who are upset about this are Google, and people loyal to Google (most /. users).
In other news, Toshiba supplies Apple with multi touch displays. Apple can't deliver!
Kinect games brought absolutely no innovation in gameplay.
Because optical full body skeleton tracking has been done before in a $150 consumer device. I'm not sure you really know what the Kinect is. Eyetoy provides a 320×240 RGB image. Kinect provides 640x480 RGBD image, so it has an extra depth dimension which is the key innovation of the kinect.
Eyetoy enables rudimentary color detection, edge detection, and gesture tracking. The depth information provided by the kinect goes far beyond this, as the devices provides full body motion tracking, advanced gesture recognition, facial recognition, person tracking (with integrated motors), and voice recognition (with microphone array).
So far the coolest things I've seen with the kinect are automatic login via face recognition, and simultaneous full body tracking for two player games. This is something eyetoy was never capable of, or any console for that matter. Thus, innovative.
Do you have any idea what the kinect is? Eyetoy is an overpriced glorified webcam with some object segmentation and gesture recognition behind it.
Kinect is a stereoscopic camera combined with an infrared depth sensor. This provides the xbox with a 640x480 RGBD image it uses for full-body skeleton tracking, facial recognition, gesture recognition.
If that wasn't enough, it contains motors, an accelerometer, and a microphone array for body tracking and voice recognition.
With very little modification, the same sensor used for playing silly dance games on the xbox can be used for everything from teleoperating humanoid robots to an augmented reality midi device.
Seriously, comparing the kinect to an eyetoy is just so far off base it's not funny.