Google Nabs Patent To Monitor Your Cursor Movement 198
bool2 writes "Google has been awarded a patent for displaying search results based on how you move your mouse cursor on the screen... Google's plans are to monitor the movements of the cursor, such as when a user hovers over a certain ad or link to read a tooltip, and then provide relevant search results, and ads, based on that behaviour. It means that it does not require users to actually click a link to know that they were interested in it, opening a world of opportunity for even more focused ads."
Adverts... (Score:3, Insightful)
Fuck adverts.
Re:Adverts... (Score:5, Funny)
Re: (Score:3, Funny)
Re:Adverts... (Score:4, Insightful)
Fuck adverts.
Really? So you're willing to pay a monthly subscription to all your favorite search engines, news sites, etc. then?
Re: (Score:3, Interesting)
They've been on the fence with me for some time. I'm looking at setting up an email server on a VPS with a Roundcube webmail interface as a Gmail replacement. I wouldn't be so creeped out at Google if they could keep my email separate from my other activities, but when I'm logged into Gmail they track where I'm going on Google SERP pages >_>
Creepier still, I set up another Gmail account based on my real name for dealing with potentially uncool business contacts that forwards to my personal account. Th
Hover on this comment (Score:5, Funny)
Hover on this comment and it will change to something relevant.
Re:Hover on this comment (Score:4, Funny)
Re:Hover on this comment (Score:4, Funny)
Not if I install the latest version of MouseBlocker+!
Re: (Score:3, Insightful)
My thoughts exactly. This goes beyond searches submitted to Google. Hopefully it will be opt-in only.
Re: (Score:3, Funny)
Just hover your mouse near the opt-in checkbox and it'll automatically be checked for you.
Re: (Score:3, Interesting)
Hmm... I wonder if it is possible for a plug-in to trap mouse movements and invoke/ave the link pointed at but report back to the browser engine that something else happened. *face thinking*
Re: (Score:2)
Or better yet, just have it capture all of the links on a page and give you a separate, untracked window with which to choose one (possibly disabling all scripting on that window and any other trick it can pull off).
Re: (Score:2)
Re: (Score:3, Informative)
Just block all google related scripts
Nothing to see here, move along
Re: (Score:2)
Except if they freeware their scripts, then anyone could do it.
Re: (Score:2)
Man in the middle attack? Similar has been done already with drive by clicking where mouse clicks are redirected.
Re: (Score:2)
How does javascript constitute monitoring?
Better Google than Amazon... (Score:5, Funny)
Re:Better Google than Amazon... (Score:5, Funny)
Re: (Score:3, Funny)
This gives me incentive to shake my mouse while browsing.
Is that what they call it nowadays?
Re:Better Google than Amazon... (Score:5, Funny)
My wife already handles this for me. :(
Not Accurate Metrics. (Score:3, Insightful)
What about people who inadvertently leave their cursor at a certain spot that happens to be a link while reading the results? It seems to me that this wouldn't produce very useful information.
Re: (Score:3, Insightful)
Re: (Score:2)
And you're slow :)
I'm glad I was born before graphical interfaces so that I don't my mouse on actions where it's slower and less precise than my keyboard.
That and I'm not terrified of C or Assembly either.
Re: (Score:3, Insightful)
Seems like that would be easy enough to sort out. If someone moves their cursor toward something, then away that might be a potential interest. If they leave the cursor in one place for an extended period of time, probably not interest, it's probably "reading." Seems like it wouldn't be too hard to tell if the page had a lot of text as well. Similarly, if you walk away from the computer, the cursor is going to be in one place for a comparatively long time, they'd presumably be able to tell that it was i
Re:Not Accurate Metrics. (Score:4, Insightful)
I now see a bright side to the 'touchscreen devices can't support mouse movement and Javascript hover behavior' complaints about web development for Android, iPhone, iPad, WebOS etc.
Re: (Score:2)
You're conflating "data that contains some noise" with "data that isn't valuable".
I'm also not sure there'll be that much noise, really. I don't know about you, but when I read I usually put my mouse where I'm sure it will be out of the way while I scroll around. For me, usually that's whitespace in the left margin. Even if I cast the cursor aside randomly, what are the odds it lands on an ad-sensitive link?
Re: (Score:2)
New Google motto (Score:2)
Re: (Score:2)
A great user experience awaits in 2030 (Score:5, Interesting)
Thanks Google, for the disclosure of this invention which society will be free to benefit from in 2030.
Some will say that the game is broken and Google is just obliged to play the game too, but in that case, they could make a promise not to use this patent aggressively. Since there's no such promise, all we can say is that they're stockpiling dangerous patents.
Re:A great user experience awaits in 2030 (Score:4, Interesting)
Re: (Score:3, Interesting)
If anyone could post the link to that interview, it would be good to have.
I'm certainly happier that this patent is going to Google than to MS or a troll, but companies change and twenty years is a long time.
Re: (Score:3, Insightful)
Microsoft used to be in the same position. Bill Gates wanted patents for defensive purposes. For a long time they didn't sue anybody over them. That changed.
Re: (Score:2)
Seriously, "display ads determined by *FOO*" is the new "do *BAR*, but on the internet".
The only problem with that... (Score:5, Insightful)
...is most people I know use their eyes instead of their mouse to see. Why would you need to move your mouse over to a certain part of the screen when you can just look there? Also, there's times where the mouse is just sitting in a portion of the screen idly, or sometimes people randomly move their mouse around to fulfill their OCD-ish needs (I'd know, I do that). A better alternative would simply be to see which links people end up clicking, which I'm pretty sure lots of search engines already do, and it works very well from what I've seen.
Re: (Score:2)
They're going to show their search result in a really narrow strip of screenspace, with a scrollbar next to it. Then they'll measure how long you keep looking at a certain part of the search results.
Re:The only problem with that... (Score:4, Informative)
Re:They know that (Score:3, Interesting)
Of course they do.
Welcome to the art of Inverse Patents.
You patent the "sexy" form of the Patent concept, but you implement it 1-X. "Draw a burst radius around what you moved your mouse away from to read and correlate with subsequent clicks".
Re: (Score:2)
I ODC-ishly select things randomly onscreen.
I think I started it as a way to save my place in walls of words while my eyes wandered or I scrollwheeled, and now it's just part of the browsing motion. Though it's definitely found my OCD neurons and they amplify the effect.
So my mouse is almost never where I'm looking, and it's often selecting the wrong words.
Re:The only problem with that... (Score:4, Insightful)
As with all the various habits mentioned in this topic, your case will either be factored into the algorithm, or disable it for you if you are truly too random for the algorithm. There's no reason to assume the algorithm won't be personally tailored to the extent you provide a unique visitor profile.
People tend to forget that algorithms are where Google excels. They shouldn't to be underestimated so easily.
Re:The only problem with that... (Score:4, Funny)
I don't have OCD. I have CDO. That's like OCD but the letters are in the right order.
Re: (Score:3, Funny)
Unfortunately, he did too much LDS and it interferes with his ODC.
Re: (Score:2)
...or sometimes people randomly move their mouse around to fulfill their OCD-ish needs...
Damn you!
I'll show you...
You are AWARE OF YOUR TONGUE!
Re: (Score:2, Insightful)
You aren't kidding. Back in the earlier days of the Internet, I used to have my color palette inverted, so I was more comfortable with white text on a black background. I got fond of it, but as the internet began to rely more and more heavily on embedded and background images and that started resulting often in pages I couldn't see or wouldn't render properly, I finally went back to the 'normal' palette. Now I find I co
Re: (Score:3, Interesting)
Why would you need to move your mouse over to a certain part of the screen when you can just look there?
Because something useful pops up when you do, obviously. Try the new Google Image Search [google.com] to see what I mean.
Re: (Score:2)
sometimes people randomly move their mouse around to fulfill their OCD-ish needs
I know this is what they'd see from my data sometimes.
"It looks like he was looking back a forth in a inch inch range about 3 times a second."
Hmm... (Score:2)
Hellllloooooooooo NoScript.
sadly, my first thought was (Score:4, Funny)
Michael J. Fox isn't worried about this patent.
Re: (Score:2)
even sadder, he's probably terrified of it.
as bad as we say ad-targeting is, totally un-targeted ads are pure noise.
Good luck with that ... (Score:4, Interesting)
I wiggle the mouse and randomly highlight text while I'm reading -- it used to confuse and baffle co-workers. Mostly it's just keeping my hand busy.
If they can infer anything meaningful from what is essentially doodling with the mouse, good luck with that. What I'm highlighting or hovering over has little to do with how they might be able to advertise to me. Heck, I think it would be funny to see the results.
And, I somewhat agree with the observations in TFA that there might be some privacy issues here. I already block google analytics on most of my machines.
Re: (Score:2)
"Mostly it's just keeping my hand busy."
Well, I think I can guess what's entertaining the non-mouse hand.
Re: (Score:2)
*laugh* You know, as I typed that I pretty much knew someone would say something along those lines. Thanks for not disappointing. :-P
No, actually the left hand is sitting on the home row where it belongs. The right hand has become accustomed to being on the mouse a good chunk of the time.
Re: (Score:2)
No, actually the left hand is sitting on the home row where it belongs.
Speak for yourself...
Mine rests on WASD. :)
Re: (Score:2)
The end guy is hard with all that two-finger gestures.
When the end guy is REALLY hard, you end up reducing that to repeated one-finger gestures. ;)
Re: (Score:3, Interesting)
I highlight text, too.
Some people hover over the text that they are reading, moving the mouse in parallel lines across the screen and indicating to Google the speed at which they read,
Some people don't move the mouse at all while reading.
Some people throw the mouse to the corner of the screen while they're reading.
Some people hover over ads but don't click. Others avoid hovering over ads.
Some people's behavior changes when they use a laptop versus a desktop.
Most of the people I know consistently perform a
Re: (Score:2)
Any ad that gets through noscript will be hovered over, just long enough for me to use firebugs inspect element and then edit it out of the damn page.
Does this mean that google will think I am interested in those ads (some google domains are permitted to use javascript due to gmail)?
Re: (Score:2)
AdBlock Plus can already block HTML elements.
Re: (Score:2)
This isn't a story about "spying". (Score:2)
This is a story about patent abuse. There's a language and an environment which fires events based on other events. Now it turns out that actually using these features is so frapping ingenious that nobody but Google can do it for 20 years!
Re: (Score:2)
Re: (Score:3, Informative)
People have done this and got bored with it 2 years ago or more.
If you had RTFP, you'd know that Google's patent [uspto.gov] application was initially filed in Dec., 2004. That's a little over 2 years ago in case you couldn't figure it out.
Re: (Score:2)
Legally (Score:4, Interesting)
What is the difference between this and a keylogger?
It's one thing to record commands I have sent to their computers by clicking. It's another thing entirely to track things I do on MY computer. I foresee a lot of legislation in Google's future.
Re: (Score:2)
I'm a bit out of date on the nuts and bolts, mainly because I'm not in web development, but my guess is that they can only track hover actions not raw mouse data. It's not terribly different than using a tracking pixel.
Don't get me wrong, I'm usually one of the first to start worrying about privacy issues. But I think that here, the likelihood of the data set being dominated by noise and leading to extremely weird marketing behaviors is a larger concern than the privacy concerns - assuming that the patent e
Re: (Score:2)
I'm a bit out of date on the nuts and bolts, ..., but my guess is that they can only track hover actions not raw mouse data
If you mean "raw mouse data" as in bits coming down the wire from the mouse (or receiver for wireless), they definitely can't get at that.
Assuming no software is downloaded, web sites only have access to stuff the browser gives them. This is in the form of actions (mouse move, clicks, etc) and keyboard codes/characters being performed on a Javascript object (e.g. a textbox, buttons or the page itself). Assume any input done while a browser is active could be tracked by the page in the current tab.
As to th
Re: (Score:2)
I'm a bit out of date on the nuts and bolts, mainly because I'm not in web development, but my guess is that they can only track hover actions not raw mouse data.
There's also the onmousemove event, which provides the coordinates of the mouse. While you're on a page with the JavaScript (and don't have JavaScript blocked or disabled), they can track the position of the mouse relatively to the browser window.
Re: (Score:2)
ESPECIALLY if you use the onboard keyboard Ease of Access tool!
Though generally this only deals with your mouse on the browser (I believe?)
Re:Legally (Score:4, Informative)
Re: (Score:3, Insightful)
They wouldn't be tracking your mouse movements anywhere other than when you are on their page.
When I click on something, it is implied that I am giving them permission to do something. That's how the internet works, after all. Google and others have up to now taken liberties with click data, assuming that they can do what they want with it. And so far, no one has seriously objected.
However just visiting a page, with no warning that I am going to be "tracked
Re: (Score:3, Interesting)
That's only half true. As a web developer, I have never personally developed any sort of mouse tracking software but I can tell you unequivocally that I could easily register a listener for mouse
Someone, quick patent eye movement. (Score:2)
For me, it'll be incorrect data (Score:3, Insightful)
If I'm reading something, I move the mouse out of the way. So, if Google want to track what I'm interested in, they'll need to look at what the mouse is _not_ hovering over, or certainly not stopped over.
Re: (Score:2)
Re: (Score:2)
I am sure this would be possible too. Since if you are looking at a page of google search results they can easily check the distance + time quality vs whatever you eventually click on to see if you move your mouse away from what you eventually click on... or more your mouse close to what you eventually click on.
Then they can just set that correlation coefficient in a cookie / server side property of your google account and use it to modify the search behavior appropriately.
There is all sorts of prior art on this.... (Score:3, Insightful)
No, there isn't, nor is it obvious... (Score:3, Insightful)
I've worked at two companies where we created libraries for monitoring cursor movement, what the business folks used it for I'm not certain but this has been done over and over. What is so new and innovative about their implementation that it is patentable?
First, to say something is prior art, you have to read the claims of the patent, not the title of the Slashdot summary. For one, were your two companies providing search results and modifying the relevance of the results based on the cursor movement? Probably not.
Second, flip through the comments here on Slashdot:
Good luck with that [slashdot.org]
For me, it'll be incorrect data [slashdot.org]
The only problem with that... [slashdot.org]
Not accurate metrics. [slashdot.org]
Apparently, ordinary "skilled in the art" programmers and computer folks think that thi
Hooray! (Score:5, Insightful)
I am all for more focused ads. I dream of the day i will get an advert for something i will actually buy.
Re: (Score:2)
I am all for more focused ads.
Yeah, on a page I don't even visit.
This isn't fair. (Score:5, Funny)
What am I supposed to do now?
Boxy Ads (Score:2, Funny)
Hrm. (Score:2)
Blocking? (Score:2)
I wonder (Score:2)
Not new at all (Score:2)
We wrote code to monitor mouse movements to detect click fraud years ago. However, we didn't deploy it for general use because it's a violation of the end-user's privacy - we only used it for pages and IPs that we suspected were either bots or paid-to-click.
So yes, by your mouse movement (not just the movement, but the timing) we could take specific advertisement-related actions.
it has to be said... (Score:3, Funny)
Data poisoning (Score:2)
It seems to me there is a wonderful opportunity here to do some data poisoning. A nice plugin to send Google as much false data and noise as possible, to reduce the value of this technology as close to zero as can be arranged. Identification of advertisers who knowingly use this information would be good too, to do a little "targeted" activity as a return favor. Nothing drastic or illegal, just boycotts and public shaming and attempts to poison their data caches too.
If they do this, my default search eng
but, seriously (Score:2)
P.S. I, slick7 do hereby claim the concept of a MAC address mouse (IPv4,IPv6,IPv8,IPV16,IPv32) with biometric fingerprint scanner on this date,27 July, 2010 at the time of 2059 GMT. This claim supe
Re: (Score:2, Funny)
Re: (Score:2)
Why? Are you making obscene gestures with it? Moving it up and down really rapidly?
Re:Too invasive (Score:4, Funny)
He suffers from premature clicking. There's an app for that, but he's too embarrassed to buy it.
Re: (Score:2)
Then again, if he clicks prematurely, he may have clicked the "buy now!" button before realising the embarrassement.
Re: (Score:3, Informative)
Re:I assume webkit will be properly protected (Score:5, Informative)
This can be done right now in any browser unless you turn off or restrict JavaScript.
Re: (Score:2)
Even with that, it's pretty easy for the rendering engine to know where the mouse is at any given moment (query the OS) and relate that back to the contents of the page. All without ever touching the code on the web page. That being said, reporting that information back to the web site or to Google directly would be pretty easy to spot.
Re: (Score:2)
But doing it that way would only work in Chrome, or would require users to install an extension of some kind (Google toolbar?).
It would also be much easier to block because I assume Google would be the only one using that functionality, and/or you could just uninstall the extension or use a different browser if it really bugs you.
Personally I don't really see what the issue is as long as they're just watching your mouse cursor on their pages.
Re: (Score:2)
Re: (Score:2)
from snooping via this sort of BS too ... right?
It is trivially esy to avoid this sort of snooping: Use bing. Nobody forces you to use google as a search engine.
Re: (Score:2, Interesting)
Uh, Google is already doing that. Search suggestions, tracking the typing speed in form fields (they really do this, apparently so they can detect bots, but it opens up all of those possibilities)... They also see what link you click on their search results via background http request when you click it.
Re: (Score:3, Interesting)
You’re missing the point. As long as you have some sort of “hover” action, this applies – in fact, since useful information is often hidden until you mouse-over something, hardware designers are constantly trying to find better ways of implementing hover on touch interfaces.
Go check out Google’s new-and-improved image search results page [google.com] for a perfect example of this sort of thing. They’ve completely done away with the text surrounding each image – hovering over one